Addressing Model Inaccuracies in Transmission Network Reconfiguration via Diverse Alternatives

TL;DR

This work addresses grid congestion from high renewable penetration by proposing a human-in-the-loop modeling-to-generate alternatives (HITL-MGA) framework for topology reconfiguration that generates diverse near-optimal options and iteratively incorporates operator feedback to mitigate unmodeled constraints. It advances the MGA approach by (i) formulating an augmented DC-OPF/MILP-based topology reconfiguration model with a diversity term and a buffer, (ii) introducing baseline HITL-MGA and dynamic weighting variants (HITL-MGA-M and HITL-MGA-MDy) to integrate feedback, and (iii) defining six evaluation functions to quantify model inaccuracies and guide refinement. Case studies on IEEE 57-bus and 118-bus systems demonstrate that HITL-MGA can improve upon standard MGA in several objectives, particularly when substation switching expands the action space, though gains depend on alignment between evaluation functions and the MGA objective and on exploration strength. The results suggest a practical pathway to more robust, operator-aligned topology decisions under model limitations, with future work focused on enhancing exploration, feedback mechanisms, and integration of dispatch and sequencing considerations.

Abstract

The ongoing energy transition places significant pressure on the transmission network due to increasing shares of renewables and electrification. To mitigate grid congestion, transmission system operators need decision support tools to suggest remedial actions, such as transmission network reconfigurations or redispatch. However, these tools are prone to model inaccuracies and may not provide relevant suggestions with regard to important unmodeled constraints or operator preferences. We propose a human-in-the-loop modeling-to-generate alternatives (HITL-MGA) approach to address these shortcomings by generating diverse topology reconfiguration alternatives. Case studies on the IEEE 57-bus and IEEE 118-bus systems show the method can leverage expert feedback and improve the quality of the suggested topology reconfigurations.

Figures (7)

• Figure 1: The proposed approach, generating initial topological alternatives with MGA, utilizing feedback from the evaluation step to guide the generation of further alternatives with HITL-MGA.
• Figure 2: Modeling of substation switching, visualized for two substations with two busbars each, adapted from heidarifarNetworkTopologyOptimization2016.
• Figure 3: A two-dimensional representation of the proposed HITL-MGA-M & HITL-MGA-MDy approach applied to binary decision variables. The dots represent topological alternatives in the near-optimal solution space. The black and blue arrows represent MGA objectives, and the dots in the respective colors represent the alternatives found by them.
• Figure 4: Performance of the MGA-generated alternatives with respect to the primary objective $f$ (generation cost) and to the evaluation function value of cumulative quadratic load ($U_5$) for the IEEE-118 bus system with line switching.
• Figure 5: Performance comparison of the different HITL-MGA approaches (baseline, M, MDy) with respect to different model inaccuracies, evaluated by evaluation functions for the IEEE 118-bus system with line switching. For the specific switching actions, $f^* = U^{\varepsilon *}$. The aim is to maximize the number of specific switching actions, while the other evaluation functions should be minimized.