Research on combination of data-driven and probability-based prognostics techniques for equipments

For fault prognosis process historical data is not complete, such problems as lack of prior knowledge, a combination of data-driven and probability-base prognostics method is proposed. First, the nonlinear relationship of failure rate and state parameters is established with Weibull Proportional Hazard Model. Second, the condition monitoring data of actual operated equipment is put into least-squares nonlinear regression to obtain the failure rate trend. Finally, the failure criteria are determined with statistical knowledge, and then the remaining useful life of the equipment is predicted. The proposed approach is applied to prognosis of marine diesel engine, and the testing results show that the proposed method can reliably predict the remaining useful life of the equipment in real time with state parameters.