基于神经网络的固态功率控制器的故障诊断研究  

Fault diagnosis research of solid state power controller based on neural network

作　　者：汤恒先[1] 林蓉[2] 庞骥庭 周军涛[1] 

机构地区：[1]西北工业大学第365研究所,陕西西安710065 [2]空军工程大学,陕西西安710051

出　　处：《电子设计工程》2015年第5期187-189,192,共4页Electronic Design Engineering

摘　　要：通过分析总结固态功率控制器(SSPC)的故障经验,训练样本,设计实现一个基于神经网络的固态功率控制器故障诊断系统。神经网络权值和阈值会随着实际的排故结果不断进行更新,而且神经网络正向推理是一种并行推理,较高了推理速度,克服了冲突,具有很强的故障诊断能力。仿真结果表明该方法收敛速度快,诊断精度高,自适应性强,能够有效地诊断故障。Based on the analysis and sum-up of fault experience of solid state power controller, a neural network-based solid state power controller fault diagnosis system is designed and realized by learning sample. The weight value and threshold value of neural network can be renewed with practical fault results, and forward inference of neural network is a apposite inference,which enhances inference speed and overcomes confliction, so it has a strong fault diagnosis capability. The simulation results show that the method has fast convergence speed, high accuracy, strong adaptability, can effectively diagnose the faults.

关 键 词：BP神经网络 故障诊断 固态功率控制器 网络训练 

分 类 号：TN710[电子电信—电路与系统]