Distributed robust optimization strategy for multi-energy virtual power plant clusters Authors: Ziyang Wang, Hao Guo, Ruijin Zhu, Zixuan Liu Published: 2025 ABSTRACT As renewable energy penetration continues to rise, the demand for coordinated optimization of decentralized source-load-storage. Virtual power plant (VPP) addresses this need by aggregating and coordinating diverse resources. However, effective mechanisms for multi-agent VPP coordination remain limited. To address this, this paper proposes a distributed robust optimization strategy for multi-energy VPP clusters in high-altitude regions. This strategy combines a dual-norm uncertainty set with a Nash bargaining mechanism to coordinate multi-agent interactions and mitigate operational risks. First, addressing the integrated planning-operation problem for multi-energy VPPs, a two-stage distributed robust optimization model is established. The dual-norm uncertainty set characterizes scenario probability uncertainties. Second, aiming to minimize interaction costs, a bargaining model is proposed based on Nash bargaining theory. Finally, case studies are conducted on three multi-energy VPPs. Results demonstrate that the proposed optimization strategy effectively enhances overall system revenue, environmental benefits, and source-load matching capability while ensuring fair competition among VPPs.Supplementary InformationThe online version contains supplementary material available at 10.1038/s41598-025-28736-6. Introduction Due to the low oxygen content in the air in plateau regions, the resulting hypoxia significantly affects people’s production and daily lives. The government and relevant scholars have been dedicated to addressing oxygen supply issue in these high-altitude areas. The 14th Five-Year Plan of Tibet in 2021 explicitly emphasized the necessity and development goals for oxygen supply in plateau regions1–3. Simultaneously, according to China’s lofty objectives of attaining the peak of carbon by 2030 and being carbon–neutral by 2060, the development of renewable energy and energy-saving emission reduction has become urgent. Tibet, as the national base for clean energy, advanced several strategic objectives and timelines. However, the geographical distribution of renewable energy in Tibet is excessively scattered4, leading to a lack of coordination, output volatility, and challenges in accurate prediction. The intermittency of renewable energy constrains the dispatch capability of integrated energy systems, hindering the effective integration of the Integrated Energy System (IES). Achieving reliable operation under fluctuating conditions has become a critical concern. VPP have emerged to address this challenge by aggregating and coordinating distributed energy resources, thereby enhancing dispatch capability and system integration levels5–10. VPP as one of the important technologies for realizing the smart grid, integrates, optimizes, and coordinates decentralized distributed energy systems within a region through advanced technical means. This promotes the rational optimization of resources and increases the absorption rate of renewable energy11,12. Current research on VPPs is extensive and has achieved certain results. Literature13 suggest an effective robust self-arrangement approach for mitigating uncertainties related to energy costs, wind energy production, and power demand. Literature14 considers the geographical differences of distributed energy and numerous small-scale energy producers and consumers, proposing a day-ahead thermoelectric scheduling algorithm for optimizing large-scale VPPs. Literature15 proposes a bi-level real-time economic dispatch strategy for VPPs, considering the different interests of VPP systems and operators and the uncertainties of distributed renewable energy and loads. Literature16 to address the issues of output fluctuations in remote sustainable energy and time variations in load demand, this study introduces a load aggregating for regulation and suggests an economic operation approach appropriate for VPP. Literature17 proposes a VPP collaborative operational framework that incorporates optimized scheduling, data security and day-ahead price forecasting (Prophet), and completes case validation in VPP scenarios. However, Nash bargaining is not employed for settlement or revenue distribution. In a cluster containing multiple VPP entities, the ADMM algorithm combined with Nash bargaining theory is typically used to address cooperation and competition issues among multiple entities, considering their mutual benefits and energy exchanges. The Nash bargaining model is a non-convex nonlinear optimization problem18, distributed solution schemes are common. A two-stage coordination mechanism based on ADMM for VPP clusters was introduced in literature19 with day-ahead Nash bargaining and intraday pricing. However, the study did not consider a distributionally robust treatment of scenario probabilities. It also did not make use of the dual-norm ambiguity set, nor was the worst-case expected operating cost optimized within a distributionally robust framework. Literature20 proposes a distributed multi-microgrid energy-sharing strategy based on Nash games. However, the following were not modelled: probability ambiguity through distributionally robust optimization with a dual-norm ambiguity set; the post-dispatch price-alignment layer; and oxygen/VPSA coupling. A P2P-to-grid trading framework with dynamic operating envelopes and communication-aware ADMM was presented in literature21. However, network and communication aspects were prioritized, whereas dual-norm DRO, worst-case expectation, and a post-dispatch bargaining layer were not integrated. Literature22 proposes a collaborative operation scheme for multiple virtual power plants and load aggregators based on Nash game theory and the dynamic penalty ADMM algorithm. However, adversarial scenario-probability reweighting was not included, a two-stage iterative master–subproblem procedure to optimize the worst-case expected cost was not employed, and without considering the oxygen load. Given the variability in renewable energy generation and User’s spontaneous behavior, the safe and economic operation of the power system is greatly affected23–25. Currently, to address the uncertainty of renewable energy output, research mainly utilizes stochastic optimization26–28 and robust optimization methods29–31 to effectively solve system operation problems. However, these two optimization methods are overly conservative. Some scholars have also adopted distributed robust optimization methods to study the uncertainty of renewable energy output32–35. Literature36 suggests a novel, robust optimization approach that’s designed to handle the complex interplay of various regional integrated energy systems, accounting for the intricate multi-energy flow connections. The study reveals that this innovative strategy holds the potential to not only enhance the overall economic efficiency of these systems but also bolster their overall stability. Literature37 presents a two-stage distributionally robust optimization model for virtual power plant operation, coordinating day-ahead scheduling under renewable uncertainty and validating the framework with case studies. However, a dual-norm ambiguity set is not adopted, a post-dispatch Nash price-coordination layer is not separated from physical scheduling. Literature38 develops a scenario-based distributed robust optimization approach for optimal scheduling of virtual power plants, enabling scalable coordination across multiple entities. However, the worst-case expected operating cost is not optimized via an iterative master–subproblem procedure with a dual-norm probability set. Currently, there is relatively little research on integrated energy systems in plateau regions, with only my research group conducting related studies. Literature39 establishes an energy cooperation framework for high-altitude prosumers and proposes an asymmetric profit distribution model for multi-energy sharing, ultimately enhancing the profits of individual entities and the overall alliance. Literature40 introduces a two-tiered energy collaboration model designed for high-altitude prosumer networks. This framework addresses the evolving energy exchange dynamics in plateau areas, boosting economic returns while optimizing collective benefits. It also fosters long-term viability for the shared prosumer market in these elevated regions. Literature41 based on Tibet’s power supply and growth planning, provides an innovative optimization strategy for integrated energy system clusters, increasing the multi-energy sharing system’s economic efficiency. Based on the above research, this paper proposes a two-stage distributed robust optimization strategy for virtual power plant clusters in high-altitude regions, addressing the interplay between oxygen supply requirements and multi-energy flows. The model employs dual-norm fuzzy sets and an iterative master–slave problem approach to handle worst-case expected costs, enabling multi-energy flow interactions among VPPs. Nash bargaining is applied to optimize economic dispatch. Additionally, sensitivity analysis is conducted to validate result robustness, ensuring the model’s reliability under varying conditions. Under the simultaneous constraints of 1-norm and ∞-norm, a scheduling strategy is constructed based on the mechanism of multi-energy flow interactions, which synergizes system robustness and economy. The principal contributions of this paper are outlined below:(1) This paper examines energy demand and development strategies for high-altitude regions, has constructed the electrical, thermal, oxygen, and other multi-energy flow VPP system model framework by coupling devices like vacuum pressure swing adsorption (VPSA) and gas turbine (GT).