TE Process Fault Diagnosis Based on Improved Crow Search Algorithm Optimization ELM
ZHAO Wenhu
CAI Shenghong
WANG Wen
Abstract：As a new single hidden layer feedforward neural network,extreme learning machine(ELM)has obvious advantag-es in solving classification and data regression problems.However,the selection of its own parameters will also affect the accuracy.Therefore,an improved crow search algorithm(ICSA)is proposed to optimize the Tennessee Eastman(TE)process fault diagnosis model of ELM.In order to improve the optimization ability of crow search algorithm,chaos theory and Levy flight strategy are used to improve the crow search algorithm,and then the improved crow algorithm is used to optimize the weight and threshold of the extreme learning machine.Finally,it is applied to the fault classification of TE process.The results show that,compared with other algo-rithms,the improved crow algorithm has faster iteration and better performance.ELM can also accurately identify faults,improve the classification accuracy and achieve better results.
Keywords：extreme learning machineTennessee Eastman processcrow search algorithmfault diagnosisclassification accuracy
Publication Date：2025-08-20
Online Publishing Date：2025-12-12(First online date of this platform, not the publication date of the document)
Pages：6( 2122-2126,2139 )