Evaluation method for accelerated degradation testing with interval analysis

Traditional evaluation methods of accelerated degradation testing (ADT) are based on precise degradation data to conduct reliability and lifetime assessment. However, with interfere of the uncertainties from human factors, the test data can be imprecise represented by interval rather than precise data. Under this consideration, an interval analysis method for ADT evaluation was proposed based on Wiener process, which included possibility and necessity models. Interval regression method was firstly used to transfer the problems of modeling interval degradation data under different accelerated stress levels into quadratic programming problems. The interval drift coefficients under different stress levels with possibility model and diffusion coefficient were obtained. Then the interval drift coefficients were extrapolated to normal stress condition with accelerated model under necessity model, and further to analyze the relationship between measurement uncertainty and reliability and lifetime evaluation results. Finally, the numerical study was used to present and verify the proposed methodology, and conduct uncertainty sensitivity analysis. The results show that both reliability and lifetime evaluation results are effected by epistemic uncertainty of measurement, and their correctness can be ensured with decreasing epistemic uncertainty.