(2) By combining the uncertainty probability set of norm constraints, it describes the uncertainty of scenarios, and the worst-case scenario’s probability value is determined by updating the scenario possibility values.(3) Aiming at the planning and operation problems of multi-energy VPP, a distributed robust optimization model considering double norm constraints has been established. Secondly, the Nash negotiation model has been constructed with the goal of minimizing interaction costs. Ultimately, the methodology outlined in this document has been confirmed as effective. (1) This paper examines energy demand and development strategies for high-altitude regions, has constructed the electrical, thermal, oxygen, and other multi-energy flow VPP system model framework by coupling devices like vacuum pressure swing adsorption (VPSA) and gas turbine (GT). (1) This paper examines energy demand and development strategies for high-altitude regions, has constructed the electrical, thermal, oxygen, and other multi-energy flow VPP system model framework by coupling devices like vacuum pressure swing adsorption (VPSA) and gas turbine (GT). (2) By combining the uncertainty probability set of norm constraints, it describes the uncertainty of scenarios, and the worst-case scenario’s probability value is determined by updating the scenario possibility values. (2) By combining the uncertainty probability set of norm constraints, it describes the uncertainty of scenarios, and the worst-case scenario’s probability value is determined by updating the scenario possibility values. (3) Aiming at the planning and operation problems of multi-energy VPP, a distributed robust optimization model considering double norm constraints has been established. Secondly, the Nash negotiation model has been constructed with the goal of minimizing interaction costs. Ultimately, the methodology outlined in this document has been confirmed as effective. (3) Aiming at the planning and operation problems of multi-energy VPP, a distributed robust optimization model considering double norm constraints has been established. Secondly, the Nash negotiation model has been constructed with the goal of minimizing interaction costs. Ultimately, the methodology outlined in this document has been confirmed as effective. The contents of each chapter of this article are summarized as follows. Chapter 2 presents the VPP system model and the double norm constraint model; Chapter 3 will expound upon the two-stage distributed optimized robust paradigm; Chapter 4 introduces the VPP cluster negotiation model based on Nash negotiation; Chapter 5 presents various data and case simulation results; Chapter 6 introduces the conclusion. VPP System model VPP System operation model Figure 1 is the energy flow architecture diagram of the VPP cluster, which implements information transmission and exchange through a control center. An energy management center exists in each VPP. The dispatching commands can only be conveyed between various VPPs through the energy management center. The power generation units, oxygen supply units, and heating units in each VPP will first meet their own load requirements. If there is excess electricity at this time, it can be directly transmitted to other VPP to ensure their normal operation. When the power generation capacity of a certain VPP cannot meet its own load demand, it can receive electricity from other VPP or purchase electricity from the power grid. The natural gas network undertakes the task of one-way delivery of natural gas from natural gas suppliers to power generation and heating equipment in each VPP. At the same time, these energy conversion devices can convert natural gas into electricity and heat to meet the demand for different energies of each VPP.Fig. 1VPP energy sharing architecture diagram. VPP energy sharing architecture diagram. Figure 2 is a single VPP system architecture diagram, which includes a gas turbine (GT), wind turbine (WT) or photovoltaic (PV), a gas boiler (GB), an electric cooler (EC), an absorption chiller (AC), and a vacuum pressure swing adsorption (VPSA). The load includes electricity load, heat load, cold load, and oxygen load. The electricity load in VPP can only obtain electricity from the external grid, WT, PV, and GT. The demand for natural gas for each equipment within each VPP comes from the natural gas supplier. In summary, the external grid, WT, PV, and GT are equivalent to the power supply of the VPP cluster, and the natural gas supplier is the source of natural gas for the entire VPP cluster. A variety of energy conversion devices in the system can convert electrical energy and natural gas into energy required by different types of loads to meet various needs.Fig. 2Architecture of a single VPP system. Architecture of a single VPP system. The VPP system adopts the general model for some models, and the system modeling of VPP is not the focus, so this article will not elaborate on it, for details refer to literature42,43. Model framework introduction An iterative two-stage distributed robust optimization with a separate price-coordination module is used to coordinate a multi-energy VPP cluster. The framework diagram is in Fig. 3. Bilateral exchanges of electricity and oxygen are permitted, and antisymmetric constraints ensure pairwise conservation. In each iteration, the master problem selects device capacities and enforces an upper bound on the expected operating cost; given these capacities, a joint 24-h scheduling subproblem is solved for all historical scenarios to obtain per-scenario costs and planned P2P trades while satisfying balances, unit limits, and buy–sell exclusivity. Scenario probabilities are updated within a dual-norm ambiguity set to yield the worst-case expected cost and cut-and-column generation proceeds until the upper and lower bounds meet. The scheduled trades are then priced in a Nash-bargaining/ADMM layer, where bilateral prices are aligned within a price floor and time-of-use ceiling, leaving physical dispatch unchanged. The formulation is indexed by the VPP set and scales from three to any finite number of participants without altering the model structure.Fig. 3A Distributed Robust Optimization Scheduling Framework for Multi-Energy VPP Clusters. A Distributed Robust Optimization Scheduling Framework for Multi-Energy VPP Clusters. VPP system cost model (1) Construction cost and operation cost (1) Construction cost and operation cost (1) Construction cost and operation cost In the VPP system, the system cost consists of construction cost and operation cost, where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{b}^{r}$$\end{document} is the system construction cost, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{r}^{i}$$\end{document} is the operation cost, and the calculations of construction cost and operation cost are as shown below:1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left\{ \begin{gathered} f_{b}^{r} = \frac{{P_{ri,GT} M_{GT,i} + P_{ri,AC} M_{AC,i} }}{{T_{years} (1 + \gamma )}} \hfill \\ \, + \frac{{P_{ri,GB} M_{GB,i} + P_{ri,EC} M_{EC,i} }}{{T_{years} (1 + \gamma )}} \hfill \\ \, + \frac{{P_{ri,pip} L_{pip} + P_{ri,VPSA} M_{VPSA} }}{{T_{years} (1 + \gamma )}} \hfill \\ f_{r}^{i} = C_{buy,i} + C_{dr,i} \hfill \\ C_{buy,i} = f_{e} + f_{gas} \hfill \\ \end{gathered} \right.$$\end{document} In Eq. (1): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,GT}$$\end{document} is the comprehensive unit cost of the GT (CNY/kW), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GT,i}$$\end{document} is the actual construction capacity of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} GT, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,AC}$$\end{document} is the comprehensive unit cost of the absorption chiller (CNY/kW), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{AC,i}$$\end{document} is the actual construction capacity of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} absorption chiller, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,GB}$$\end{document} is the comprehensive unit cost of the waste heat boiler (CNY/kW), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GB,i}$$\end{document} is the actual construction capacity of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} waste heat boiler, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,EC}$$\end{document} is the comprehensive unit cost of the electric refrigerator (CNY/kW), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{EC,i}$$\end{document} is the actual construction capacity of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} electric refrigerator, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,pip}$$\end{document} is the comprehensive cost per unit of the pipeline (CNY/km),\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{pip}^{i}$$\end{document} is the length of the oxygen transport pipeline constructed by the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} VPP, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,VPSA}$$\end{document} is the comprehensive unit cost of VPSA (CNY/kW), \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{VPSA}^{i}$$\end{document} is the actual construction capacity of VPSA of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} VPP, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T_{years}$$\end{document} is the number of days in a year, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma$$\end{document} is the discount rate, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{buy,i}$$\end{document} is the electricity purchase and gas purchase costs of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} VPP, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{dr,i}$$\end{document} is the demand response cost of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} VPP.