摘要
为了准确地评价采煤机的健康状况,提出一种改进麻雀搜索算法(ISSA)优化模糊神经网络(FNN)的评估模型。针对麻雀搜索算法(SSA)的不足,引入精英混沌反向学习策略生成初始麻雀种群,提高算法收敛性能,采用柯西高斯变异策略提升种群多样性保持能力和局部空间逃逸能力。实验表明,所提方法提高了FNN的收敛精度和泛化能力,ISSA-FNN在采煤机健康评估中优于SSA-FNN与PSO-FNN模型。
In order to accurately evaluate the health status of shearer,an evaluation model to optimize fuzzy neural network(FNN)with improved sparrow search algorithm(ISSA)was proposed.Aiming at the shortcomings of sparrow search algorithm(SSA),an elite chaotic reverse learning strategy was introduced to generate the initial sparrow population and improve the convergence performance of the algorithm,and Cauchy Gaussian mutation strategy was used to improve population diversity maintenance and local space escape ability.Experiments show that the proposed method improves the convergence accuracy and generalization ability of FNN,and ISSA-FNN outperforms SSA-FNN and PSO-FNN models in shearer health assessment.
作者
李晓真
张海波
王光远
Li Xiaozhen;Zhang Haibo;Wang Guangyuan(Licun Coal Mine,Lu’an Chemical Group Co.,Ltd.,Changzhi 046000,China;School of Electrical and Control Engineering,Liaoing Technical University,Huludao 125105,China)
出处
《煤矿机械》
2024年第3期168-171,共4页
Coal Mine Machinery