Sand Cat Swarm Optimization Algorithm Based on Weibull Flight and Warning Mechanism and Its Application
Aiming at the problems of slow convergence and low optimization accuracy of sand cat swarm optimization algorithm,a multi-strategy improved sand cat swarm optimization algorithm is proposed.Firstly,Latin hypercube sam-pling is used for initialization to improve population diversity.Secondly,Weibull flight is proposed in the search stage to enhance the search ability of the algorithm.Finally,an alert mechanism is proposed to further improve the optimization ability and convergence speed of the algorithm.The challenging CEC2017 function is used for functional testing,and comprehensive evaluation is performed based on qualitative analysis of benchmark function,optimization precision analy-sis,effectiveness analysis of improved strategy,convergence curve analysis,Wilcoxon rank sum test and Friedman test.The experimental results show that compared with other three sand cat swarm algorithms and six meta-heuristic algo-rithms,the proposed algorithm has significant advantages in the optimization accuracy and convergence of complex func-tions.The algorithm is applied to transformer fault diagnosis cases,and the effectiveness of ESCSO algorithm is further verified.
万方数据
维普
