Open-source implementation of distribution network reconfiguration methods: Analysis and comparison

TL;DR

This paper reviews and compares various distribution network reconfiguration algorithms, from heuristic to mathematical methods, and provides an open-source repository to support research and development in the field.

Contribution

It systematically analyzes seven reconfiguration methodologies and offers an open-source implementation to foster reproducibility and further innovation.

Findings

Meta-heuristic methods outperform classical heuristics in complex scenarios

Mathematical approaches like MILP show high accuracy but higher computational cost

Open-source repository enhances accessibility and reproducibility

Abstract

This paper presents a critical and practical approach to the evolution of distribution network reconfiguration algorithms, tracing their development from foundational heuristic methods introduced in 1975 to contemporary state-of-the-art techniques. The article systematically reviews seven different methodologies, including classical heuristic algorithms (Merlin, Baran, and others), advanced meta-heuristic methodologies (particle swarm optimization (PSO) and genetic algorithms), and purely mathematical approaches (MILP-based), analyzing their theoretical foundations, implementation strategies, computational complexity, and performance metrics based on extensive literature review and our own empirical testing. Each methodology is assessed through standardized test systems, considering multiple objectives such as power loss minimization and voltage profile improvement. The comparative analysis reveals the strengths and limitations of each approach under various network conditions and operational constraints. Furthermore, this work provides significant value to the research community by offering an open-source repository containing documented implementations of all reviewed algorithms. This resource facilitates accessibility for newcomers to the field, promotes reproducible research, and accelerates the development of next-generation distribution network optimization solutions. The repository includes comprehensive documentation, test cases, and performance benchmarks.