Rolling bearing modeling and model updating method and system based on digital twinning

Abstract

The invention discloses a rolling bearing modeling and model updating method and system based on digital twinning. The method includes using a sensor for monitoring operation data of a rolling bearing in real time and preprocessing the operation data; carrying out coupling modeling to obtain a dynamically updated digital twin model; inputting the normalized operation data into a digital twinborn model in real time, performing simulation calculation on output data of the rolling bearing by using the digital twinborn model, comparing simulation data of the digital twinborn model with the operation data in a corresponding time interval, and if a residual sequence is white noise with a zero mean value, collecting a corresponding rolling bearing digital twin model; otherwise, adjusting and correcting the parameters of the rolling bearing digital twinning model according to the criterion function to obtain a real-time synchronous rolling bearing digital twinning model. The invention can reflect the time history of the health state of the bearing, and provides a basis for fault diagnosis and life prediction of the rolling bearing.

Description

technical field [0001] The invention belongs to the technical field of mechanical diagnosis intelligence and digitization, and in particular relates to a method and system for modeling and updating a rolling bearing based on digital twins. Background technique [0002] As a key basic component in rotating machinery, rolling bearings are widely used in fields such as aero-engines, wind power and high-speed rail. Usually, the service environment is harsh and the working conditions are complex. Despite regular maintenance and inspections, unpredictable failures will inevitably occur. The failure of rolling bearings will reduce the working performance of mechanical equipment, and cause the overall damage of equipment, resulting in huge economic losses and casualties. In order to realize the real-time monitoring of rolling bearings and reasonably predict the working state of the bearings, it is urgent to establish a model that is highly similar to the real rolling bearings. Sin...

AI Technical Summary

Problems solved by technology

Most of the existing rolling bearing performance evaluation technologies are driven by traditional physical models or data, but the multi-source monitoring information under big data is coupled with each other, which is characterized by high-dimensional heterogeneity and uneven distribution. The situation is changeable, and its degradation trend cannot be accurately predicted solely based on physical model information or data information

Images