(2) Electricity purchase and sale costs and gas purchase costs (2) Electricity purchase and sale costs and gas purchase costs (2) Electricity purchase and sale costs and gas purchase costs The cost of buying natural gas and electricity is shown below: The cost of buying and selling electricity:2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{e} = \sum\limits_{t = 1}^{24} {(P_{ri,buy}^{t} P_{buy,e}^{t} - P_{ri,sell}^{t} P_{sell,e}^{t} )}$$\end{document} In Eq. (2): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{e}$$\end{document} is the cost of electricity purchase, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,buy}^{t}$$\end{document} is the electricity purchase price, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{buy,e}^{t}$$\end{document} is the quantity of electricity purchased, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,sell}^{t}$$\end{document} is the sell electricity price, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{sell,e}^{t}$$\end{document} is the electricity sales. The cost of buying gas:3\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{gas} = \sum\limits_{t = 1}^{24} {P_{ri,gas}^{t} } P_{buy,gas}^{t}$$\end{document} In Eq. (3): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{gas}$$\end{document} is the cost of gas purchase, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,gas}^{t}$$\end{document} is the gas purchase price,\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{buy,gas}^{t}$$\end{document} is the gas purchase volume.(3) Demand-side response cost4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{dr} = \delta_{tr} \sum\limits_{t = 1}^{24} {P_{tr}^{t} } + \delta_{cut} \sum\limits_{t = 1}^{24} {(P_{cut}^{t} + H_{cut}^{t} + O_{cut}^{t} )} P_{buy,gas}^{t}$$\end{document} (3) Demand-side response cost4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{dr} = \delta_{tr} \sum\limits_{t = 1}^{24} {P_{tr}^{t} } + \delta_{cut} \sum\limits_{t = 1}^{24} {(P_{cut}^{t} + H_{cut}^{t} + O_{cut}^{t} )} P_{buy,gas}^{t}$$\end{document} (3) Demand-side response cost4\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{dr} = \delta_{tr} \sum\limits_{t = 1}^{24} {P_{tr}^{t} } + \delta_{cut} \sum\limits_{t = 1}^{24} {(P_{cut}^{t} + H_{cut}^{t} + O_{cut}^{t} )} P_{buy,gas}^{t}$$\end{document} In Eq. (4): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\delta_{tr}$$\end{document} is the conversion coefficient of the demand side translation cost, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\delta_{cut}$$\end{document} is the conversion coefficient of the demand side reduction cost, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tr}^{t}$$\end{document} is the transferable electric load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{cut}^{t}$$\end{document} is the reducible electric load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$H_{cut}^{t}$$\end{document} is the reducible heat load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$O_{cut}^{t}$$\end{document} is the reducible oxygen load. VPP system normal operation constraint condition To ensure the safe and stable operation of VPP, relevant operation constraints must be met. The specific constraints are as follows. GT Upper and lower limit constraints on construction capacity:5\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GT}^{\min } \le M_{GT,i} \le M_{GT}^{\max }$$\end{document} In Eq. (5): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GT,i}$$\end{document} is the GT cast capacity of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{i}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GT}^{\max }$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GT}^{\min }$$\end{document} are the upper and lower limits of the construction capacity of GT. GB Upper and lower limit constraints on construction capacity:6\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GB}^{\min } \le M_{GB,i} \le M_{GB}^{\max }$$\end{document} In Eq. (6): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GB,i}$$\end{document} is the GB cast capacity of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{i}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GB}^{\max }$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GB}^{\min }$$\end{document} are the upper and lower limits of the construction capacity of GB. (3) AC Upper and lower limit constraints on construction capacity:7\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{AC}^{\min } \le M_{AC,i} \le M_{AC}^{\max }$$\end{document} In Eq. (7): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{AC,i}$$\end{document} is the AC cast capacity of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{i}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{AC}^{\max }$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{AC}^{\min }$$\end{document} are the upper and lower limits of the construction capacity of AC. (4) EC Upper and lower limit constraints on construction capacity:8\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{EC}^{\min } \le M_{EC,i} \le M_{EC}^{\max }$$\end{document} In Eq. (8): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{EC,i}$$\end{document} is the EC cast capacity of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{i}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{EC}^{\max }$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{EC}^{\min }$$\end{document} are the upper and lower limits of the construction capacity of EC. (5) Upper and lower limit constraints on VPSA construction capacity:9\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{VPSA}^{\min } \le M_{VPSA,i} \le M_{VPSA}^{\max }$$\end{document} In Eq. (9): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{VPSA,i}$$\end{document} is the VPSA cast capacity of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{i}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{VPSA}^{\max }$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{VPSA}^{\min }$$\end{document} are the upper and lower limits of the construction capacity of VPSA. (6) The response part of the electricity/heat/oxygen load demand. Electricity load demand response balance constraint:10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{e} = L_{e0} - P_{e}^{cut} + P_{e}^{tran}$$\end{document} In Eq. (10): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{e}$$\end{document} is the actual electrical load after demand response, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{e0}$$\end{document} is the electrical load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{cut}$$\end{document} is the electrical load that VPP can reduce, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{tran}$$\end{document} is the electrical load that VPP can transfer. Heat load demand response balance constraint:11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{h} = L_{h0} - P_{h}^{cut}$$\end{document} In Eq. (11): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{h}$$\end{document} is the actual heat load of VPP after demand response, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{h0}$$\end{document} is the heat load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{h}^{cut}$$\end{document} is the heat load that VPP can reduce. Oxygen load demand response balance constraint:12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{o} = L_{o0} + P_{o}^{VPSA} - O_{cut}^{t}$$\end{document} In Eq. (12): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{o}$$\end{document} is the actual oxygen load after demand response, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{o0}$$\end{document} is the demand oxygen load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{*{20}c} {P_{o}^{VPSA} } \\ \end{array}$$\end{document} is the oxygen production power of the VPSA device. Constraints on the upper and lower limits of reducible electrical power:13\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha L_{e0} \le P_{e}^{cut} \le 0$$\end{document} In Eq. (13): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{e0}$$\end{document} is the electrical load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{cut}$$\end{document} is the electrical load that VPP can reduce, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} is the percentage of electrical power reduction. Constraints on the upper and lower limits of transferable electrical power:14\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$- \alpha L_{e0} \le P_{e}^{tran} \le \alpha L_{e0}$$\end{document} In Eq. (14): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{tran}$$\end{document} is the electrical