Health Management And Fault Prediction Model For Complex Equipment

In recent years,failure prediction and health management(Prognostics and Health Management,PHM)technology in the field of aerospace and electronic equipment is developing rapidly,but the research in ground weapons equipment is still in its infancy,has not formed a complete system structure.The newly developed self-propelled artillery is a complex equipment,involving multiple disciplines and fields,and its reliability and stability are key conditions to ensure the successful completion of the equipment’s operational tasks.Therefore,the study of PHM technology in self-propelled artillery to assess the health status of equipment and predict unknown failures is of great significance and practical application value to improve the efficiency of equipment maintenance and guarantee.The research content of the paper mainly includes the following aspects:(1)A variable-weight fuzzy comprehensive assessment model of equipment health status based on deterioration degree is studied and established.As the health of self-propelled guns is a gradual degradation process,their failure occurrence has a certain ambiguity.In this paper,the equipment health status is divided into 5 levels:healthy,good,attention,abnormal and fault.The deterioration degree is used as a unified measure,based on the hierarchical analysis method,and the weight vector is modified using the principle of variable weight,so as to determine the weight value of each assessment index.The deterioration degree values of each assessment index are calculated,and the ridge-shaped affiliation function is used to construct the affiliation functions corresponding to the five health status classes and build the affiliation matrix.The health status of the equipment is then assessed using the idea of fuzzy integrated judgement and the principle of maximum subordination.Finally,taking a certain type of self-propelled artillery full gun electrical system as an example,the algorithm proposed in this paper was applied to assess the health status of the equipment and obtained a "good" rating.From the subsequent inspection and maintenance records,the evaluation result is close to the actual operating state of the equipment,which verifies the reasonableness and validity of the model.(2)An improved grey Markovian fault prediction model based on this model was studied and developed.In response to the problem that the division of state intervals in the traditional grey Markov prediction model is more arbitrary and the prediction accuracy is lower,this paper improves the calculation of state transfer probability and proposes a grey Markov fault prediction model based on the improvement.The method first uses the grey prediction model to obtain the initial prediction value of the fault interval,then uses the Markov chain prediction model to correct the initial prediction result,and when the state interval is divided,the probability average expectation value of the state in which the relative value of each interval is located is used as the next transfer value,so as to obtain the prediction result,and finally uses the relative error test method to determine the accuracy of the model.In this paper,the historical data of a sub-system failure interval of a type of self-propelled artillery gun is used as an example for analysis and validation,and the results show that the average relative error of the traditional prediction model is 6.8%,and the average relative error of the improved prediction model is 2.79%,which is 4.01%less than that of the traditional prediction model,improving the accuracy of the prediction results and verifying the effectiveness of the method proposed in this paper.