Research On Radar System Fault Modeling And Diagnosis Technology

Studying the fault diagnosis of radar system is of great significance to ensure the reliability of system operation.With the development of radar technology,the structure of radar systems is becoming more and more complex,the types of faults are increasing,the coupling and ambiguity between faults are getting higher and higher,and fault diagnosis becomes more and more difficult.How to quickly locate faults based on fault symptoms has become a research hotspot in the field of radar.Based on the radar fault data provided by the cooperative research institute,this paper takes a certain type of airborne radar as the research object,and conducts in-depth research on radar fault diagnosis.The main research of this article is as follows:Firstly,the hierarchical division of the radar system is realized.In view of the complex structure of the radar system,it is very difficult to directly conduct fault analysis.In this paper,based on the radar system structure and the hierarchical characteristics of fault propagation,the radar system is divided into layers to lay the foundation for subsequent radar fault diagnosis.Secondly,selection of test points for radar system is completed.At present,at the beginning of radar system design,a large number of test points are often set up internally.In order to avoid the problems of increased test cost,large fault diagnosis model and increased calculation volume caused by redundant tests,before the radar fault modeling,the test point selection of the radar system is studied,and an improved particle swarm algorithm is proposed.Experimental results show that the improved particle swarm optimization algorithm has a better overall effect in the selection of test points.Thirdly,a fault Petri net diagnosis model is established.Unlike the traditional fault Petri net model,which only focuses on the fuzzy correlation between faults,this paper introduces the fuzzy correlation between the test and the fault into the Petri net model,which takes into account the uncertainty of the test,which is more in line with the actual radar fault system and improves the model reliability.In addition,in order to avoid the impact of the huge structure of the Petri net model on fault diagnosis,based on the idea of hierarchical modeling,this paper implements the hierarchical modeling of the radar Petri net model.Fourthly,based on the improved BP algorithm,the parameters of the Petri net model are optimized.In order to get rid of the model's excessive dependence on expert experience,improve the reliability and self-adaptive ability of the model,the problem of Petri net model parameter optimization was studied,and an improved BP algorithm was proposed.Experimental results show that improving the BP algorithm is of great significance for improving the accuracy and generalization ability of the model.Fifthly,based on the mixed positive and negative reasoning method,radar fault diagnosis is realized.Based on the hierarchical model of Petri net,and adopting positive and negative mixed inference algorithm,the hierarchical fault diagnosis of radar system is realized.The comparison with the fault tree diagnosis method shows that the fault Petri net model adopted in this paper has obvious advantages in model complexity,model description ability and diagnosis efficiency,and has high engineering application value.Lastly,the radar fault diagnosis system based on Petri net model is developed.The system is based on Petri net fault diagnosis technology,using QT for graphical interface development,MATLAB for data processing and My SQL database for data storage,and back-end and frontend data interaction through dll files.In addition,the model information is retained through the mat file,which realizes one-time modeling and multiple uses.