Analog Circuit Fault Diagnosis Method Based on IWOA-ELM
Aiming at the hard problem of consciousness caused by nonlinear and high latitude output signals in analog circuit fault diagnosis,an analog circuit fault diagnosis method based on improved whale algorithm(IWOA)optimized extreme learning machine(ELM)was proposed.Firstly,principal component analysis(PCA)was used to reduce the dimensionality of initial fault circuit features;Secondly,based on the whale algorithm,a Tent map was introduced to initialize the population,and nonlinear time-varying factors,adaptive weights and random differential mutation strategies were added.Then the improved whale algorithm was reused to optimize ELM;Finally,the dimensionality reduced fault feature vectors were input into ELM to obtain the fault diag-nosis results.The simulation test examples of Sallen-Key bandpass filter circuit and CSTV filter circuit show that IWOA optimized ELM fault diagnosis method has better fault diagnosis performance with a fault diagnosis accuracy of up to 99.41％.
analog circuitsfault diagnosisfeature extractionprincipal component analysislimit learning machinewhale algorithm
NETL
NSTL
万方数据
