Exact and Evolutionary Algorithms for Sequential Multi-Objective Transmission Topology Planning
Job Groeneveld, Miguel Muñoz, Jan Viebahn, Alessandro Zocca, Job Groeneveld, Miguel Muñoz, Jan Viebahn, Alessandro Zocca · 2026
Read Exact and Evolutionary Algorithms for Sequential Multi-Objective Transmission Topology Planning by Job Groeneveld, Miguel Muñoz, Jan Viebahn, Alessandro Zocca, Job Groeneveld, Miguel Muñoz, Jan Viebahn, Alessandro Zocca (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-2605-03753.txt.
Introduction
Job Groeneveld, Miguel Muñoz, Jan Viebahn, and Alessandro Zocca study day-ahead control of high-voltage transmission grid topology under N-1 security constraints, meaning the grid must survive any single failure. Operators can switch busbar couplers at substations to relieve line overloads, but switching has costs. The task is to choose a sequence of topologies over twenty-four hours.
Four objectives from transmission operator practice are balanced: worst-case line loading, topological depth, number of topology changes, and time spent away from the reference topology. An exact block algorithm groups consecutive hours sharing one topology and enumerates the complete Pareto front; an NSGA-III evolutionary heuristic is tested against it. On real Dutch TenneT data, the exact front for a highly congested day took under three minutes.
Catalog record
| CREATOR | Job Groeneveld, Miguel Muñoz, Jan Viebahn, Alessandro Zocca, Job Groeneveld, Miguel Muñoz, Jan Viebahn, Alessandro Zocca |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 79 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.03753
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Exact and Evolutionary Algorithms for Sequential Multi-Objective Transmission Topology Planning Job Groeneveld, Miguel Muñoz, Jan Viebahn, Alessandro Zocca, Job Groeneveld, Miguel Muñoz, Jan Viebahn, Alessandro Zocca Abstract We study day-ahead transmission topology control for high-voltage grid operation under N − 1 N-1 security constraints. The operational task is to select, over a 24-hour horizon, a sequence of substation topologies obtained via busbar-coupler switching to relieve line overloads while limiting switching effort and topological complexity. We formulate this task as a sequential multi-objective optimization problem with four objectives used in TSO decision making: worst-case N − 1 N-1 line loading, maximum topological depth, number of topology changes, and time spent outside the reference topology. We propose an exact block algorithm that exploits the temporal structure of topology plans: consecutive hours with the same topology are represented as blocks, enabling enumeration of the complete Pareto front over the admissible set of topologies under fixed operational bounds on depth and switching. We also develop a tailored NSGA-III-based evolutionary heuristic and evaluate it against the exact front. Using real operational data from the Dutch high-voltage transmission grid operated by TenneT, the block algorithm computes the exact front for a highly congested day in under three minutes after topology-level load-flow preprocessing. The exact front reveals low-switching plans with no DC N − 1 N-1 thermal overloads that the tested evolutionary search fails to find. The proposed method, therefore, provides both a practical day-ahead decision-support tool for transmission operators and a benchmark for heuristic and learning-based topology-control methods. Keywords: Transmission topology control ; N-1 security ; Congestion management ; Optimal transmission switching ; Busbar switching ; Multi-objective optimization ; TSO decision support 1 Introduction Transmission system operators (TSO) routinely assess whether the day-ahead high-voltage grid schedule remains secure under N-1 contingency criteria [ 17 , 14 ] . When forecasted injections lead to overloads, operators can deploy remedial actions to modify power flows before more expensive measures, such as redispatch, are required. One operationally important class of remedial actions is transmission topology control: changing the network configuration so that power is redistributed over the existing grid while relevant line-loading limits remain respected [ 15 , 26 ] . This paper focuses on substation-level topology control through busbar-coupler switching. Opening or closing busbar couplers changes the electrical grouping of busbars inside a substation and can be modeled as node splitting in the transmission network [ 18 ] . This differs from line switching, which is the dominant abstraction in much of the optimal transmission switching literature [ 11 , 17 , 14 ] . From a control-room perspective, busbar-coupler switching creates a multi-objective decision problem: a topology may reduce the worst N-1 line loading, but at the cost of additional switching actions, greater topological depth, or more time spent operating away from the reference topology. The day-ahead version of this problem is inherently sequential. A topology