load that VPP can transfer. Constraints on the upper and lower limits of reducible heat power:15\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$0 \le P_{h}^{DR} \le \beta L_{h0}$$\end{document} In Eq. (15): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta$$\end{document} is the percentage of reducible heat load. Constraint that the total amount of transferred electrical load and oxygen load is 0:16\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sum\limits_{i = 1}^{24} {P_{e}^{tran} } = 0$$\end{document}17\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sum\limits_{i = 1}^{24} {P_{o}^{tran} } = 0$$\end{document} In Eq. (17): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{o}^{tran}$$\end{document} is the transferred oxygen load. (7) Electricity/Heat/Cooling load balance constraint: Electric load:18\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{GT} + P_{e}^{rew} + P_{buy} = P_{sell} + L_{e} + P_{e}^{EC} + P_{e}^{VPSA}$$\end{document} In Eq. (18): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{GT}$$\end{document} is the electrical power of the GT, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{rew}$$\end{document} is the electrical power of renewable energy, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{EC}$$\end{document} is the electrical power consumed by the EC, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{buy}$$\end{document} is the electrical power purchased from the external power grid, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{sell}$$\end{document} is the electrical power sold to the external power grid, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{VPSA}$$\end{document} is the power consumption of the VPSA device. Heat load:19\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{h}^{GT} + P_{h}^{GB} = L_{h} + P_{h}^{AC}$$\end{document} Cool load:20\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{c}^{EC} + P_{c}^{AC} = L_{c}$$\end{document} Oxygen load:21\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{o}^{VPSA} = L_{o}$$\end{document} In Eqs. (19)-(21): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{h}^{GT}$$\end{document} is the heat power of the GT, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{h}^{GB}$$\end{document} is the heat power of the GB, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{h}^{AC}$$\end{document} is the heat absorption power of the AC, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{c}^{EC}$$\end{document} is the cooling power of the EC, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{c}^{AC}$$\end{document} is the cooling power of the AC, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{c}$$\end{document} is the cooling load, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{*{20}c} {P_{o}^{VPSA} } \\ \end{array}$$\end{document} is the oxygen production power of the VPSA equipment, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$L_{o}$$\end{document} is the oxygen load. (8) Power constraint of the connecting line22\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tran}^{\min } \le P_{tran} \le P_{tran}^{\max }$$\end{document} In Eq. (22): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tran}$$\end{document} is the power of the connecting line, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tran}^{\max }$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tran}^{\min }$$\end{document} are the upper and lower limits of the power of the connecting line. (9) The VPSA model is as follows:23\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left\{ {\begin{array}{*{20}l} {V_{VPSA}^{t} = \eta_{VPSA} \frac{{P_{e}^{VPSA} }}{{P_{VPSA,e} }}} \hfill \\ {P_{e,\min }^{VPSA} \le P_{e}^{VPSA} \le P_{e,\max }^{VPSA} } \hfill \\ {\Delta P_{VPSA}^{lp} \le P_{VPSA}^{t} - P_{VPSA}^{t - 1} \le \Delta P_{VPSA}^{up} } \hfill \\ \end{array} } \right.$$\end{document} In Eq. (23): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$V_{VPSA}^{t}$$\end{document} is the oxygen production volume per unit time, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{VPSA}$$\end{document} is the most economical efficiency of oxygen production, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e}^{VPSA}$$\end{document} is the electric power of the VPSA device, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{VPSA,e}$$\end{document} is the power consumption per unit, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e,\max }^{VPSA}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{e,\min }^{VPSA}$$\end{document} are the upper and lower limits of the VPSA’s electric power, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta P_{VPSA}^{up}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Delta P_{VPSA}^{lp}$$\end{document} are the upper and lower limits of the VPSA device’s ramp-up power. Double norm constraint model In order to ensure that the obtained probability distribution values are as close to the actual data as possible, this two-stage distributed robust optimization model builds a constraint method that centers on the initial probability distribution values and adopts 1-norm and the ∞-norm as constraint conditions., used to limit the range of probability distribution values in discrete cases. Their feasible domains are \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Omega_{1}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Omega_{\infty }$$\end{document} respectively. Wind and PV are treated as exogenous, scenario-based injections (10 historical scenarios × 24 h per VPP); scenario probabilities vary within the dual-norm ambiguity set specified below, and the worst-case expected cost is optimized in the outer loop.24\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Omega = \left\{ {\begin{array}{*{20}l} {P_{i} \ge 0,\;i = 1,2,3, \ldots ,k} \hfill \\ {\sum\limits_{i = 1}^{k} {P_{i} } = 1} \hfill \\ {\Pr \left\{ {\sum\limits_{i = 1}^{k} {\left| {P_{i} - P_{i}^{0} } \right|} \le \theta_{1} } \right\} \ge \alpha_{1} } \hfill \\ {\Pr \left\{ {\max \left| {P_{i} - P_{i}^{0} } \right| \le \theta_{\infty } } \right\} \ge \alpha_{\infty } } \hfill \\ \end{array} } \right.$$\end{document} In Eq. (24): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{i}$$\end{document} is the scenario probability, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\sum\limits_{i = 1}^{k} {\left| {P_{i} - P_{i}^{0} } \right|} \le \theta_{1}$$\end{document} is \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Omega_{1}$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\max \left| {P_{i} - P_{i}^{0} } \right| \le \theta_{\infty }$$\end{document} is \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\Omega_{\infty }$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta_{1}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\theta_{\infty }$$\end{document} are the maximum deviation values of the probability, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha_{1}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha_{\infty }$$\end{document} denote the confidence level of the probability. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\{ P_{k} \}$$\end{document} satisfy the following confidence level:25\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\{ P_{k} \} = \left\{ {\begin{array}{*{20}l} {\Pr \left\{ {\sum\limits_{i = 1}^{k} {\left| {P_{i} - P_{i}^{0} } \right|} \le \theta_{1} } \right\} \ge 1 - 2ke^{{ - \frac{{2M\theta_{1} }}{k}}} } \hfill \\ {\Pr \left\{ {\max \left| {P_{i} - P_{i}^{0} } \right| \le \theta_{\infty } } \right\} \ge 1 - 2ke^{{ - 2M\theta_{\infty } }} } \hfill \\ \end{array} } \right.$$\end{document} In Eq. (25): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M$$\end{document} represents the number of scenarios, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$K$$\end{document} represents the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$K$$\end{document} scenario. Based on the above, we can obtain:26\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left\{ {\begin{array}{*{20}l} {\theta_{1} = \frac{k}{2M}\ln \frac{2k}{{1 - \alpha_{1} }}} \hfill \\ {\theta_{\infty } = \frac{1}{2M}\ln \frac{2k}{{1 - \alpha_{\infty } }}} \hfill \\ \end{array} } \right.$$\end{document} Two-stage distributed robust optimization model The unpredictable nature of distributed robust optimization outcomes and forecasting inaccuracies can negatively impact VPP cluster planning and operations. To address this challenge, a novel VPP cluster planning and scheduling approach was developed, leveraging dual norm constraints for enhanced reliability. This paper decomposes the Nash negotiation cluster scheduling model into three problems, namely the main problem (MP) and two sub-problems (SP1 and SP2). The main problem (MP) aims to establish the lower limit of the objective function, that is \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$LB$$\end{document}.27\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left\{ {\begin{array}{*{20}l} {LB = \left\{ {\min \left( {f_{b}^{r} + \eta } \right),LB} \right\}} \hfill \\ {UB = \left\{ {\left( {f_{b}^{r} + \max \sum\limits_{i = 1}^{k} {\left( {P_{i}^{T + 1} \min f_{r}^{i} } \right)} } \right),UB} \right\}} \hfill \\ {\eta \ge \sum\limits_{i = 1}^{k} {P_{i}^{T} } f_{r}^{i} } \hfill \\ {P_{i}^{T} = P_{i}^{0} ,\quad if \, T = 1} \hfill \\ \end{array} } \right.