A New Deep Learning Model for Fault Diagnosis with Good Anti-Noise and Domain Adaptation Ability on Raw Vibration Signals

TLDR

Intelligent fault diagnosis techniques have replaced time‑consuming and unreliable human analysis, and deep learning models enhance accuracy through multilayer nonlinear mapping. This paper proposes the WDCNN model to improve the low accuracy of conventional CNN‑based fault diagnosis. The method uses raw vibration signals with data augmentation, wide first‑layer kernels to extract features and suppress high‑frequency noise, small kernels in subsequent layers for nonlinear mapping, and AdaBN for domain adaptation. WDCNN achieves 100 % classification accuracy on normal signals and outperforms the state‑of‑the‑art frequency‑feature DNN under varying loads and noisy conditions.

Abstract

Intelligent fault diagnosis techniques have replaced time-consuming and unreliable human analysis, increasing the efficiency of fault diagnosis. Deep learning models can improve the accuracy of intelligent fault diagnosis with the help of their multilayer nonlinear mapping ability. This paper proposes a novel method named Deep Convolutional Neural Networks with Wide First-layer Kernels (WDCNN). The proposed method uses raw vibration signals as input (data augmentation is used to generate more inputs), and uses the wide kernels in the first convolutional layer for extracting features and suppressing high frequency noise. Small convolutional kernels in the preceding layers are used for multilayer nonlinear mapping. AdaBN is implemented to improve the domain adaptation ability of the model. The proposed model addresses the problem that currently, the accuracy of CNN applied to fault diagnosis is not very high. WDCNN can not only achieve 100% classification accuracy on normal signals, but also outperform the state-of-the-art DNN model which is based on frequency features under different working load and noisy environment conditions.