Signal Status Recognition Based on 1DCNN and Its Feature Extraction Mechanism Analysis

Sensors

Yangyang Wang¹  Shuzhan Huang¹  Juying Dai²  Jian Tang² 

[1] Graduate School, Army Engineering University of PLA, Nanjing 210000, China;School of Field Engineering, Army Engineering University of PLA, Nanjing 210000, China;

关键词: convolutional neural network;    intelligent fault diagnosis;    convolution kernel;    feature extraction mechanism;   

DOI  :  10.3390/s19092018

来源: DOAJ

【 摘 要 】

In this paper, we construct a one-dimensional convolutional neural network (1DCNN), which directly takes as the input the vibration signal in the mechanical operation process. It can realize intelligent mechanical fault diagnosis and ensure the authenticity of signal samples. Moreover, due to the excellent interpretability of the 1DCNN, we can explain the feature extraction mechanism of convolution and the synergistic work ability of the convolution kernel by analyzing convolution kernels and their output results in the time-domain, frequency-domain. What’s more, we propose a novel network parameter-optimization method by matching the features of the convolution kernel with those of the original signal. A large number of experiments proved that, this optimization method improve the diagnostic accuracy and the operational efficiency greatly.

【 授权许可】

Unknown