$$\end{document} In Eq. (27): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$LB$$\end{document} is the lower bound of the initial stage operating cost, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$UB$$\end{document} is the upper bound of the second stage operating cost, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{i}^{T} (T \ge 1)$$\end{document} is the scenario probability value of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$T$$\end{document} generation, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{i}^{0}$$\end{document} is the initial probability value of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} VPP as shown in Table 1, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{b}^{r}$$\end{document} is the daily average construction cost, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$f_{r}^{i}$$\end{document} is the operating cost, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta$$\end{document} is an auxiliary variable.Table 1Initial scenario probability values.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document}Scene 10.170.070.06Scene 20.100.050.10Scene 30.090.120.12Scene 40.110.040.16Scene 50.070.060.09Scene 60.090.040.08Scene 70.050.070.08Scene 80.120.050.06Scene 90.120.310.09Scene 100.080.090.16 Table 1Initial scenario probability values.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document}\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document}Scene 10.170.070.06Scene 20.100.050.10Scene 30.090.120.12Scene 40.110.040.16Scene 50.070.060.09Scene 60.090.040.08Scene 70.050.070.08Scene 80.120.050.06Scene 90.120.310.09Scene 100.080.090.16 Initial scenario probability values. As can be seen from the above, compared with the traditional robust optimization model, Unlike conventional robust optimization approaches that focus solely on minimizing risks in the most extreme circumstances, while the two-stage distributed robust optimization model solves the first stage objective function under the worst-case scenario through the lower bound \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$LB$$\end{document}, estimates its economic operation through \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\min f_{r}^{i}$$\end{document}, determine the worst-case operational scenarios that the cluster is likely to encounter and update the scenario probability values through \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\max \sum\limits_{i = 1}^{k} {\left( {P_{i}^{T + 1} \min f_{r}^{i} } \right)}$$\end{document}, synergizing economy and resilience through constant repetition. VPP Group bargaining model based on nash negotiation Through Nash negotiation, all parties can achieve a win–win situation. Assume that the three VPPs in this paper belong to different stakeholders. The VPP cluster bargaining model based on Nash negotiation in this paper is shown below.28\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left\{ {\begin{array}{*{20}l} {\max \prod\limits_{i = 1}^{k} {\left( {U_{i} - U_{i}^{0} } \right)} } \hfill \\ {U_{1} \ge U_{1}^{0} } \hfill \\ {U_{2} \ge U_{2}^{0} } \hfill \\ {U_{3} \ge U_{3}^{0} } \hfill \\ \end{array} } \right.$$\end{document} In Eq. (28): \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$U_{i}$$\end{document} is the profit of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} VPP after Nash negotiation, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$U_{i}^{0}$$\end{document} is the profit of the \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} VPP before participating in the Nash negotiation.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$U_{i} - U_{i}^{0}$$\end{document} is the increased profit compared to before participating in the Nash negotiation. To maximize the model, needs \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {U_{1} - U_{1}^{0} } \right)\left( {U_{2} - U_{2}^{0} } \right)\left( {U_{3} - U_{3}^{0} } \right)$$\end{document} to be maximized, that is, each of the three needs to be maximized, which is to maximize \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\left( {U_{1} + U_{2} + U_{3} } \right) - \left( {U_{1}^{0} + U_{2}^{0} + U_{3}^{0} } \right)$$\end{document}, where \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$U_{1}^{0} + U_{2}^{0} + U_{3}^{0}$$\end{document} is a constant. If \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$U_{1} + U_{2} + U_{3}$$\end{document} is a constant, then it can meet the condition of maximizing the mean inequality. According to the above derivation, the Lagrangian remainder form of the ADMM algorithm can be finally obtained, as shown in the following formula:29\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\begin{array}{*{20}c} { - \min \left[ \begin{gathered} \ln \left( {\omega_{i}^{*} + C_{i}^{S} - U_{i}^{0} } \right) \hfill \\ + \sum\limits_{t = 1}^{T} {\lambda_{i - g} } (pri_{i - g}^{e} - pri_{g - i}^{e} ) \hfill \\ + \frac{\rho }{2}\left\| {pri_{i - g}^{e} - pri_{g - i}^{e} } \right\|_{2}^{2} \hfill \\ + \sum\limits_{t = 1}^{T} {\lambda_{i - g} } (pri_{i - g}^{o} - pri_{g - i}^{o} ) \hfill \\ + \frac{\rho }{2}\left\| {pri_{i - g}^{o} - pri_{g - i}^{o} } \right\|_{2}^{2} \hfill \\ \end{gathered} \right]} \\ {{\text{ s}}{\text{.t}}{.}\;\left\{ {\begin{array}{*{20}l} {\omega_{i}^{*} + C_{i}^{S} - U_{i}^{0} \ge 0} \hfill \\ {pri_{\min }^{e} \le pri_{i - g}^{e} \le pri_{\max }^{e} } \hfill \\ {pri_{\min }^{o} \le pri_{i - g}^{o} \le pri_{\max }^{o} } \hfill \\ \end{array} } \right.} \\ \end{array}$$\end{document} In Eq. (29):\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i \ne g$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$g,T \in i$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$pri_{i - g}^{e}$$\end{document} is the electricity price from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$g$$\end{document}, conversely, it is the price of electricity from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$g$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{\min }^{e}$$\end{document} is the minimum electricity price between the two VPPs, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{\max }^{e}$$\end{document} is the maximum electricity price between the two VPPs, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$pri_{i - g}^{o}$$\end{document} is the price of oxygen from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$g$$\end{document}, conversely, it is the price of oxygen from \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$g$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$i$$\end{document},\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{\min }^{o}$$\end{document} is the minimum interaction oxygen price between the two VPPs, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{\max }^{o}$$\end{document} is the maximum trading oxygen price between the two VPPs, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\omega_{i}^{*}$$\end{document} is the optimal solution for the sum of running and upkeep expenses and the external interaction costs, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$C_{i}^{S}$$\end{document} is the energy interaction cost between the various VPPs. Each VPP in this model employs the same local decision module. Energy sharing between VPPs is solely reflected through pairwise exchanges and the existing global exchange balancing constraint. Consequently, the model can be scaled to n VPPs. Case analysis Device related parameters The various parameters of the VPP and real-time electricity prices, and natural gas and oxygen price tables are shown in Tables 2, 3, 4, 5: Table 2Parameters of the VPP system.ParametricNumericalParametricNumericalParametricNumerical \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tran}^{\max }$$\end{document} 500 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tran}^{\min }$$\end{document} −500 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GT,e}$$\end{document} 0.35 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GT,h}$$\end{document} 0.42 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GB,h}$$\end{document} 0.9 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$COP_{EC}$$\end{document} 4 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$COP_{AC}$$\end{document} 0.83 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma$$\end{document} 0.08 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} 0.15 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta$$\end{document} 0.2 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{VPSA}$$\end{document} 0.85 Table 3Upper limit of VPP equipment construction capacity.Name \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GT}^{\max }$$\end{document} 500030003000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{AC}^{\max }$$\end{document} 200020002000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{EC}^{\max }$$\end{document} 100010001000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GB}^{\max }$$\end{document} 800500500 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{VPSA}^{\max }$$\end{document} 1000700700 Table 4Comprehensive Cost per Unit of VPP Equipment.NameNumerical(CNY/kW) \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,GT}$$\end{document} 5000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,AC}$$\end{document} 1000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,GB}$$\end{document} 3000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,EC}$$\end{document} 3000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,VPSA}$$\end{document} 