Rotor monitoring method based on deep learning signal reconstruction

Abstract

The invention discloses a rotor monitoring method based on deep learning signal reconstruction. The rotor monitoring method comprises the following steps: 1, obtaining initial vibration signal data ofa rotor; 2, arranging the signal data into a matrix form and standardizing, and dividing to obtain a training set and a verification set; 3, building and obtaining a deep learning neural network model; 4, training a deep learning neural network model to obtain a trained reconstruction and classification model; 5, continuing training according to the error with the actual monitoring signal until the error between the vibration signal output by the model and the monitoring signal meets the preset requirement, and obtaining a trained reconstruction and classification model; and 6, realizing rotor monitoring by utilizing the trained reconstruction and classification model obtained in the step 5. According to the method, accurate rotor full-measuring-point signals can be obtained by means of monitoring signals of a few measuring points, the whole rotor and key parts can be monitored, and the accuracy of rotor fault recognition is improved.

Description

technical field [0001] The invention belongs to the technical field of mechanical operation monitoring, in particular to a rotor monitoring method based on deep learning signal reconstruction. Background technique [0002] The rotor is an important part in the industrial production process. Its structure and working environment are relatively complex, and it is prone to failure. Once a failure occurs and cannot be checked in time, it will cause great economic losses and even safety accidents. Therefore, the real-time monitoring and fault warning of the rotor is of great significance. [0003] Due to the lack of number of measuring points, difficulty in setting, and deviation in the position of the current real-time monitoring, the monitored data often cannot truly reflect the vibration state of the rotor, especially the vibration of the core part, resulting in a decline in the accuracy of fault diagnosis. In summary, the current rotor monitoring and fault diagnosis have the...

AI Technical Summary

Problems solved by technology

[0004] 1. The number of measuring points is limited, and the monitoring situation is often difficult to reflect the operation status of the entire rotor;

[0005] 2. It is difficult to arrange the measuring points in the core parts such as the vulnerable parts of the rotor, and there are certain blind spots in the monitoring;

[0006] 3. Fault diagnosis using monitoring data often leads to unsatisfactory classification and diagnosis results due to the above problems

Images