A Cutting Pattern Recognition Method for Shearers Based on ICEEMDAN and Improved Grey Wolf Optimizer Algorithm-Optimized SVM

When the shearer is cutting, sound signal generated by cutting drum crushing coal and rock contains a wealth of status information. In order to effectively process accurately identify mode, this paper proposed recognition method based on an improved complete ensemble empirical mode decomposition with adaptive noise (ICCEMDAN) grey wolf optimizer (IGWO) algorithm-optimized support vector machine (SVM). First, approach applied ICEEMDAN obtained several intrinsic function (IMF) components. It used correlation coefficient select characteristic component. Meanwhile, calculated composite multi-scale permutation entropy (CMPE) components as eigenvalue. Then, introduced differential evolution algorithm nonlinear convergence factor improve GWO algorithm. realize selection SVM parameters established model. According proportioning plan, made simulation walls for experiments collected signals pattern recognition. The experimental results show that in can shearer, average accuracy model reached 97.67%.