4500 Table 5Real-Time Electricity, Natural Gas and Oxygen Price.CategoriesTime periodPricePrice of electricity00: 00–7: 0022: 00–24: 000.4 CNY/kWh7: 00–11: 0014: 00–19: 000.75CNY/kWh11: 00–14: 0019: 00–22: 001.2CNY/kWhGas00: 00–24: 002.2CNY/\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$m^{3}$$\end{document}Oxygen00: 00–24: 001.5CNY/\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$m^{3}$$\end{document} Table 2Parameters of the VPP system.ParametricNumericalParametricNumericalParametricNumerical \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tran}^{\max }$$\end{document} 500 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{tran}^{\min }$$\end{document} −500 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GT,e}$$\end{document} 0.35 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GT,h}$$\end{document} 0.42 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GB,h}$$\end{document} 0.9 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$COP_{EC}$$\end{document} 4 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$COP_{AC}$$\end{document} 0.83 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\gamma$$\end{document} 0.08 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha$$\end{document} 0.15 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\beta$$\end{document} 0.2 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{VPSA}$$\end{document} 0.85 Parameters of the VPP system. Table 3Upper limit of VPP equipment construction capacity.Name \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GT}^{\max }$$\end{document} 500030003000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{AC}^{\max }$$\end{document} 200020002000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{EC}^{\max }$$\end{document} 100010001000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{GB}^{\max }$$\end{document} 800500500 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$M_{VPSA}^{\max }$$\end{document} 1000700700 Upper limit of VPP equipment construction capacity. Table 4Comprehensive Cost per Unit of VPP Equipment.NameNumerical(CNY/kW) \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,GT}$$\end{document} 5000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,AC}$$\end{document} 1000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,GB}$$\end{document} 3000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,EC}$$\end{document} 3000 \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{ri,VPSA}$$\end{document} 4500 Comprehensive Cost per Unit of VPP Equipment. Table 5Real-Time Electricity, Natural Gas and Oxygen Price.CategoriesTime periodPricePrice of electricity00: 00–7: 0022: 00–24: 000.4 CNY/kWh7: 00–11: 0014: 00–19: 000.75CNY/kWh11: 00–14: 0019: 00–22: 001.2CNY/kWhGas00: 00–24: 002.2CNY/\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$m^{3}$$\end{document}Oxygen00: 00–24: 001.5CNY/\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$m^{3}$$\end{document} Real-Time Electricity, Natural Gas and Oxygen Price. 00: 00–7: 00 22: 00–24: 00 7: 00–11: 00 14: 00–19: 00 11: 00–14: 00 19: 00–22: 00 In the above table, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GT,e}$$\end{document} is the power generation efficiency of the GT, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GB,h}$$\end{document} is the thermal efficiency of the GB, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$COP_{EC}$$\end{document} is the efficiency of the electric chiller, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$COP_{AC}$$\end{document} is the efficiency of the absorption chiller, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\eta_{GT,h}$$\end{document} is the heat production efficiency of the GT. In the above table, the lower limit of the construction capacity of each equipment is 0. Cost analysis of independent and joint operation of each VPP As shown in Table 6, a cost comparison has been made between independent and joint operation of each VPP. After calculation, the operating costs in independent operation are 128442.92 CNY, 66016.72 CNY, and 77823.32 CNY respectively. After the joint operation, the operating costs of each VPP are 104852.84 CNY, 86597.09 CNY, and 70561.29 CNY respectively. The respective income of each VPP increases by 23590.08 CNY, −20580.37 CNY, 7262.03 CNY. The rates of increase in income for each VPP are 18.37%, −31.17%, and 9.33%, respectively. The total cost of independent operation is 272282.96 CNY, the total cost of joint operation is 262011.22 CNY, after joint operation, the total income increases by 10271.74 CNY, and the total income rate is 3.77%. The data also further indicates that the control strategy proposed in this paper enhances the income of VPP cluster and ensures fairness in the distribution of income among the various parties.Table 6Comparison of operating costs of independent and joint operation of each VPP.VPP NumberStand-alone running costs/CNYJoint operating costs/CNYBenefits enhancement/CNYBenefit enhancement ratio\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}128,442.92104,852.8423,590.0818.37%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document}66,016.7286,597.09−20,580.37−31.17%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document}77,823.3270,561.297262.039.33% Table 6Comparison of operating costs of independent and joint operation of each VPP.VPP NumberStand-alone running costs/CNYJoint operating costs/CNYBenefits enhancement/CNYBenefit enhancement ratio\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}128,442.92104,852.8423,590.0818.37%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document}66,016.7286,597.09−20,580.37−31.17%\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document}77,823.3270,561.297262.039.33% Comparison of operating costs of independent and joint operation of each VPP. In the simulation analysis, this paper uses the Yalmip toolbox in MATLAB software to call the CPLEX solver for solutions. The computer parameters are AMD(R7) 3.20 GHz and 16 GB RAM. Convergence analysis of the algorithm This paper is based on the Nash bargaining theory and obtains the interactive electricity price through the ADMM algorithm. The objective function is a multiple-entity single-variable, and the constraint conditions are the electricity price constraint, oxygen price constraint, and the constraint that the income of a single VPP system is greater than 0. Figure 4 is the iterative convergence graph of each VPP and the distributed robust optimization algorithm. The distributed robust optimization algorithm converges after 7 iterations with a convergence accuracy of 0.01. The algorithm detailed herein demonstrates efficient computation and convergence, adequately fulfilling Virtual Power Plant (VPP) dispatch requirements..Fig. 4VPP iteration convergence. VPP iteration convergence. Instance data This paper takes the cooperative optimization between three VPPs as an example. Considering the diversity and differences between each VPP include wind farms, and include PVs. Where the confidence level is set to 0.5 and 0.99, the scenario is set to 10. The output of VPP wind farms and PVs is shown in Figure 5.Fig. 5VPP Wind and PV Outputs. VPP Wind and PV Outputs. From Figure 5, we can see that the wind power output of all day, overall, the wind power output is large at night and small during the day. Therefore, the wind power generation at night can meet the electricity demand of the remaining VPPs. Due to the participation in power generation by PVs and the relatively few wind power during the day, the wind power output is relatively small during the day. The electrical, thermal, cooling and oxygen load power diagrams for each VPP are shown in Figure 6:Fig. 6(a) \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} load power. (b) \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} load power. (c) \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} load power. (a) \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} load power. (b) \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} load power. (c) \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} load power. Analysis of example results Analysis of Energy Interaction Volumes and Interaction Prices between Each VPP The interaction results of electrical energy between each VPP are shown in Figure 7. The analysis of this figure is as follows:Fig. 7Interactive electrical energy between VPPs. Interactive electrical energy between VPPs. Taking VPP12 as the research object. From the figure, it can be seen that during 00:00–01:00, 04:00–05:00, and 07:00–19:00, the value of VPP12 is positive, indicating that during these times, the power output of system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} generation unit does not meet its own electricity load demand. Therefore, the other VPPs supply some of their electricity to system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}. During 01:00–02:00, 06:00–07:00, and 19:00–24:00, the value of VPP12 is negative, indicating that during these times, the power output of system VPP1 generation unit meets its own demand. Thus, system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} supplies excess electricity to the other VPPs to meet their electricity load demands. The electricity interactions between the other VPPs follow the same principle and will not be elaborated here. The transaction tariffs between each VPP as well as