Development And Research Of The Fault Diagnosis System For The Loading Device Of A Large-caliber Artillery

Automatic transmission system is one of the important subsystems of artillery.Its safe and stable response and matching with the overall performance will directly affect the reliability and feasibility of the artillery.Automatic loader is an integrated system of machine,electricity and liquid,the machinery is complex and the its working process is various.And it is one of subsystems that have higher failure rate.With the continuous improvement of the intelligentization of modern weaponry,the original artificial fault judgement has been unable to meet the actual demand,so the research of intelligent diagnosis system for automatic loader is imminent.This paper is based on the research project of "the 13th five-year plan",the reliability loading platform of large-caliber artillery as the research object.Through the establishment of artificial neural network expert system,the on-line fault diagnosis of automatic transmission system is realized.The main researches in this paper are listed as follow:(1)Intensive study of the failure phenomenon of the automatic transmission system of large caliber artillery is necessary,and based on the high performance requirements of modern warfare for intelligent weapons and equipment,the basic concept of fault diagnosis technology is put forward.In this paper,the detailed fault mode and effect analysis(FMEA)are carried out for the large-caliber artillery automatic filling system,which provides theoretical basis for the establishment of fault diagnosis expert system.(2)The expert diagnosis technology is introduced into the fault diagnosis by combining the fault characteristics of each subsystem of automatic loading of artillery.The structure and working principle of the expert system are introduced in detail,and the mathematical model of the fault diagnosis is established,and the fault diagnosis strategy of the subsystem is designed.(3)Aiming at the deficiency of traditional expert system,a method of expert system based on neural network is proposed.This paper analyzes the process of BP neural network and taking modular dispensing fault as an example,introduces the process of realizing BP neural network fault diagnosis using Matlab.(4)The software system of automatic loading prediction for large caliber artillery is developed on the Labview platform.The hardware structure and software realization of the system are introduced,and the effect of fault diagnosis and fault prediction are achieved through experimental demonstration.The fault diagnosis technology has been successfully applied in the practical application.