基于小波时频图和多尺度卷积神经网络的发动机工况识别研究

Research on engine condition recognition based on wavelet time-frequency graph andmulti-scale convolutional neural network

针对传统工况识别方法对非平稳的汽车发动机音频信号难以准确识别的问题,提出一种基于小波时频图和多尺度卷积神经网络的发动机工况识别方法。首先,将原始信号通过连续小波转化为小波时频图,其次,对小波时频图进行统一的预处理,最后将处理好的图片输入到卷积神经网络中提取多尺度特征并分类识别。该方法有效结合了具有处理非线性平稳信号优势的小波时频分析和卷积神经网络的图像分析能力。在测试集数据转速不同的情况下,识别准确率和鲁棒性更好。

Aiming at the problem that the traditional method is difficult to accurately identify the non-stationary automotive engine audio signal,an engine condition recognition method based on wavelet time-frequency graph and multi-scale convolutional neural network is proposed.First,the original signal is transformed into a wavelet time-frequency graph through continuous wavelet.Secondly,the wavelet time-frequency graph is pre-processed uniformly.Finally,the processed image is input into the convolutional neural network to extract multiscale features and classify them.This method effectively combines the wavelet time-frequency analysis with the image analysis capability of convolutional neural network,which has the advantage of processing nonlinear stationary signals.The recognition accuracy and robustness are better when the test set data speed is different.