Circuit breaker residual electric life prediction method and system based on multi-physical field modeling

By combining multiphysics modeling and machine learning, dynamic resistance-stroke curves are generated, solving the problem of assessing the erosion state of circuit breaker contacts, achieving high-precision prediction of remaining electrical life, and improving the condition awareness and predictive maintenance capabilities of power equipment.

CN122287307APending Publication Date: 2026-06-26CHONGQING UNIV +2
View PDF 3 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
CHONGQING UNIV
Filing Date
2026-03-17
Publication Date
2026-06-26

AI Technical Summary

Technical Problem

Existing technologies struggle to accurately assess contact erosion and predict remaining electrical lifetime without disassembling the circuit breaker. Furthermore, existing methods suffer from high computational costs, data scarcity, and unstable prediction results.

Method used

By constructing a multi-physics coupled contact erosion simulation model, generating dynamic resistance-stroke curves, extracting feature quantities, and training the model using machine learning algorithms, and combining sequential Bayesian theory to update the degradation model parameters, high-precision estimation of contact erosion size and prediction of remaining electrical lifetime are achieved.

Benefits of technology

It enables accurate and quantitative assessment of the erosion size of circuit breaker contacts, improves the accuracy and reliability of remaining electrical life prediction, eliminates the dependence on scarce field data, and provides reliable operation and maintenance support.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122287307A_ABST
    Figure CN122287307A_ABST
Patent Text Reader

Abstract

A method and system for predicting the remaining electrical lifetime of circuit breakers based on multiphysics modeling are proposed. The method includes: building a contact erosion simulation model based on multiphysics coupling to obtain the contact erosion size and its corresponding dynamic resistance-stroke curve; establishing a feature vector of the contact erosion size with respect to characteristic quantities, and training a contact erosion estimation model using a machine learning algorithm; collecting operating data of the target circuit breaker to obtain an estimated value of the current contact erosion size, establishing a degradation model of the contact erosion size, and establishing a failure state criterion for the circuit breaker; updating the degradation model parameters based on sequential Bayesian theory, and predicting the cumulative number of operations required for the contact erosion size to degrade from the current estimated value to the point where the failure state criterion is first met, as the remaining electrical lifetime of the circuit breaker. This invention reduces reliance on field test data, utilizes easily measurable data to inversely derive the contact erosion size, which is difficult to detect directly, and integrates a sequential Bayesian algorithm for dynamic prediction, providing information support for operation and maintenance decisions.
Need to check novelty before this filing date? Find Prior Art