Power System Fault Diagnosis Model Via Particle Swarm Differential Evolution-based Extreme Learning Machine
Rapid diagnosis for faults occurring in power systems is of extraordinary significance for timely restoration of power supply and reduction of fault impact.In order to effectively deal with the uncer-tainty in the operation of protective relays and circuit breakers during power system faults,this paper pro-poses an extreme learning machine-based fault diagnosis model based on particle swarm differential evolu-tion algorithm with multiple random variants(MRPSODE).The MRPSODE is used to determine the opti-mal number of nodes in the hidden layer of extreme learning machine to achieve efficient fault diagnosis.A cross-validation method is used to reduce the influence of noise on the original samples to improve the di-agnosis performance.Simulation results of actual fault cases show that the proposed method can success-fully diagnose complex faults and is competitive compared with other methods.
国家科技期刊平台
NSTL
万方数据
