LSR-Net: A Lightweight and Strong Robustness Network for Bearing Fault Diagnosis in Noise Environment

TL;DR

This work targets reliable bearing fault diagnosis in noisy environments by introducing LSR-Net, a lightweight network that combines a denoising/feature-enhancement DFEM with a Convolution-based Efficient Shuffle (CES) block to achieve high accuracy and low latency on edge devices. DFEM denoises signals with 1D convolutions and enriches representations through an activation ensemble, while AP-DM further improves robustness under strong noise. The CES design uses channel splits and attention-guided group convolutions to reduce MAC/FLOPs without sacrificing performance, enabling real-time inference on devices like Jetson Nano. Evaluations on the Padderborn dataset with Gaussian and Laplace noise show LSR-Net achieving near-perfect accuracy on clean signals and strong robustness under noise, outperforming benchmarks in both accuracy and efficiency. The proposed approach offers practical impact for industrial condition monitoring by delivering accurate, fast fault diagnosis in noisy, real-world operating conditions.

Abstract

Rotating bearings play an important role in modern industries, but have a high probability of occurrence of defects because they operate at high speed, high load, and poor operating environments. Therefore, if a delay time occurs when a bearing is diagnosed with a defect, this may cause economic loss and loss of life. Moreover, since the vibration sensor from which the signal is collected is highly affected by the operating environment and surrounding noise, accurate defect diagnosis in a noisy environment is also important. In this paper, we propose a lightweight and strong robustness network (LSR-Net) that is accurate in a noisy environment and enables real-time fault diagnosis. To this end, first, a denoising and feature enhancement module (DFEM) was designed to create a 3-channel 2D matrix by giving several nonlinearity to the feature-map that passed through the denoising module (DM) block composed of convolution-based denoising (CD) blocks. Moreover, adaptive pruning was applied to DM to improve denoising ability when the power of noise is strong. Second, for lightweight model design, a convolution-based efficiency shuffle (CES) block was designed using group convolution (GConv), group pointwise convolution (GPConv) and channel split that can design the model while maintaining low parameters. In addition, the trade-off between the accuracy and model computational complexity that can occur due to the lightweight design of the model was supplemented using attention mechanisms and channel shuffle. In order to verify the defect diagnosis performance of the proposed model, performance verification was conducted in a noisy environment using a vibration signal. As a result, it was confirmed that the proposed model had the best anti-noise ability compared to the benchmark models, and the computational complexity of the model was also the lowest.

Figures (9)

• Figure 1: Proposed lightweight network framework
• Figure 2: Denoising module (DM).
• Figure 3: Adaptive pruning-DM (AP-DM).
• Figure 4: proposed Conv-based efficient shuffle (CES) block
• Figure 5: (a) Spatial attention module (SAM) and (b) Channel attention module (CAM).