Immune evolutionary algorithm of wavelet neural network to predict the performance in the centrifugal compressor and Research

Centrifugal compressor performance prediction method mainly uses the traditional BP neural network problem is not high enough accuracy, convergence, easily falling into local optimal solution. In order to more accurately predict the performance of centrifugal compressors, find problems early implicit commit. Is the algorithm, wavelet theory, artificial neural networks, established immune algorithm of wavelet neural network model of a centrifugal compressor performance prediction. First, the initial antibody produced by the immune algorithm group, through iteration, obtain the corresponding coefficient for each antibody WNN, and then use back-propagation algorithm to train WNN approach any nonlinear function. Simulation results show that application of the prediction model, which enables the accurate prediction of centrifugal compressor performance and monitoring. The prediction model algorithm is simple, stable structure, computational convergence speed, generalization ability of the advantages of prediction accuracy of 99%. Than the traditional method of prediction accuracy of 13%. Has some theoretical research value and practical value.