Uncertainty and Autarky: Cooperative Game Theory for Stable Local Energy Market Partitioning
Saurabh Vaishampayan, Maryam Kamgarpour · 2026
Read Uncertainty and Autarky: Cooperative Game Theory for Stable Local Energy Market Partitioning by Saurabh Vaishampayan, Maryam Kamgarpour (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-05169.txt.
Introduction
Saurabh Vaishampayan and Maryam Kamgarpour model how strategic prosumers in spread grids form groups for collective self-use under uncertainty. Their cooperative game method weighs grid-operator interests next to prosumer ones, taking in grid constraints, variable load and making, and links between nearby local markets. With fixed load and making, the largest group is the optimal stable partition, while heavy crowding makes individual self-use optimal. For stochastic prosumption with moderate congestion, they give an algorithm that finds the optimal stable split, checked on benchmark and real grids.
Catalog record
| CREATOR | Saurabh Vaishampayan, Maryam Kamgarpour |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 61357 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.05169
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Uncertainty and Autarky: Cooperative Game Theory for Stable Local Energy Market Partitioning Saurabh Vaishampayan, Maryam Kamgarpour Abstract Local energy markets empower prosumers to form coalitions for energy trading. However, the optimal partitioning of the distribution grid into such coalitions remains unclear, especially in constrained grids with stochastic production and consumption. This analysis must take into account the interests of both the grid operator and the constituent prosumers. In this work, we present a cooperative game theoretic framework to study distribution grid partitioning into local energy market coalitions under uncertain prosumption and grid constraints. We formulate the optimal stable partitioning problem to balance the interests of the grid operator with that of prosumers. Under deterministic load and generation, we show that the largest market coalition is the optimal stable partition. For the case of stochastic loads and generation, we provide an algorithm to evaluate the optimal stable partition. Numerical experiments are performed on benchmark and real world distribution grids. Our results help in understanding how uncertainty affects local energy market partitioning decisions in constrained distribution grids. I Introduction Recent legislations [ 15 ] empower prosumers in distribution grids to form Local Energy Markets (LEMs) for collective self-consumption. However, the scale and composition of the LEM coalitions that prosumers would form remains unclear. In a large distribution grid, such market coalitions could range anywhere from a small neighborhood to a coalition of all prosumers in the distribution grid. A distribution grid may have multiple such coalitions, so an open question is which grid partitions will emerge when prosumers are empowered to form LEM coalitions . To answer this, we must first evaluate the costs and benefits of LEMs. Since LEMs are typically incentivized for self-consumption, this amounts to assessing the impacts of self-consumption (analogous to the concept of autarky ) in different LEM configurations. Following this, one can analyze distribution grid partitioning into self-consuming LEM coalitions. The problem of grid partitioning must be studied from the perspective of two parties: 1. The Distribution System Operator (DSO), who manages the distribution grid and cares about aggregate costs for the grid; and 2. Prosumers, who have the power to form LEMs, and may only care about local costs. 1. From the DSO’s perspective , some degree of autarky via localized self-consumption may be preferred to reduce power flows in constrained grids. This can minimize the risk of line overloading and voltage violations under uncertain load and generation [ 6 ] [ 7 ] . On the other hand, such localized self-consumption may come at the expense of economic efficiency by comprising on globally optimal dispatch [ 13 ] . Grid partitioning may help the DSO balance the tradeoffs between risks associated with network constraint violations under uncertain load and generation, and economically efficient dispatch [ 4 ] . Note that studying grid partitioning only from the DSO’s perspective is not sufficient, since it is the prosumers who actually have the power to form LEMs. If the DSO selects a partition where the realized economic gains are not allocated appropriately to the prosumers, some of the prosumers may deviate to form alternative LEM coalitions. This brings us to the perspective of prosumers in analyzing LEM partitioning. 