Gear performance degradation evaluation method based on AR model and dictionary learning

Abstract

The invention discloses a gear performance degradation evaluation method based on an AR model and dictionary learning. The method comprises the following steps: firstly, extracting features of a fault-free sample and a to-be-detected sample by using an autoregression model (AR), constructing an over-complete dictionary model by using AR model coefficients of the fault-free sample, and then inputting the AR model coefficients of the to-be-detected sample into the dictionary model as feature vectors to obtain reconstructed AR model coefficients; and finally, respectively constructing an autoregressive model by the original AR model coefficient and the reconstructed AR model coefficient, and completing time sequence modeling of the to-be-measured signals respectively; taking the root-mean-square error of the residual sequence obtained by the two autoregression models as a performance degradation degree index, andsetting an adaptive early warning threshold. Experimental analysis shows that the evaluation index obtained by the performance degradation method provided by the invention can monitor the performance degradation trend of the gear in real time and can discover the early failure of the gear in time.

Description

technical field [0001] The invention relates to a gear performance degradation evaluation method based on AR model and dictionary learning, and belongs to the technical field of mechanical product quality reliability evaluation and fault diagnosis. Background technique [0002] As one of the key components of rotating machinery, the gear's performance directly determines the performance of the equipment. Once it breaks down, it will directly affect the normal and safe operation of mechanical equipment and even cause major safety accidents. Therefore, how to realize in-service condition monitoring and performance degradation assessment of gears is of great significance. Reducing downtime costs and achieving near-zero downtime is the ultimate goal of pre-diagnostics. However, without an accurate prediction of the remaining service life before a failure actually occurs, it is impossible to realize the full benefits of pre-diagnostics. Inaccurate forecast information may lead...

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a gear performance degradation evaluation method based on AR model and dictionary learning, which can solve the problems of complex model and easy premature saturation in the probabilistic similarity measurement evaluation method, so as to carry out online state monitoring of gears and realize Maintenance and prevention of major accidents

Images