Abstract: In view of the problem that the existing engine health state assessment model cannot effectively describe the engine health state when it is affected by local ignorance, an engine health assessment method is proposed based on belief rule base with power set (PBRB) in this paper. An interpretable description of the engine health condition assessment problem from the point of view of symbolic rules was provided. The system knowledge and observation data were effectively combined to build a PBRB-based health status assessment model, and the parameters were optimized to further improve the accuracy of the assessment results. The WD615 diesel was used as a case to prove the effectiveness of the model proposed in this paper.