Hybrid CNN-BYOL Approach for Fault Detection in Induction Motors Using Thermal Images

TL;DR

This paper tackles the challenge of fault detection in induction motors using thermal imagery, where labeled data can be scarce. It proposes a CNN-BYOL hybrid, culminating in a lightweight BYOL-IMNet that combines four BYOL-inspired blocks with a compact CNN backbone to enable real-time inference. By pretraining with BYOL on unlabeled thermal images and fine-tuning on labeled data, the approach achieves state-of-the-art performance, exemplified by 99.89% test accuracy and 5.7 ms per-image inference, with strong generalization evidenced by 5-fold cross-validation (average ~99.11% accuracy and ~99.81% AUC). The work demonstrates the practicality of self-supervised, hardware-efficient fault detection for online industrial motor monitoring, reducing downtime and energy waste while enabling scalable deployment.

Abstract

Induction motors (IMs) are indispensable in industrial and daily life, but they are susceptible to various faults that can lead to overheating, wasted energy consumption, and service failure. Early detection of faults is essential to protect the motor and prolong its lifespan. This paper presents a hybrid method that integrates BYOL with CNNs for classifying thermal images of induction motors for fault detection. The thermal dataset used in this work includes different operating states of the motor, such as normal operation, overload, and faults. We employed multiple deep learning (DL) models for the BYOL technique, ranging from popular architectures such as ResNet-50, DenseNet-121, DenseNet-169, EfficientNetB0, VGG16, and MobileNetV2. Additionally, we introduced a new high-performance yet lightweight CNN model named BYOL-IMNet, which comprises four custom-designed blocks tailored for fault classification in thermal images. Our experimental results demonstrate that the proposed BYOL-IMNet achieves 99.89\% test accuracy and an inference time of 5.7 ms per image, outperforming state-of-the-art models. This study highlights the promising performance of the CNN-BYOL hybrid method in enhancing accuracy for detecting faults in induction motors, offering a robust methodology for online monitoring in industrial settings.

Figures (4)

• Figure 1: Steps of Our Methodology.
• Figure 2: Architecture of our proposed model BYOL-IMNet.
• Figure 3: CNN-BYOL Hybrid Framework.
• Figure :