Method for evaluating credibility of prior data

By constructing the acceleration factor invariance theory and gamma process modeling, identifying stress-independent and stress-dependent parameters, quantifying the parameter credibility and weighting them, the problem of mismatch between prior data and actual working conditions is solved, and the applicability and accuracy of the model in multiple stress environments are improved.

CN120805748AActive Publication Date: 2025-10-17HANGZHOU INTERNATIONAL INNOVATION INSTITUTE OF BEIHANG UNIVERSITY
2 Cites 0 Cited by

Patent Information

Application Number
CN202511324962.7
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-09-17
Publication Date
2025-10-17
Estimated Expiration
2045-09-17

AI Technical Summary

Technical Problem

When processing prior data, existing technologies have the problem of mismatch between prior data and actual working conditions, which leads to reduced model accuracy. In particular, it is difficult to ensure the applicability and accuracy of the model in a multi-stress environment.

Method used

By constructing a degradation model based on the acceleration factor invariance theory, stress-independent and stress-dependent parameters are identified. Combined with gamma process modeling, the credibility of parameters under different stress levels is quantified, and weighted calculation is performed through the hazard coefficient to evaluate the comprehensive credibility of the data.

Benefits of technology

The applicability of degradation modeling in multi-stress environments and the robustness of model reasoning are improved, the risk of misjudgment is reduced, and the representativeness and reliability of accelerated degradation experimental data in Bayesian modeling are ensured.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention relates to the technical field of reliability engineering, in particular to a priori data credibility evaluation method, which comprises the following steps: S1, parameter identification: identifying parameters irrelevant to stress and parameters relevant to stress in a degradation model; s2, consistency relationship construction: constructing a parameter relational expression which needs to be met by failure mechanism consistency; s3, consistency discrimination: quantizing the deviation between parameter values at the reference stress level; s4, parameter consistency evaluation: evaluating the consistency degree between the parameter set under each acceleration stress and the parameter set under the normal stress, and calculating the credibility of the data under the corresponding stress level according to the consistency degree; s5, credibility evaluation: weighting the importance of the data under different stress levels to obtain comprehensive credibility; according to the method, the credibility of the prior data is quantified, comprehensive credibility weighted calculation is realized in combination with the hazard coefficient, and the effectiveness and accuracy of the accelerated degradation data in reliability modeling are improved.
Need to check novelty before this filing date? Find Prior Art