Android-Based Real-Time Classification of Electric Fire Short-Circuit Traces Using Lightweight Deep Learning Model
Open Access
Abstract
This paper presents a lightweight deep learning framework for classifying electric fire short-circuit traces to enhance safety and fault diagnosis in electrical energy systems. Accurate differentiation between primary (PSCT) and secondary short-circuit traces (SSCT) is essential for identifying failure origins, yet conventional manual inspection is time-consuming and subjective. To address these limitations, we systematically evaluate three lightweight convolutional neural network (CNN) architectures MobileNetV2, MobileNetV3, and EfficientNet using transfer learning on a domain-specific image dataset. The models are assessed based on accuracy, loss, precision, recall, and F1-score. Experimental results show that EfficientNet achieves the highest classification accuracy, while MobileNetV3 demonstrates the lowest validation loss and superior generalization stability. Based on a performance–efficiency trade-off analysis, MobileNetV3 is deployed on an Android platform using TensorFlow Lite, enabling real-time, offline, and on-device inference. To the best of our knowledge, this is among the first studies to integrate lightweight CNN-based short-circuit trace classification with real-time mobile deployment for on-site energy system fault analysis. By bridging the gap between deep learning and field deployment, the proposed mobile system ensures low-latency execution and provides a rapid, reliable, and portable solution for improving operational safety in electrical fire investigations.Keywords
All Related Versions
Funding Information
- Korea Institute of Energy Technology Evaluation and Planning (RS-2022-KP002707)
This publication has 34 references indexed in Scilit:
- Deep Learning2026
- Characteristics of Short Circuit and Combustion in Electrical Cables Under Various External Heat FluxesFire and Materials, 2025
- A comprehensive survey of deep learning-based lightweight object detection models for edge devicesArtificial Intelligence Review, 2024
- Deep learning on medical image analysisCAAI Transactions on Intelligence Technology, 2024
- Denoising and segmentation in medical image analysis: A comprehensive review on machine learning and deep learning approachesMultimedia Tools and Applications, 2024
- A review of convolutional neural networks in computer visionArtificial Intelligence Review, 2024
- SegNet: A segmented deep learning based Convolutional Neural Network approach for drones wildfire detectionRemote Sensing Applications: Society and Environment, 2024
- Detection Algorithms of Parallel Arc Fault on AC Power Lines Based on Deep Learning TechniquesJournal of Electrical Engineering & Technology, 2022
- Classification and Understanding of Cloud Structures via Satellite Images with EfficientUNetSN Computer Science, 2021
- A survey on deep learning for big dataInformation Fusion, 2018