ITD Sample Entropy and Probabilistic Neural Network Bearing Fault Diagnosis Model

A new fault diagnosis method based on ITD sample entropy and probabilistic neural network is proposed. Firstly, the vibration signal of the equipment is decomposed using intrinsic time-scale decomposition (ITD) to obtain several proper rotation components (PRC) and a trend item that spreads from high frequency to low frequency. Secondly, the components that have a high degree of correlation with the original signal are selected as the research object, and their sample entropy is used as the feature vector. Finally, the feature vector is used to train the probabilistic neural network, and fault identification is performed using the probabilistic neural network. The proposed method is verified using experimental data to accurately identify the type and extent of equipment failure.