Rolling bearing residual life prediction method based on data and model adaptive matching

Abstract

A rolling bearing residual life prediction method based on data and model adaptive matching comprises the following steps: firstly establishing a state space model related to the accumulative number of rotation turns of a rolling bearing, then determining three thresholds, and determining an initial degradation time and an initial prediction time based on the thresholds; then, on the basis of an observation value sequence after the initial degradation time, updating model parameters by a maximum likelihood method, and automatically selecting an optimal state space model by using a Bayesian information criterion; and finally, according to the selected model and the updated parameters, calculating a probability density function of failure time to obtain a residual life prediction result. The method can be driven by state monitoring data of the rolling bearing without referring to monitoring data of any other rolling bearing, the monitoring data is matched with the optimal state space model in real time, the degradation condition of the rolling bearing in the operation process in industrial practice is effectively represented, and the prediction precision of the residual life of the rolling bearing is improved.

Description

Technical field [0001] The present invention belongs to the field of rolling bearing residual life prediction technique, and specifically, the residual life prediction method of data and model adaptive matching. Background technique [0002] In recent years, with the rapid development of military technology and the overall improvement of the modern industrial level, the requirements for mechanical equipment safe operation are also increasingly high. The core components in the equipment are extremely difficult to vary from degrees, resulting in failure of equipment, causing economic losses and casualties. Therefore, ensuring that the key components of the machinery and equipment are safe to serve, and the security of national defense safety and promotion of modern industrial production significance is significant. The rolling bearings play an important role in most rolling bearing systems, known as the "industrial joint", and its health status is directly affected to the operation...

AI Technical Summary

Problems solved by technology

However, for actual engineering equipment, there is often only a small amount of failure data or no failure data; even if sufficient failure data is accumulated, the degradation process of different rolling bearings is often different due to factors such as manufacturing differences, environment, and working conditions, and it is very difficult to It is difficult to guarantee the adaptability of the trained model on different devices

Therefore, the above two assumptions are unreasonable in some practical situations, resulting in poor performance of the data-driven remaining life prediction method based on this assumption

Images