论文部分内容阅读
根据雷达接收机信号特点,提出了将小波变换和最小二乘支持向量机相结合的雷达故障诊断新方法。首先,根据专家经验选取电路中恰当的测试点,运用小波变换对采样数据进行处理和特征提取,然后建立了雷达故障诊断模型。最后再运用最小二乘支持向量机(LS-SVM)的基本原理和分类方法进行故障诊断,并在某型雷达接收机故障诊断中进行了实际应用研究。采样信号先经过小波降噪处理,以减少采样引人的误差,再进行小波分解提取能量系数作为雷达的故障特征向量,经归一化处理后,作为输入向量,经诊断模型输出后完成雷达接收机典型故障的诊断。MATLAB实例仿真结果表明,该方法有很好的分类能力,提高了雷达故障诊断的正确性和效率。
According to the characteristics of radar receiver signals, a new radar fault diagnosis method based on wavelet transform and least squares support vector machine is proposed. First of all, according to the expert’s experience, we select the appropriate test points in the circuit, and use the wavelet transform to process the sampled data and extract the features. Then, the radar fault diagnosis model is established. Finally, the basic principles and classification methods of least square support vector machine (LS-SVM) are used to diagnose the fault. The practical application of a certain type of radar receiver fault diagnosis is also studied. The sampled signal is first processed by wavelet denoising to reduce the error of sample introduction, and then the wavelet decomposition is used to extract the energy coefficient as the fault eigenvector of the radar. After normalized, the sampled signal is output as input vector, and the radar receives Machine typical fault diagnosis. The result of MATLAB simulation shows that this method has good classification ability and improves the correctness and efficiency of radar fault diagnosis.