Fault Diagnosis Method of Metallurgical Hydraulic System Based on Improved Neural Network
A fault diagnosis method of metallurgical hydraulic system based on improved neural network is presented.The method combines wavelet packet transform to extract features,uses genetic algorithm to optimize network parameters,and introduces attention mechanism to improve diagnostic accuracy.In the multi-fault mode test of 1 000 t hydraulic press,the diagnostic accuracy of the proposed method is up to 98.75%,and the average diagnostic time is 0.85 s,which is better than the traditional Back Propagation(BP)neural network and Support Vector Machine(SVM)method.
NETL
NSTL
万方数据
