VGG16 and Morlet Wavelet-Based Anomaly Detection for Automotive Actuators

This paper proposes a method for anomaly detection in automotive air conditioning actuator(AACA) based on acoustic feature extraction and deep learning. The Morlet wavelet transform is used to convert the acoustic signal sequence of the AACA into a time-frequency graph, extracting the transient features and spectral information of the signal. Then, the pretrained VGG16 model is employed to extract high-level image features from the time-frequency graph. Finally, the Support Vector Data Description (SVDD) classification algorithm is used for anomaly detection. The test results show that this method can effectively distinguish between normal and abnormal states of the automotive air conditioning actuator, achieving high detection accuracy and robustness. This provides a new solution for real-time monitoring and fault prediction of automotive air conditioning actuators.