Research On The Technology Of Prognostics And Health Management Based On Aerospace Special Vehicle

Aerospace special vehicle is one of the indispensable equipment in the field of aerospace,and it plays a vital role in missions.With the improvement of related technology,the vehicle performance develops rapidly and the system tends to be complicated,which greatly increases the development cycle and cost,and brings great challenges to the daily maintenance and repairing work.At present,only relying on traditional empirical analysis methods for fault detection will increase a lot of labor costs,which has been difficult to meet the task requirements.Therefore,in order to achieve the goal of fault prediction and reducing the maintenance cost for aerospace special vehicle,a certain aerospace special vehicle is taken in this paper as the research object,and based on neural network prediction method,fault tree analysis(FTA)and support vector machine(SVM)classification algorithm a new fault prediction method and health management algorithm are proposed.At the same time,an integrated and intelligent system designed and established,which can predict unknown faults and evaluate the health status of aerospace special vehicles.In this paper,firstly,combining with the requirements of fault prediction and health management for special vehicles,we analyses the current research situation and the development of PHM technology in aerospace special vehicles.Secondly,based on some technical data,the software framework of PHM system for special vehicles is designed.Then,according to SOM neural network theory,and combined with the vehicle running data,we completes the analysis of fault prediction to improve the accuracy of prediction by optimizing parameters and improving methods.And then completes health status assessment of aerospace special vehicles by FTA and SVM algorithm.Finally,based on the mixed programming method of MATLAB and MFC,the application of prediction algorithm and health assessment algorithm is realized,based on MFC framework and Microsoft Visual Studio 2015 IDE,code to complete the functions of each module.The innovations of this paper include the following two aspects:(1)This paper uses the fault prediction algorithm based on SOM neural network to give detailed prediction results for unknown faults of aerospace special vehicles,and recompiles the model files by mixed programming method,which makes the prediction model applicable to multi-type equipment that achieves the generality of this model;(2)This paper uses the combination of FTA and SVM algorithm to complete the health status assessment.Visual graphics and lists are used to display the health assessment results of the system in detail and intuitively,which is convenient for maintenance personnel to operate the software and understand the system.After testing functional modules of the aerospace special vehicle prognostics and health management software,the results of the prediction and health assessment are verified.The software meets the actual needs with practicability and versatility,which is helpful to monitor and maintenance of the aerospace special vehicle.