Research On Aircraft Maintenance And Tracking System Based On Data Classification And Prediction Technology

With the development of computer technology and Internet technology,new demands and new technologies in the aviation sector are emerging.With the technical update,aviation maintenance continue to develop and change.The traditional maintenance methods can not meet the development needs.In order to ensure the continuous airworthiness of the aircraft,it is necessary to carry out systematic management of the aircraft and make full use of the existing technology to predict and analyze the possible fault.In this paper,a aircraft maintenance and tracking system is designed and developed according to airline needs,and introduces the classification prediction algorithm into the system.The system can not only improve the efficiency of aircraft maintenance staff,but also can arrange maintenance timely by predicting the aircraft engine fault types,thereby reducing the failure rate of the aircraft to ensure safe flight of the aircraft.Firstly,the paper analyzes the development and application of maintenance track system and technology of data classification and prediction.At the same time,introduce the method of data classification and prediction,propose a comprehensive evaluation model based on the neural network algorithm and random forest algorithm,and carry out simulation experiment of aircraft engine fault with this model,the experiment proves the feasibility of the comprehensive evaluation model for aircraft engine fault prediction.Secondly,demand analysis and design of the aircraft maintenance and tracking system,the main functions of the system include:1)Engine fault prediction:According to aircraft engine typical fault characteristic data and classification predictive algorithm predict the fault of the existing data,and a visualization chart of aircraft engine fault status is realized for the maintenance staff intuitive to show the current engine failure status.2)Maintenance warning:According to the parameters of the aircraft such as flight log,maintenance interval,utilization(number of landings per month,monthly flight hours)give the specific maintenance strategy and the expected maintenance date,and warning the component which is close to or beyond the expected maintenance date.3)Maintenance program:Defines the rules for all aircraft maintenance,different models have different definitions of maintenance rules,in order to meet the needs of different users of different models,users can apply for the corresponding maintenance program permissions according to their needs.4)Aircraft management:For user management and maintenance the basic information of their own aircraft,and different types of aircraft can bind the corresponding maintenance program.Finally,the practical case is given in the text which verifies the feasibility andvalidity of the system.This paper introduce machine learning algorithm to the practical application of aircraft maintenance for the purpose of providing a convenient maintenance service platform for airline which can not only improve the efficiency of the maintenance staff and save the cost of the airline but also reduce the failure rate of the aircraft and ensure personal safety.