基于BP网络的飞机燃油系统故障诊断方法研究  

Research on the Method of Fault Diagnosis of Aircraft Fuel System Based on BP Network

作　　者：龙浩[1] 王新民[2] 

机构地区：[1]北京联合大学自动化学院,北京100101 [2]西北工业大学自动化学院,西安710072

出　　处：《北京联合大学学报》2007年第1期52-56,共5页Journal of Beijing Union University

基　　金：航空科学基金项目(04E09002)

摘　　要：对于飞机燃油系统故障诊断问题,提出了一种利用BP神经网络进行飞机燃油系统故障诊断的新方法。通过设计飞机燃油系统层次分类诊断模型,构建了专家系统知识库和推理机,利用CB工具开发了飞机燃油神经网络故障诊断专家系统。仿真结果表明,该系统很好地发挥了专家系统的智能性,能够快速准确地诊断出燃油系统的故障,完全适合于飞机燃油系统的故障诊断。A new method for fault diagnosis of aircraft fuel system using BP network as a way to prevent the problem of fault diagnosis of aircraft system is presented. By designing muhilayer neural network model of aircraft fuel system, knowledge base and infer engine are built, and by means of CB tool , the relevant software environment is set up . Consequently the neural network fault diagnostic expert system of aircraft fuel system is successfully developed . The emulational results prove that the intelligence of expert system works perfectly and the faults of fuel system are diagnosed well and fast ; furthermore, effective methods of reconstruction can be brought forward in the expert system. It is entirely practicable in aircraft fuel fault diagnosis system.

关 键 词：飞机燃油系统 故障诊断 神经网络 专家系统 

分 类 号：V228.1[航空宇航科学与技术—飞行器设计]