Wei Du,∗ Guozhu Yang,∗ Maojie Tian,∗ Wei Hu,∗ and Chuntian Ma∗
Multiverse optimisation algorithm (MOA), active distribution network (ADN), distributed power generation, fault location, fault tolerance performance
To address the issues of slow convergence, susceptibility to local optima, and poor fault tolerance in existing optimisation algorithms for active distribution network fault location, an improved multiverse optimisation (IMVO) algorithm is proposed. A fault-tolerant network model is established, and the multiverse population is encoded discretely based on fault location features. The algorithm incorporates an adaptive elite strategy to enhance population quality and a nonlinear update mechanism for wormhole existence probability (WEP) and travel distance rate (TDR) to balance early-stage exploration and late-stage precision. Adaptive mutation operations further strengthen local search and global optimisation. Simulations show that IMVO achieves optimal solutions within ten iterations in single-point and multi-point fault cases, with iteration times around 1.8 s. The improved WEP and TDR ensure iteration stability, while adaptive strategies enhance population quality and exploration. IMVO outperforms other algorithms in accuracy and convergence speed for fault location in distribution networks.
Important Links:
Go Back
