小波神经网络在雷达故障诊断中的应用研究  

Application Research of Wavelet Neural Network in Radar Fault Diagnosis

作　　者：涂望明[1,2] 宋执环[1] 孟庆志[1] 张国超[1] 苏帅[1] 

机构地区：[1]浙江大学工业控制技术国家重点实验室,工业控制技术研究所,浙江杭州310027 [2]武汉军械士官学校雷达系,湖北武汉430075

出　　处：《计算机测量与控制》2012年第4期921-923,共3页Computer Measurement &Control

摘　　要：根据某型雷达接收机的信号特点,提出了基于小波神经网络和专家知识相结合的雷达智能故障诊断方法,探讨了该方法在某型雷达接收机故障诊断中的应用,采样信号经过小波去噪和小波特征向量提取后再进行归一化处理,作为小波神经网络的输入向量,小波神经网络隐含层的激活函数选用小波函数,完成雷达接收机典型故障的诊断;MATLAB实例仿真结果表明,该方法诊断准确,智能化水平高,有很好的自学习能力,提高了故障诊断的正确性和效率。According to the signal characteristic of some kind of radar, proposed the intelligence fault diagnosis method based on the wavelet neural network and expert knowledge, the application of this method in the middle frequency receiver of the radar is introduced, the sampled signals, after wavelet de--noising and wavelet feature vector extraction and then normalization, as the input vector of the wavelet neural network, the wavelet function is selected as the activation transfer function of the hidden layer to complete the categories of the typical faults in the radar receiver. The actual example results of the simulation with MATLAB tools indicate that this method has accurate diagnosis and higher intelligence level and better self--learning power and improves the accuracy and efficiency of the fault diagnosis.

关 键 词：雷达 小波神经网络 接收机 故障诊断 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] TP273[自动化与计算机技术—控制科学与工程]