the grid tariffs are shown in Fig. 8, from which it can be seen that the tariffs of each VPP after the interaction are much lower than the grid tariffs. The tariffs of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} fluctuate over time, with the highest point occurring at 12:00–15:00 at 1.2 CNY/kWh. The tariffs of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} as well as \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} to \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} both fluctuate over time, but they are also much lower than the grid tariffs.Fig. 8Interactive tariffs between VPPs. Interactive tariffs between VPPs. The interaction volumes of oxygen for each VPP are shown in Fig. 9. From this figure, it can be seen that when VPP12 is positive, it indicates that system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} is supplying oxygen to system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document}. Conversely, when VPP12 is negative, it indicates that system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} is supplying oxygen to system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}. The same applies to VPP13. VPP23 remains at 0 throughout, indicating that systems \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} only need to interact with system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} to meet their oxygen load demands and do not require oxygen interaction with the other VPPs.Fig. 9Interactive oxygen between VPPs. Interactive oxygen between VPPs. The oxygen trading prices for each VPP are shown in Fig. 10. From the figure, it can be seen that the interaction prices of oxygen are much lower than the market price. The oxygen interaction price between system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} and system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} is a stable 0.2 CNY/m3. The oxygen interaction price between system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} and system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} fluctuates over time, with the highest price being 1.2 CNY/m3 and the lowest price being 0.2 CNY/m3. Since there is no oxygen interaction between system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} and system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document}, the price is 0.Fig. 10Oxygen price of interaction between VPPs. Oxygen price of interaction between VPPs. Power Balance Analysis of Each VPP. Taking the first scenario as an example, the electricity, heat, and oxygen power balance for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} is shown in Figs. 10, 11, and 12. The power balance diagrams of other VPPs can be found in Appendix A.Fig. 11\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} Electrical power balance.Fig. 12\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} Thermal power balance. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} Electrical power balance. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} Thermal power balance. Analyzing Fig. 11, it can be seen that the GT and renewable energy (wind power) generate power throughout the day. From 14:00–17:00,the electricity for \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} is purchased by the system from the outside grid. From 00:00–08:00 and 18:00–24:00, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} excess electricity is supplied to systems \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} to ensure their normal operation. From 08:00–18:00, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} excess electricity is supplied to system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} to ensure its normal operation. From 07:00 to 10:00, 12:00–13:00, and 20:00–24:00, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} electricity is sold to the grid. The electric chiller and VPSA equipment work almost all day to meet cooling and oxygen load demands. Comparing the power load curve before and after demand response, it can be seen that the peak of the power load curve decreases after demand response, with multiple peak points appearing, which reduces the impact on the system and thus improves system stability. Analyzing Fig. 12, as may be observed, the GT generates thermal power throughout the day due to its continuous involvement in power generation. Since the heat generated by the GT is sufficient for thermal load, the waste heat boiler does not produce additional heat. Meanwhile, the absorption chiller runs almost all day, with relatively stable power output to meet cooling load demands. Contrasting the heat load curves prior to and following demand response installation, after demand response is put into place, it is evident that the heat load curve’s peak value drastically drops, and the heat load curve becomes smoother than before. This change helps to improve the stability of system operation. Analyzing Fig. 13, The VPSA plant operates nearly all day to meet the oxygen loading requirements of \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}. Meanwhile, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} from 00:00–03:00 and 20:00–24:00, the excess oxygen is supplied to system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} From 03:00–04:00, 05:00–07:00, 16:00–20:00, and 23:00–01:00, oxygen is supplied to system \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document}. During the remaining time, \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} the excess oxygen is supplied to systems \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{2}$$\end{document} and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{3}$$\end{document} to meet their oxygen load demands. Comparing the oxygen load before and after demand response, it is evident that the peak of the oxygen load curve decreases after demand response, which reduces the impact on the system and improves stability.Fig. 13\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} oxygen power balance. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$VPP_{1}$$\end{document} oxygen power balance. Analysis of operating costs and benefits of various VPPs Comparison of Operating Costs under Double-Norm and Single-Norm. In distributed robust optimization, different confidence levels can result in varying degrees of cluster conservatism. This paper analyzes the model calculation results by setting different confidence intervals. As shown in Tables 7 and 8,When dealing with unpredictable power systems, greater investment leads to increased operational caution, reduced risk exposure, and larger capacity buffers. This approach compensates for the substantial output fluctuations inherent in distributed renewable sources while addressing the persistent challenges of wind and solar energy curtailment.Table 7Comparison between double-norm and 1-norm.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha_{1}$$\end{document}Double-Norm/CNY1-Norm/CNY0.2015,17515,2070.5015,17815,2370.9015,18015,304Table 8Comparison between double-norm and ∞-norm.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha_{\infty }$$\end{document}Double-Norm/CNY∞-Norm/CNY0.5015,23715,3100.9015,23715,3490.9915,17815,181 Table 7Comparison between double-norm and 1-norm.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha_{1}$$\end{document}Double-Norm/CNY1-Norm/CNY0.2015,17515,2070.5015,17815,2370.9015,18015,304 Comparison between double-norm and 1-norm. Table 8Comparison between double-norm and ∞-norm.\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\alpha_{\infty }$$\end{document}Double-Norm/CNY∞-Norm/CNY0.5015,23715,3100.9015,23715,3490.9915,17815,181 Comparison between double-norm and ∞-norm. (1) Comparison between Double-Norm and 1-Norm: (2) Comparison between Double-Norm and ∞-Norm: By comparing Table 7, it is found that when the confidence level of the 1-norm increases from 0.20 to 0.90, the operating cost under the double -norm constraint decreases by 32 CNY, 59 CNY, and 124 CNY, respectively, compared to the operating cost under the 1-norm constraint. By comparing Table 8, it is found that when the confidence level of the ∞-norm increases from 0.50 to 0.99, the operating cost under the double -norm constraint decreases by 73 CNY, 112 CNY, and 3 CNY, respectively, compared to the operating cost under the ∞-norm constraint. The double-norm constraint jointly bounds aggregate and per-scenario probability deviations, yielding a tighter and more balanced ambiguity set that mitigates overconcentration on extreme scenarios and produces robust yet not overly conservative decisions.Their intersection (double-norm) restricts both dimensions simultaneously. Under identical confidence levels, the intersection moderates the adversarial reweighting of scenarios and avoids concentrating probability mass solely through total-mass relocation or through a few extreme scenarios. A smaller risk premium is therefore observed, and the worst-case expected operating cost is slightly lower under the double-norm than under the single-norm cases. Operationally, this corresponds to hedging that reduces opportunistic selling, slightly increases self-supply at the margin, and moderates P2P exchanges, while all device limits and feasibility constraints remain satisfied. In summary, the operating cost under double-norm constraint is lower than that under single-norm constraint. Sensitivity analysis of