Research on Vibration Signal Classification and Recognition System Based on Computer Deep Learning Technology

Reducing equipment downtime by detecting function decay signals. Then the fault feature recognition technique of vector extreme learning machine is proposed. The mixed singular value decomposition method based on wavelet transform can effectively extract the vibration signal features containing impact noise. Finally, the obtained eigenmatrix is normalized and sent to the extreme learning machine for recognition. A representative sample of training was tested and the results showed that the accuracy of each training method improved. By comparing with BP neural network and support vector machine, the advantages of this algorithm in recognition accuracy and training speed are verified. Based on vibration analysis of intelligent fault diagnosis, BP neural network is used to extract eigenvalues of each type of fault to obtain diagnosis results of each type of fault. The results show that this method can identify the fault type effectively with high accuracy. The results show that the algorithm can accurately identify different fault types and fault degrees of equipment, and improve the reliability of equipment operation.