TE Process Fault Diagnosis Based on Improved Butterfly Algorithm Optimization SVM
Aiming at the problems of instability and low classification accuracy of support vector machine in Tennessee East-man process fault diagnosis,easy to fall into local optimization and slow convergence of butterfly optimization algorithm in optimiz-ing support vector machine,a model of using improved butterfly algorithm to optimize support vector machine is proposed.Firstly,reverse learning strategy,adaptive adjustment strategy and chaotic local search strategy are used to optimize the butterfly algorithm to improve the optimization performance and development ability of the butterfly algorithm.Then the improved butterfly algorithm is used to optimize the parameters of support vector machine.Finally,the support vector machine model optimized by the improved butterfly algorithm is applied to the fault diagnosis of Tennessee Eastman process.Through function test and comparative experi-ment,the results show that the improved butterfly algorithm has better optimization effect,and the support vector machine opti-mized by the improved butterfly algorithm has better stability and higher classification accuracy.
support vector machinebutterfly optimization algorithmfault diagnosisfunction testclassification accuracy
国家科技期刊平台
NSTL
万方数据
