A Fault Analysis And Prediction Of Aircraft Based On Association Rules And BP Neural Network

With the development of the times and the improvement of living standards,aircraft travel has become common,and aircraft accidents reported have also occurred from time to time.Therefore,the safety of aircraft in different fields,especially mechanical equipment,has received more and more attention,and more and more attention has been paid to aircraft equipment failure analysis and prediction methods.In-depth research and analysis of aircraft equipment maintenance has very important practical value and practical significance,especially for aircraft equipment failure analysis and prediction.For example,the analysis of aircraft equipment failures can help maintenance departments understand the correlation between various equipment,and the prediction of aircraft equipment failures can help maintenance departments find problems in a controlled manner in a timely manner.These methods can minimize the occurrence of disasters and minimize human casualties.For traditional single aircraft equipment inspection technology,this form not only consumes a lot of human and material resources,but also may have the disadvantages of long maintenance work hours and untimely discovery of hidden problems.Therefore,aircraft equipment failure analysis and prediction technology has become a hot trend in aircraft maintenance research topics.By using computer technology to analyze and predict hidden faults of aircraft equipment,not only can real-time simulation be performed on real data,but it can also provide more reasonable and effective opinions and suggestions for aircraft maintenance.The main work and innovations of this article are summarized as follows:At present,there are some technical shortcomings in the existing aircraft equipment fault detection technology,which cannot completely and efficiently detect the potential safety hazards of various aircraft equipment.In order to solve the existing shortcomings in the field,this paper proposes an aircraft equipment fault analysis and prediction algorithm based on association rules and BP neural network,and applies it to real data.This algorithm improves the shortcomings of the two aircraft maintenance methods when they are separately applied to aircraft equipment detection and prediction faults by combining aircraft equipment correlation and aircraft fault prediction methods.The aircraft equipment failure analysis method is based on the original Apriori algorithm,and the prefix path graph is added to improve the algorithm.By traversing the aircraft equipment fault database,the candidate set is constructed into the prefix path map in a specific form,and all frequent items and specific association rules are obtained by decomposing the prefix path map.The method of optimizing the algorithm structure by adding a prefix path graph can not only obtain the ability of association rules based on high support,It can also speedup traversing the candidate set to improve the algorithm's operation speed and the accuracy of obtaining association rules.The aircraft equipment failure prediction method uses the BP neural network algorithm based on Weibull distribution to analyze the equipment's first failure time database to derive the equipment failure prediction model.Combining the two methods can effectively improve the efficiency of detecting and predicting aircraft equipment failures,not only ensuring the rapid and effective search and discovery of hidden safety issues during aircraft maintenance,but also facilitating the design and implementation of equipment maintenance programs.This method provides guidance and assistance for real aircraft maintenance.The aircraft equipment failure analysis and prediction algorithm based on association rules and BP neural network is used to predict multiple equipment failures,which can provide a fast prediction retrieval scheme for aircraft maintenance,which not only makes the use of maintenance funds more reasonable,It also greatly shortened the maintenance time of the aircraft.