key parameters As shown in Tables 9 and 10, a partial sensitivity analysis is performed to evaluate the objective function’s response to price and capacity perturbations. Disturbances are implemented through influence factors. The influence factors for electricity sales prices are 0.6, 0.7, and 0.8, while those for natural gas procurement prices are 1.8, 2.2, and 2.6. The influence factors for all equipment construction capacity boundaries are uniformly 0.8, 1.0, and 1.2. By analyzing the upper bound of the second stage operating cost \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$UB$$\end{document}, assess the sensitivity of key parameters.Table 9Price parameter sensitivity.Factors influencing electricity sales pricesUB/CNYFactors affecting gas purchase priceUB/CNY0.6260,173.111.8244,783.220.7258,741.162.2258,741.160.8257,384.122.6272,656.75Table 10Capacity limit sensitivity.Equipment Investment Capacity Boundary Adjustment FactorUB/CNY0.8258,797.821.0258,741.161.2258,740.48 Table 9Price parameter sensitivity.Factors influencing electricity sales pricesUB/CNYFactors affecting gas purchase priceUB/CNY0.6260,173.111.8244,783.220.7258,741.162.2258,741.160.8257,384.122.6272,656.75 Price parameter sensitivity. Table 10Capacity limit sensitivity.Equipment Investment Capacity Boundary Adjustment FactorUB/CNY0.8258,797.821.0258,741.161.2258,740.48 Capacity limit sensitivity. Results indicate that increasing the electricity sales price slightly reduces \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$UB$$\end{document}, with relatively weak price sensitivity. A change from 0.6 to 0.8 results in \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$UB$$\end{document} ranging from −0.53% to + 0.55%. The impact of fluctuating investment equipment capacity within a 20% range above and below the boundary is negligible. By contrast, varying the gas price between 1.8 and 2.6 leads to ≈ ± 5.4% changes around the baseline (244,783.22 to 272,656.75 CNY). Overall, the objective exhibits low sensitivity to selling-price discounts and capacity bounds within the tested ranges, and moderate sensitivity to gas prices—confirming that the proposed distributed robust strategy maintains feasibility and controls cost fluctuations under real-world disturbances, ensuring the model’s reliability under various conditions30. Conclusion This paper constructs a multi-energy VPP cluster using multiple sources such as electricity, heat, cooling, and oxygen. Based on a robust optimization model with double -norm constraints and the Nash bargaining model, an optimal operation model for the multi-energy VPP cluster is established. The conclusions drawn are as follows:(1) Multi-energy VPP Cluster for High Plateau Regions: To address the oxygen supply needs and energy development planning for high plateau regions, a multi-energy VPP cluster coupling electricity, heat, cooling, and oxygen is established. Assessing Tibet’s present energy conditions and national “dual carbon” goals, various energy devices are coupled, reduce the overall intensity of carbon emissions and improve the consumption capacity of renewable energy. Through a reasonable pricing mechanism, the enthusiasm of the demand side is improved without compromising the interests of all parties.(2) Under the dual-norm constraint robust optimization model, operating costs at different confidence levels are lower than those under single-norm constraints: Compared to the 1-norm, costs decrease by 32, 59, and 124 CNY respectively; compared to the ∞-norm, costs decrease by 73, 112, and 3 CNY respectively. This approach enables coordinated optimization between robustness and economic efficiency. Consequently, distributed robust optimization demonstrates excellent performance in addressing the uncertainty of renewable energy output within VPP clusters. Simultaneously, the two confidence intervals of the dual-norm serve as parameters reflecting the decision-maker’s risk preference. Lower conservatism yields lower operating costs, requiring decision-makers to select different confidence intervals based on their varying risk preferences.(3) Compared to operating independently, the multi-functional interaction and resource sharing within a cluster can reduce overall costs and enhance efficiency. By coordinating exchanges to smooth out temporal and spatial variations in renewable energy output, the model improves overall utilization rates and reduces curtailment of wind and solar power. Additionally, sensitivity analysis was conducted to validate the robustness of the results, ensuring the model’s reliability under various conditions. (1) Multi-energy VPP Cluster for High Plateau Regions: To address the oxygen supply needs and energy development planning for high plateau regions, a multi-energy VPP cluster coupling electricity, heat, cooling, and oxygen is established. Assessing Tibet’s present energy conditions and national “dual carbon” goals, various energy devices are coupled, reduce the overall intensity of carbon emissions and improve the consumption capacity of renewable energy. Through a reasonable pricing mechanism, the enthusiasm of the demand side is improved without compromising the interests of all parties. (1) Multi-energy VPP Cluster for High Plateau Regions: To address the oxygen supply needs and energy development planning for high plateau regions, a multi-energy VPP cluster coupling electricity, heat, cooling, and oxygen is established. Assessing Tibet’s present energy conditions and national “dual carbon” goals, various energy devices are coupled, reduce the overall intensity of carbon emissions and improve the consumption capacity of renewable energy. Through a reasonable pricing mechanism, the enthusiasm of the demand side is improved without compromising the interests of all parties. (2) Under the dual-norm constraint robust optimization model, operating costs at different confidence levels are lower than those under single-norm constraints: Compared to the 1-norm, costs decrease by 32, 59, and 124 CNY respectively; compared to the ∞-norm, costs decrease by 73, 112, and 3 CNY respectively. This approach enables coordinated optimization between robustness and economic efficiency. Consequently, distributed robust optimization demonstrates excellent performance in addressing the uncertainty of renewable energy output within VPP clusters. Simultaneously, the two confidence intervals of the dual-norm serve as parameters reflecting the decision-maker’s risk preference. Lower conservatism yields lower operating costs, requiring decision-makers to select different confidence intervals based on their varying risk preferences. (2) Under the dual-norm constraint robust optimization model, operating costs at different confidence levels are lower than those under single-norm constraints: Compared to the 1-norm, costs decrease by 32, 59, and 124 CNY respectively; compared to the ∞-norm, costs decrease by 73, 112, and 3 CNY respectively. This approach enables coordinated optimization between robustness and economic efficiency. Consequently, distributed robust optimization demonstrates excellent performance in addressing the uncertainty of renewable energy output within VPP clusters. Simultaneously, the two confidence intervals of the dual-norm serve as parameters reflecting the decision-maker’s risk preference. Lower conservatism yields lower operating costs, requiring decision-makers to select different confidence intervals based on their varying risk preferences. (3) Compared to operating independently, the multi-functional interaction and resource sharing within a cluster can reduce overall costs and enhance efficiency. By coordinating exchanges to smooth out temporal and spatial variations in renewable energy output, the model improves overall utilization rates and reduces curtailment of wind and solar power. Additionally, sensitivity analysis was conducted to validate the robustness of the results, ensuring the model’s reliability under various conditions. (3) Compared to operating independently, the multi-functional interaction and resource sharing within a cluster can reduce overall costs and enhance efficiency. By coordinating exchanges to smooth out temporal and spatial variations in renewable energy output, the model improves overall utilization rates and reduces curtailment of wind and solar power. Additionally, sensitivity analysis was conducted to validate the robustness of the results, ensuring the model’s reliability under various conditions. In conclusion, this paper establishes a VPP cluster for multi-energy flow interaction and sharing, including electricity, heat, and oxygen, to meet the national “dual carbon” goals and the energy development planning for Tibet. However, it is acknowledged that the work has limitations. For example: Excluding energy storage and EV flexibility, ignoring network losses, etc.Given the increasing renewable uptake and resource diversity in Tibet, the framework can be further extended. In future work, geothermal and biomass resources, storage, flexible EV charging with potential vehicle-to-grid services, and the water–energy nexus will be incorporated into the multi-energy interaction and sharing layer44–47. The aforementioned research further enhances the cluster efficiency of virtual power plants and the absorption rate of renewable energy. Supplementary Information Supplementary Information. Supplementary Information.