A Novel Radiality Maintenance Algorithm for the Metaheuristic Based Co-Optimization of Network Reconfiguration with Battery Storage

Abstract

Network reconfiguration (NR) is a well-accepted technique to decrease power losses and enhance the voltage profile of the distribution network. The practical challenges in solving the NP-hard NR problem using metaheuristics are to randomly generate and, in each iteration, check and repair non-radial distribution network configurations without compromising on solution space in the least amount of time. Inefficient handling of non-radial configurations results in large computational time and high standard deviation. This paper mitigates the aforementioned challenges by proposing a novel radiality maintenance algorithm (RMA) that involves the novel concept of junction nodes and a selection set to produce only radial configurations. The proposed approach can potentially improve the standard deviation and computational efficiency of metaheuristics in solving the NR problem. The proposed RMA is generic, model-independent, and scalable, as it can be seamlessly integrated into any metaheuristic approach to solve the NR problem involving feeders of different sizes. The proposed RMA, combined with the accelerated particle swarm optimization, is implemented to solve: 1) the standard snapshot NR problem; and 2) the multiperiod co-optimization problem that simultaneously computes optimal network configuration and control setpoints of the photovoltaic system and battery energy storage system. Simulation results suggest a 27.3% standard deviation reduction in achieving the best results reported in the literature on NR-based power loss reduction within a comparable timeframe of 1.6 seconds. The effectiveness and reliability of the proposed algorithm are demonstrated on IEEE 33-bus test system.