Machine Learning Methods as Fast Heuristics for Network Topology Optimization

IEEE Mediterranean Electrotechnical Conference (MELECON), pp. 127–132

Abstract

Europe’s ongoing turn away from conventional electricity generation towards renewable generation results in higher line utilization leading to more grid congestion situations. Congestion management aims to resolve grid congestion using various methods with one of them being transmission reconfiguration such as line switching and bus splitting. This work compares the performance of various machine learning approaches that predict advantageous bus splits. The proposed method is applied to a modified IEEE 118-bus network using one year of hourly grid usage scenarios. The results demonstrate that the proposed model is able to suggest suitable bus splits and thus lower resulting congestion management cost as well as provide a good starting solution for subsequent network topology optimization.

Authors 3

1. Felix Preuschoff Aachen Institute for High Voltage Equipment and Grids

   RWTH Aachen University

   Affiliation as printed

   Institute for High Voltage Equipment and Grids, Digitalization and Power Economics (IAEW) RWTH Aachen University,Aachen,Germany

2. Luna Zirkel Aachen Institute for High Voltage Equipment and Grids

   RWTH Aachen University

   Affiliation as printed

   Institute for High Voltage Equipment and Grids, Digitalization and Power Economics (IAEW) RWTH Aachen University,Aachen,Germany

3. Albert Moser Aachen Institute for High Voltage Equipment and Grids

   RWTH Aachen University

   Affiliation as printed

   Institute for High Voltage Equipment and Grids, Digitalization and Power Economics (IAEW) RWTH Aachen University,Aachen,Germany