Equipment fault diagnosis method based on improved 1DCNN-BiLSTM

Abstract

The invention discloses an equipment fault diagnosis method based on an improved 1DCNN-BiLSTM, and the method comprises the following steps: S1, preprocessing an original vibration acceleration signal by a self-adaptive white noise complete empirical mode decomposition (CEEMDAN) technology, and taking the preprocessed signal as input of a model; S2, constructing a 1DCNN-BiLSTM dual-channel model, inputting the preprocessed signal into two channels of a bidirectional LSTM model and a one-dimensional CNN model, and fully extracting the time sequence correlation characteristics of the signal, the non-correlation characteristics of the local space and the weak periodicity rule; S3, improving a SENet module and acting on two different model channels aiming at the problem that the signal is mixed with strong noise; and S4, fusing the two-channel extraction characteristics in a full connection layer, and realizing accurate identification of equipment faults by means of a Softmax classifier. To solve the problems of time sequence and noise inclusion of fault data in the industrial field, filtering and denoising preprocessing is carried out on original signals, a 1DCNN-BiLSTM dual-channel feature extraction module is constructed, a modified SENet module is integrated to realize weighting of feature channels, and the fault diagnosis efficiency of mechanical equipment is effectively improved.

Description

technical field [0001] The invention relates to the technical field of industrial equipment fault diagnosis and deep learning model construction, in particular to an equipment fault diagnosis method based on improved 1DCNN-BiLSTM. Background technique [0002] With the rapid development of modern industry, the mechanical equipment in the "smart factory" is also developing vigorously in the direction of integration and complexity. Rolling bearings are one of the widely used parts in rotating machines. With the continuous operation of mechanical equipment, various failures of bearings are inevitable. According to statistics, among the faults of rotating machinery, the fault of bearing damage accounts for about 30%. The causes of failures are often complex and diverse. Condition monitoring and fault diagnosis of rolling bearings are important contents of fault diagnosis technology for mechanical equipment. Therefore, the fault diagnosis of mechanical equipment bearings is of ...

AI Technical Summary

Problems solved by technology

For the problem of signal mixed with strong noise, improve the SENet module and apply it to two different channels

Images