Gearbox Fault Diagnosis Method Based on psoobp-cs Algorithm

Abstract

The invention provides a gearbox fault diagnosis method based on the PSOOBP-CS algorithm, extracts the fault characteristic data of the gearbox based on the ICEEMDSN and MPE algorithms, first judges the existence of the intermittent component, and adopts the corresponding decomposition algorithm accordingly, which improves the computing efficiency and reduces In the traditional feature extraction method, only the ICEEMDAN algorithm is used to remove the white noise residue added in the signal decomposition process. The present invention collects test data through the gear box working condition data acquisition device, and transmits to the fault diagnosis unit in the computer, utilizes the PSOOBP-CS algorithm to optimize the Elman-TCL neural network stored in the fault diagnosis unit, and combines the search ability of PSO with the CS The superposition of the global search ability of the PSO algorithm optimizes the particle velocity and position in the PSO algorithm, and optimizes the global optimal position of the particle in the PSO algorithm, so that it can get rid of the shortcomings of premature convergence and easy to fall into local optimum, and the fault diagnosis effect is good.

Description

technical field [0001] The invention belongs to the technical field of fault diagnosis, in particular to a gearbox fault diagnosis method based on PSOOBP-CS algorithm. Background technique [0002] As the main component of rotating machinery, gearboxes are widely used in wind power, heavy industry, automobiles, people's livelihood and other fields. Once the gearbox fails, it will bring unpredictable consequences, affect the operation of the equipment, bring large production costs and time costs to the enterprise, cause huge economic losses, and even endanger people's lives. Therefore, real-time diagnosis of the health status of the gearbox is of great significance for maintaining the health of equipment and ensuring the safety of people's lives. [0003] With the continuous development of modern science and technology, the research on gearbox fault diagnosis methods has become more and more abundant. The analysis methods include time-domain analysis, frequency-domain analys...

AI Technical Summary

Problems solved by technology

Among them, for the method of feature extraction, at present, the parameter indicators in the time domain and frequency domain of the signal are extracted as features to identify faults, and the means are relatively simple. Therefore, it is necessary to conduct in-depth research and continuous improvement on the method of feature extraction

At the same time, the neural network, vector machine and other models used in traditional pattern recognition methods have limitations, which have a great impact on the diagnostic accuracy of the results.

Images