Industrial equipment fault diagnosis system and method based on GA-BP-CBR

Abstract

The invention provides an industrial equipment fault diagnosis system and method based on GA-BP-CBR, and the system comprises a GA-BP neural network module, a case-based reasoning CBR network module,a result correction module, and an output module. The input module inputs a fault training data set to the neural network module, and the GABP neural network module trains a pre-classification network. Meanwhile, CBR case reasoning is performed on the case description to form a case library file; indexes are established for the cases by utilizing output results of the trained pre-classification network, an original case library are divided into a plurality of sub-case libraries, test data are input into the trained pre-classification network during diagnosis, similar case sets are searched forin the corresponding sub-case libraries according to output of the network, and finally, the result correction module evaluates and corrects the obtained case set by referring to the output of the neural network to obtain a final diagnosis result. According to the invention, the fault diagnosis retrieval time is reduced, and the production efficiency is improved.

Description

technical field [0001] The invention belongs to the combined field of artificial intelligence and production line equipment, and relates to the research on fault diagnosis mechanism of production line equipment based on optimized neural network and case reasoning. Background technique [0002] The reliability of industrial equipment and production system operation has an important impact on the profitability and competitiveness of production enterprises, which makes enterprises pay more and more attention to the importance of maintenance strategies for industrial production processes and production equipment. my country's industrial development is in full swing, and the Industry 4.0 Smart Manufacturing 2025 strategy has been proposed. This is a huge challenge and opportunity for my country's industrial development. We must seize this opportunity to develop our country's industrial undertakings, raise our country's manufacturing level to a higher level, and become a manufact...

AI Technical Summary

Problems solved by technology

The limitation of GA-BP is that the performance of the neural network fault diagnosis device mainly depends on the uncertainty of the training samples; when adding new training samples, the entire network needs to be retrained, which makes the self-learning of the network very inconvenient; In the network system, the way to express the diagnostic object information is single, because the usual neural network can only process numerical information, etc., and the main limitation of CBR is reflected in the establishment of cases and the selection of cases, and secondly, the similarity between cases is difficult to determine, etc.

Images