Presentation Information
[7p-N221-11]Seismic Vibration Monitoring Utilizing Optical Fiber Sensors and YOLO-v7
〇(DC)Pradeep Kumar1, Ruchi Dahiya2, KM Priyanka2 (1.Department of Electro-Optical Engineering National Taipei University of Technology, Taipei, Taiwan., 2.Department of Electronics and Communication Engineering Netaji Subhas University of Technology, Delhi, India.)
Keywords:
optical fiber sensor,machine learning,vibration signals
This study introduces a novel approach to seismic vibration monitoring by integrating Fiber Bragg Grating (FBG) sensors with a YOLOv7-based deep learning model. FBG sensors are known for their high sensitivity, immunity to electromagnetic interference, and suitability for distributed sensing. A single FBG sensor, modeled on the CP-8000 system, was used to simulate seismic vibrations at four intensity levels, with corresponding wavelength shifts ranging from 1549.2 to 1550.6 nm. These synthetic signals were used to train a customized YOLOv7 model, repurposed from object detection to classify seismic vibration patterns in real time. The trained model demonstrated effective performance in distinguishing varying levels of vibration intensity based on time-series data. Results showed clear differentiation among “No Seismic” and three increasing levels of vibration, validating the feasibility of using YOLOv7 for time-series classification. This hybrid FBG–YOLOv7 system offers a scalable, low-cost, and real-time solution for intelligent structural health monitoring. Future work will focus on multi-sensor deployment and real-world testing to further enhance system robustness and practical applicability.