Health assessment method of centrifugal pump based on T-S fuzzy fault tree and remaining useful life
Aiming at the problem that the existing health assessment methods of centrifugal pump only consider single component,and can not objectively reflect the real health state of centrifugal pump,a health assessment model of centrifugal pump based on T-S fuzzy fault tree and remaining useful life was proposed.Firstly,a T-S fuzzy fault tree was built for the centrifugal pump fault.The T-S gates were used to connect the components to the centrifugal pump in the fault tree,and the T-S gate rules were used to characterize the relationship between the component failure probability and the centrifugal pump failure probability.Then,according to the deterioration monitoring of the components,the remaining useful life(RUL)prediction methods based on the Wiener process or Weibull distribution were used to calculate the components failure probability according to the condition monitoring data or the historical failure records of the components.Finally,the failure probability and health index of the centrifugal pump were calculated by combining the component failure probability with the T-S gate algorithm.And the health status of the centrifugal pump and the use or maintenance suggestions were obtained to realize the health assessment of the centrifugal pump.The health assessment method was applied to the centrifugal pump test bench dataset and the XJTU-SY dataset.The research results show that the health index of the centrifugal pump in the centrifugal pump test bench is 0.789,and the health state is relatively healthy.The method can comprehensively evaluate the state of the centrifugal pump considering multiple components.For the verification of Xi'an Jiaotong University cooperates vith SY company(XJTU-SY)dataset,comparing with the single-component health assessment method,the health index assessed by this method is always lower than 0.6,the time of reduction to 0.4 is significantly earlier than that of the single-component health assessment method,and the assessment results are more accurate.The method provides a reference for the improvement of the health assessment method of centrifugal pump.
centrifugal pumpT-S fuzzy fault treehealth status assessmenthealth indexfailure probability calculationT-S gate algorithm
万方数据
维普
