Rolling bearing reliability prediction method based on mathematical morphology and IFOA-SVR

Abstract

The invention discloses a rolling bearing reliability prediction method based on mathematical morphology and IFOA-SVR, and relates to the technical field of rolling bearing reliability prediction. The method aims at guaranteeing the prediction precision and prolonging a prediction step at the same time. The method comprises the steps: firstly extracting an envelope signal of a vibration signal, calculating a mathematical morphology fractal dimension of the envelope signal, and enabling the mathematical morphology fractal dimension to serve as the performance degeneration state characteristics of the rolling bearing; secondly carrying out the optimizing of the parameters C, g and epsilon in SVR at the same time through IFOA and building a prediction model, and, meanwhile, building a Weibull proportion fault rate model through employing MLE (Maximum Likelihood Estimation) and combining with IFOA, thereby obtaining a reliability model; finally enabling the performance degeneration state characteristics to serve as the input of an IFOA-SVR prediction model, obtaining a characteristic prediction result through employing a long-time iteration prediction method, enabling the result to be embedded into the reliability mode, and predicting the reliability of the operation state of the rolling bearing. An experiment indicates that the method prolongs the prediction step while guaranteeing the prediction precision.

Description

technical field [0001] The invention relates to a rolling bearing reliability prediction method based on mathematical morphology and IFOA-SVR, and relates to the technical field of rolling bearing reliability prediction. Background technique [0002] Rolling bearings are key components in rotating machinery, once a failure occurs, it will cause a lot of economic losses and even endanger people's lives [1-2] . Therefore, accurate prediction of the working state of rolling bearings in the next stage is the premise basis for rationally formulating mechanical equipment maintenance plans [3-4] . [0003] At present, the research on feature extraction method of vibration signal of rolling bearing has attracted extensive attention of scholars. Literature [5] proposed a rolling bearing fault diagnosis method based on morphological component analysis and envelope spectrum, which can effectively extract the fault features in the vibration signal of the rolling bearing. Literature ...

AI Technical Summary

Problems solved by technology

However, both of the above two solving methods need to set the search initial value based on experience, and the calculation time is relatively long

Images