System level health condition assessment method of complex equipment under uncertainty based on D-S evidence theory

Different from component level health condition assessment which only uses single type of certainty information; system level health condition assessment is often a multiple attribute decision making problem with both quantitative and qualitative information under uncertain environment. In this paper, the analysis process and algorithm for system level health condition assessment problem is proposed on the basis of D-S evidence theory and entropy weight. Various types of uncertain information, including nondeterministic linguistic information, deterministic quantitative information and interval grey information is studied. First of all, system level health condition assessment and health condition classification are defined. Secondly, the assessment process is given. To get the entropy weights, standardized methods of different types of health information are put forward. Thirdly, the 3 types of health information are transformed into basic belief assignment functions under the same recognition framework. Finally, all types of health information are fused. Application analysis shows that the method is feasible and effective.