Research On State Evaluation And Diagnosis Method Of Fire Control Computer Based On Machine Learning

With the rapid development of national defense science and technology,the degree of integration,complexity,and intelligence of armored vehicles has also greatly increased.The fire control system is the fire control center,which directly determines the combat power of the armored vehicle.As a key component of the fire control system,the fire control computer works under severe working conditions all the year round,which makes the fire control computer extremely prone to abnormalities.The state evaluation and diagnosis of the fire control computer in time can reduce the time of exception elimination and reduce maintenance costs.,Is of great significance to the fire control system.In this paper,the state evaluation and diagnosis of the fire control computer are realized by constructing a state evaluation model and a fault diagnosis model.The fire control computer has a complex structure,a large number of data signals and a strong coupling relationship between various modules.It is difficult to evaluate the actual operating state with a single method.Therefore,this paper proposes a state evaluation method based on the combination of weights and gray clustering principles.Evaluation of the operating status of the fire control computer.Under normal circumstances,the fire control computer has five operating states,and the purpose of the state assessment of the fire control computer is to determine which state it is in.The operating status of the fire control computer is usually characterized by quantitative and qualitative indicators.The general method cannot handle both quantitative and qualitative indicators.Therefore,this paper adopts a weighting strategy to deal with quantitative and qualitative indicators,and obtains subjective weights by improving the analytic hierarchy process.The analysis method relies on expert experience to construct the judgment matrix in the process of solving,and there is a certain degree of subjectivity.Therefore,this paper uses the entropy weight method to obtain the objective weight,and combines the subjective weight with the objective weight to obtain the combined weight,which makes the construction of the weight more reasonable.According to the principle of cluster analysis in gray system theory,the final evaluation result needs to be obtained based on the combined weight and evaluation weight matrix.This paper calculates the turning threshold of the whitening weight function through the fuzzy C mean value,and then calculates the evaluation coefficient through the whitening weight function.Get the evaluation weight matrix of the evaluation index.In this paper,the state of the fire control computer is better evaluated by combining the optimization weight and the evaluation weight matrix.In addition,the power module of the fire control computer provides "power" for the entire fire control computer.Once a failure occurs,the system will be paralyzed.To ensure the normal operation of the fire control system,it is necessary to diagnose the fire control computer power module.This paper proposes a fault diagnosis method based on the historical fault data of the power module.The method first reduces the attributes of the data through the neighborhood rough set to obtain the key attribute set,and then builds the fault classification model through the support vector machine,but the support vector machine is Binary classification problem,so the binary tree method is used to construct multiple support vector machines to solve the multi-classification problem.Finally,in the course of the experiment,it was found that the quality of the parameters easily affects the classification efficiency,so the Antlion algorithm was introduced to optimize the parameters in the support vector machine to establish a more accurate classification model.Through experiments,the state evaluation model proposed in this paper can accurately evaluate the state of the fire control computer and provide guarantee for the safe operation of the fire control computer.The diagnostic model can classify the faults of the power supply module and provide a basis for subsequent troubleshooting.In this paper,the state evaluation and diagnosis of the fire control computer are better realized by optimizing the state evaluation model and the fault diagnosis model.