Rolling bearing fault diagnosis method based on convolutional neural network

Abstract

The invention discloses a rolling bearing fault diagnosis method based on a convolutional neural network (CNN). By aiming at problems of rolling bearing characteristic components such as easy submergence and difficulty in extraction and combining with rolling bearing signal own and large monitoring data quantity and other characteristics, the CNN is introduced in the rolling bearing fault diagnosis. By short time Fourier Transform, a motor vibration signal is converted into a time frequency spectrogram to be adapted to a CNN network training sample format, and then mass sample data having labels used to express different faults is established, and therefore sample diversity is guaranteed, and network overfitting is prevented. The CNN network having a proper layer number is established, and parameters are initialized, and then the preprocessed samples are input in the CNN for forward propagation. By combining with predetermined label calculation errors, a network weight is adjusted by using an error reverse propagation algorithm, and then after a plurality of times of iterations, the network used for the interconnection between the signal and equipment is established, and therefore the rolling bearing fault accurate diagnosis is realized.

Description

technical field [0001] The invention belongs to the field of fault diagnosis of mechanical equipment, and in particular relates to a rolling bearing fault diagnosis method based on a convolutional neural network. Background technique [0002] Mechanical equipment is developing towards high speed, high precision, and high efficiency. In order to ensure the healthy operation of the equipment, a large amount of data is collected to reflect the health status of the machinery, and the field of mechanical health monitoring has entered the "big data" era. Mechanical big data has the characteristics of large capacity, diversity and high speed. Researching and using advanced theories and methods, how to mine information from mechanical equipment big data, and realize efficient and accurate identification of health conditions have become a major challenge in the field of mechanical equipment health monitoring. of new problems. [0003] Rolling bearings are widely used in various rota...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problems that the characteristic components are easily submerged and difficult to extract when diagnosing rolling bearing faults, the purpose of the present invention is to provide a method for diagnosing rolling bearing faults based on convolutional neural networks. The samples processed by the leaf transform (STFT) are used for the training of the convolutional neural network, and combined with the label to supervise the fine-tuning of the whole network, so as to achieve accurate rolling bearing fault diagnosis

Images