chosen for one hour affects the number of subsequent switching actions, and the operator is not seeking a single mathematical optimum but a set of feasible trade-offs among security margin, switching effort, topological complexity, and persistence of non-reference configurations. Existing topology-control methods are often heuristic, single-objective, or applied independently at each time step [ 12 , 32 , 24 , 9 ] . As a result, they do not provide a complete view of the Pareto-optimal operational choices over the full planning horizon, nor do they provide an exact reference against which heuristic or learning-based methods can be evaluated. Increasing demand and renewable-driven flow patterns push transmission grids to operate closer to their limits, which raises the value for TSOs of corrective actions that do not require redispatch. Topology control is attractive in this setting because it can relieve overloads without immediately requiring grid expansion or generation redispatch, provided that the resulting topology remains acceptable from an operational security perspective [ 16 , 9 ] . The relevant question for a TSO is therefore not only whether a topology improves a loading metric, but whether a full day-ahead topology plan provides an acceptable trade-off between N-1 security, switching burden, and operational simplicity. The key methodological gap is that sequential, multi-objective topology control has not been solved exactly at the level needed for day-ahead operational decision support. Existing exact formulations for optimal transmission switching primarily target single operating points or economic objectives [ 11 , 27 ] . For a 24-hour horizon, direct enumeration over hourly topology choices is combinatorially prohibitive. This has encouraged the use of heuristics, dynamic programming approximations, evolutionary algorithms, and reinforcement-learning methods [ 12 , 32 , 24 , 9 ] . Such approaches can produce useful candidate actions, but without an exact Pareto front it remains difficult to know whether they miss operationally important trade-offs, especially low-switching strategies that remove DC N-1 thermal overloads. We address this gap by exploiting the block structure of day-ahead topology plans. Instead of treating each hour independently, we group consecutive hours with the same topology into blocks and optimize over the resulting block configurations. This preserves the operational objectives exactly while reducing the effective search space for fixed bounds on switching and topological depth. We then use the exact Pareto front as both a decision-support object and a benchmark for a tailored multi-objective evolutionary algorithm (MOEA). The main contributions of this paper are as follows: 1. We formalize day-ahead substation-level transmission topology control as a sequential multi-objective optimization problem under N-1 security assessment, using four operational objectives reflecting real TSO decision criteria. 2. We develop an exact algorithm, named block algorithm , that exploits the temporal block structure of strategies to enumerate the complete Pareto front over the precomputed admissible topology set for fixed maximum depth and switch count. 3. We design a tailored MOEA with structure-guided initialization and problem-specific variation operators, and evaluate it against the exact Pareto front using both \mathrm{IGD}^{+} and a novel multi-front coverage metric. 4. We demonstrate, using real operational data from the Dutch high-voltage grid operated by TenneT, that exact enumeration is feasible and recovers operationally relevant plans missed by the tested evolutionary search, providing transmission operators with a practical and provably complete decision-support tool. The rest of the paper is organized as follows. Section 2 reviews related work on optimal topology switching and optimization methods. Section 3 formulates the problem and solution algorithms. Section 4 describes the case study and evaluation framework. Section 5 reports the exact and evolutionary results, and Section 6 discusses operational implications, limitations, and future work. 2 Related Literature Transmission topology control is commonly studied under the term optimal topology switching (OTS), which covers dynamic grid reconfiguration through changes in network topology. OTS formulations differ mainly in the power-flow model, the inclusion of flexibility options such as dynamic thermal rating, storage, or renewable curtailment, and the treatment of uncertainty [ 15 , 26 ] . Most studies model line-switching decisions, whereas this paper addresses substation reconfiguration through busbar-coupler switching, or node splitting. Explicit node-breaker formulations have received less attention; a notable exception is [ 18 ] . The present work uses deterministic input data, a DC load-flow model [ 30 ] , and busbar-coupler switching as the only control action. Classical OTS research has primarily used mathematical programming formulations. The