Research On Reliability Analysis Of Hydraulic Torque Converter Based On Fuzzy Bayesian Network

Because of its strong mobility and applicability to complex landform,tracked armored walking vehicle has attracted the attention of various countries since it came out.The reliability of torque converter,as the core component of tracked fighting vehicle,directly affects the combat performance of tracked fighting vehicle.Therefore,the research on the reliability analysis method of hydraulic torque converter is of great practical significance for improving the combat ability of our army.This paper takes the torque converter as the research object,studies its reliability analysis method,and proposes a reliability analysis method based on fuzzy Bayesian network for tracked armored infantry vehicle torque converter.Firstly,the technical route of reliability analysis of hydraulic torque converter is determined.The structure and sample data of hydraulic torque converter are analyzed and the decision table of hydraulic torque converter is established based on rough set theory.The Bayesian network has a strong ability to deal with uncertain problems and can learn and reason under the condition of limited,incomplete and uncertain information.Combined with the reduced decision table,the Bayesian network is built to carry out reliability analysis.Secondly,the resolution matrix reduction algorithm based on genetic algorithm is adopted in this paper to reduce the attributes of the decision table,which solves the NP-hard problem and removes the data with the same conditional attributes in the sample data.The convergence speed and accuracy are better than the conventional attribute reduction algorithm.Thirdly,fuzzy Bayesian network is established to analyze the reliability of hydraulic torque converter.According to the actual operating state of the torque converter,the dimorphism(normal and failure)of the parent node of the Bayesian network is changed into trimorphism(normal,slight failure and failure).Based on the forward reasoning ability of Bayesian network,the improved Gaussian barrel elimination method is used to determine the fault probability of sub-nodes.Finally,two methods of posterior probability and probability importance are used to analyze the reliability of the parent node in the fuzzy Bayesian network,and then according to the analysis results,the inspection and maintenance opinions of each component of the hydraulic torque converter are put forward.