【摘要】 为提高车窗电机异常噪声特征提取的有效性及分类识别的准确性,提出一种以优化的梅尔频率倒谱系数(Mel Frequency Cepstrum Coefficient,MFCC)为特征值,以支持向量机(Support Vector Machine,SVM)为噪声辨识模型的电机异常噪声辨识方法。在MFCC提取方法基础上,针对频谱泄漏,用Hanning自卷积窗代替Hanning窗,获得优化的MFCC,并将其作为特征值输入到SVM进行异常噪声辨识。为提高SVM判别准确率,采用人工蜂群算法实现SVM参数选择优化。实验结果表明,该方法能够有效判别电机是否存在异响,准确率达到91%。更多还原

【Abstract】 In order to improve the efficiency and accuracy of classification and recognition of vehicle window motor abnormal noise,a new method based on the optimal MFCC taken as characteristic values and a SVM taken as the noise identification model was proposed. On the basis of MFCC extraction method,Hanning window was replaced with Hanning self-convolution windows aiming at spectrum leakage,and the optimized MFCC taken as characteristic values were input into SVM to identify abnormal noises. At the same time,the artificial bee colony algorithm was used to optimize the parameters of SVM and improve the accuracy of SVM. The test results showed that the proposed method can effectively distinguish the presence of the abnormal noise of a vehicle window motor,the accuracy reaches 91%.更多还原