A robust performance evaluation method based on interval evidential reasoning approach under uncertainty

Abstract

Performance evaluation (PE) is an important part of equipment health management. If the monitoring information of the equipment is interfered, the evaluation results may be erroneous. A robust performance evaluation (RPE) method is proposed to solve this problem. The performance evaluation results are obtained by distinguishing the cases of single evidence with interference and two evidence with interference, and a robustness measurement based on interval similarity is proposed. To improve the accuracy of the evaluation results, the referential values in the IER evaluation model are optimized. The robustness thresholds of the input indexes are obtained under the satisfaction of the robustness constraints. If the interference value of the input index is within the thresholds, the deviation between the evaluation results using monitoring information with interference and those using monitoring information without interference is small. Finally, the proposed method is applied to a type of electric servo mechanism performance evaluation, and the case shows the validity of the RPE method.