Application of variable strategy adaptive differential evolution with preselection mechanism to fault reconstruction of multi-load independent distribution system

Abstract

As power system capacity and network size increase, fault reconstruction techniques for multi-load independent distribution systems assume a pivotal role in ensuring the stable operation of distribution networks. The paper proposes a novel fault reconstruction method, addressing the challenges posed by the inherent characteristics of multi-load independent distribution system and the limitations of existing algorithms. The proposed approach involves three key components. First, an enhanced coding method is introduced to ensure that the branch correlation matrix of a multi-load independent distribution system effectively captures the network structure characteristics of the system. Then, a pre-selection mechanism is incorporated into the algorithm to ensure that the algorithm obtains a more optimal initial population during the initialization phase, thus optimizing the convergence speed of the algorithm. Finally, a variable strategy adaptive differential evolutionary algorithm is designed to improve the accuracy of the fault reconstruction scheme. The algorithm has been demonstrated to achieve a 24.1 % increase in combined optimization capability and a 28.34 % increase in optimization speed. The efficacy of the proposed method is demonstrated by its ability to identify the optimal fault reconfiguration scheme for multi-load independent distribution system, achieving a higher level of accuracy and efficiency.