2. From the prosumers’ perspective , LEM formation is analyzed in the literature using cooperative game theory. Such works check the coalition stability of LEMs to predict LEM formation, by analyzing the existence of cost sharing methods that prevent any subset of prosumers from seeking alternative coalitions for economic benefit. However, current literature neglects either physical network constraints ( [ 9 , 18 ] ) or prosumption uncertainty ( [ 18 , 2 ] ) to show the stability of the largest coalition for energy dispatch. As a result, it is unclear why would prosumers prefer autarky, that is, form coalitions for self-consumption instead of grid-wide trading. This analysis is further complicated because power flow constraints introduce coupling between LEMs, meaning that LEMs’ costs not only depend on its internal configuration, but may also depend on the configuration of other LEMs in the grid. This necessitates the study of partition stability , where cost allocations must disincentivize alternative partition formation while accounting for cost coupling across coalitions. Research on partition stability in electric grids is limited. In the context of energy communities, the stable partitioning problem was used by the authors [ 1 ] to study energy community formation in grids with shared infrastructure costs. In [ 5 ] , stable partitioning of transmission grids into balancing groups was studied using partition function form games [ 17 ] . However, both works neglect the presence of uncertain prosumption and the risk of network constraint violations. Summarizing the related work, an open question that we attempt to answer is: How can the distribution grid be partitioned into LEMs to balance the interests of the DSO and the prosumers, under uncertain prosumption and grid constraints? The rest of the paper is organized as follows. Section II provides the preliminaries related to graph partitioning and power flow. In Section III , we formulate a two-stage model to analyze total operating costs in a grid partition under uncertain prosumption. These reflect the total costs for the grid, which can have multiple LEM coalitions. The model for coalition costs, that is the cost faced by individual LEM coalitions in a grid partition, is presented in Section IV . Based on these cost models, the partitioning preferences of the DSO and the prosumers are formulated in Section V . Following this, the optimal stable partitioning problem is formulated in Section VI to balance these preferences. Here, we provide a solution for under perfect forecasts of prosumption (Theorem VI.1 ), and an algorithm to calculate the solution under imperfect forecasts (Algorithm 1 ). Finally, Section VII validates the theory via simulations in benchmark and real world distribution grids, with conclusions discussed in Section VIII . II Preliminaries II-A Distribution grid partitions We consider a radial distribution grid graph \mathcal{G}=(\mathcal{N}_{\mathcal{G}},\mathcal{E}_{\mathcal{G}}) with nodes (buses) \mathcal{N}_{\mathcal{G}} and edges (lines) \mathcal{E}_{\mathcal{G}} . Denote the Point-of-Common-Coupling of the distribution grid as PCC(\mathcal{G}) . Denote the set of leaf nodes as \mathcal{M}_{\mathcal{G}}\subset\mathcal{N}_{\mathcal{G}} , with each prosumer assumed to be located at a unique leaf node. Let \mathcal{S}(\mathcal{G}) denote the set of connected subgraphs of \mathcal{G} . In this work, we only consider LEMs that are connected subgraphs of \mathcal{G} , hence, any LEM F is an element of the set \mathcal{S}(\mathcal{G}) . For any LEM coalition F , denote its associated subgraph as (\mathcal{N}_{F},\mathcal{E}_{F}) , where \mathcal{N}_{F},\mathcal{E}_{F} denote the nodes and internal edges. This LEM coalition hence contains the prosumer set given by \mathcal{M}_{F}=\mathcal{N}_{F}\cap\mathcal{M}_{\mathcal{G}} . The boundary nodes for LEM F , representing the nodes that have lines connecting to the rest of the distribution grid, are denoted by \mathcal{N}_{F}^{b} , and the corresponding boundary edges by \mathcal{E}_{F}^{b} . The internal nodes \mathcal{N}_{F}\setminus\mathcal{N}_{F}^{b} are denoted as \mathcal{N}_{F}^{int} . A distribution grid partition is then defined below as Definition II.1 . (Grid partition) Let P=\{\{F_{l}\}_{l=1}^{L}\} denote a collection of connected subgraphs of the distribution grid \mathcal{G} . P is a distribution grid partition if and only if 1. \mathcal{E}_{F_{i}}\cap\mathcal{E}_{F_{j}}=\emptyset\ \forall i,j\in\{1,\cdots L\},i\neq j 2. \cup_{i=1}^{L}\mathcal{M}_{F_{i}}=\mathcal{M}_{\mathcal{G}} with the set of all possible partitions of