Bearing inner ring fault residual life prediction method using FBM-based long correlation model

Abstract

The invention relates to a bearing inner ring fault residual life prediction method using an FBM-based long correlation model. The method comprises the following steps: 1) acquiring a vibration signalby an acceleration sensor, converting an acceleration signal into a speed signal, and calculating signal energy of the vibration signal; 2) decomposing an energy sequence by adopting an empirical mode decomposition method, decomposing an original signal into a plurality of IMFs with limited bandwidths, and selecting a component reflecting a bearing degradation trend; 3) acquiring a Hurst parameter value of the decomposed energy sequence by utilizing a rescaled range analysis method; 4) estimating FBM degradation model parameters by adopting a maximum likelihood estimation method, and buildingan FBM-based degradation prediction model; and 5) predicting the model by adopting a Monte Carlo method to obtain the residual service life in a current state. Compared with the prior art, the methodhas the advantages that the prediction precision can be improved, the operation faults of bearings can be found in time, the fault occurrence rate in the normal production process can be reduced, andthe economic benefits of enterprises can be improved.

Description

technical field [0001] The invention relates to a method for predicting the remaining service life of a bearing inner ring fault, in particular to a method for predicting the remaining service life of a bearing inner ring fault based on an FBM long correlation model. Background technique [0002] In modern industrial society, machinery and equipment continue to develop towards precision and automation, and their structures are becoming more and more complex. Bearings are an integral part of mechanical equipment, and their operating status is directly related to whether the equipment can run well. Weak faults will not appear significantly during the operation of the equipment, but there is a process of time accumulation inside the equipment, and when it reaches a certain level, there will be problems in the operation of the equipment. Therefore, it is very important to monitor the operation status of mechanical equipment and predict the failure of mechanical equipment. [0...

AI Technical Summary

Problems solved by technology

However, the prediction model used is a Markov type model. The model does not have memory, that is, the prediction process is a memoryless process, and the occurrence of bearing faults is a long-term and slowly changing process. neglect

Images