Equipment fault diagnosis method based on deep learning

Abstract

The invention relates to an equipment fault diagnosis method based on deep learning. Aiming at the characteristics of equipment data, a parameter regularization and Dropout method is added to an original basic convolutional neural network model to improve the generalization ability of the model, and an adaptive batch normalization algorithm is introduced on the basis of adding a BN algorithm afteran activation function, so that the adaptive recognition ability of the model is further improved. The adopted deep learning algorithm directly and automatically learns and extracts features layer bylayer; the device state change contained in the data change can be analyzed without further signal processing or depending on experience knowledge of experts, the fault type of the monitored device can be objectively and accurately reflected, and a basis is provided for guiding subsequent device fault management and maintenance work.

Description

technical field [0001] The present application relates to a device fault diagnosis technology, in particular to a device fault diagnosis method based on deep learning. Background technique [0002] As the management of mechanical equipment becomes increasingly intelligent and complex, industrial systems also show a trend of large-scale and big data development. Fault diagnosis monitors and analyzes the operating status of system equipment, judges the type of fault and locks the fault location, gives a reasonable use plan based on the results of equipment fault diagnosis, and designs a repair plan that maximizes the efficiency of the equipment. Optimizing the production plan of the factory, conducting performance evaluation, etc., can accurately guide the operation of the equipment, which is an important means to ensure the reliable and safe use of the equipment, and it is also a powerful method to improve social production capacity and economic benefits. [0003] At present...

AI Technical Summary

Problems solved by technology

[0005] At present, there have been some studies on fault diagnosis based on deep learning, such as the Chinese patent "A Method for Fault Diagnosis of Equipment" (publication number: CN109670595A), but the deep learning method is poorly self-adaptive when the data contains noise and load changes, and the fault diagnosis The recognition rate will decrease

Images