Evaluation of the wear state for diesel engine based on RIMER with the uncertain information

Literature review indicates that a large amount of failures are caused by abnormal wear of the diesel engine components. Therefore, it is essential to evaluate the wear state of the diesel engine. A large amount of oil information can be attained based on the ferrography analysis which can be well applied in the evaluation of the wear state for the diesel engine. Considering the uncertainty and vagueness of the information, an evaluation model of the wear state for the diesel engine was constructed based on the belief rule-base inference methodology using the evidential reasoning approach (RIMER), combining the expert experience with experimental data. In this paper, an EQD xx-xx diesel engine was selected as the research object and the oil information acquired from ferrographic analysis was acquired by the 1000h reliability test. To avoid the combinational explosion problem of the belief rule-base, feature selection with the interaction information was conducted to reduce the number of the antecedent attributes initially. By considering the difference of multi-experts experience, an initial BRB model was constructed to evaluate the wear state of the diesel engine. The study demonstrates that the model constructed by RIMER can accurately reflect the wear state of the diesel engine and take full use of uncertain information to improve the accuracy of the wear state evaluation.