现代电子技术2025，Vol.48Issue(24)：25-30,6.DOI:10.16652/j.issn.1004-373x.2025.24.004
基于多特征的IHO-KELM模拟电路故障诊断
Analog circuit fault diagnosis based on multi-feature IHO-KELM
摘要
Abstract
In order to enhance the accuracy of fault diagnosis in analog circuits,a method of fault diagnosis based on multi-feature fusion and Kernel Extreme Learning Machine(KELM)optimized by improved hippopotamus optimization(IHO)is proposed.A comprehensive feature set is constructed by fusing statistical features and weighted Mahalanobis distance features.On the basis of IHO,the Sobol sequence is introduced to initialize the population,and the dynamic Lévy step size,sine-cosine algorithm oscillation and Cauchy distribution randomness strategies are added to realize the improvements to HO.The IHO algorithm is employed to optimize the regularization parameter(C)and kernel function parameter(g)of KELM,thereby establishing the IHO-KELM fault diagnosis model.The feasibility and efficiency of this method are validated by means of simulation experiments on a BUCK circuit.The simulation experimental results show that the proposed method can significantly improve the fault diagnosis accuracy and efficiency of analog circuits compared with other methods.关键词
模拟电路/故障诊断/河马优化算法/核极限学习机/加权马氏距离/多特征融合Key words
analog circuit/fault diagnosis/hippopotamus optimization algorithm/kernel extreme learning machine/weighted mahalanobis distance/multi-feature fusion分类
信息技术与安全科学引用本文复制引用
万军,王秋勇,高书苑..基于多特征的IHO-KELM模拟电路故障诊断[J].现代电子技术,2025,48(24):25-30,6.基金项目
国家自然科学基金青年基金项目(52105541) （52105541）
常州市领军型创新人才引进培育项目C类(CQ20210082) （CQ20210082）
