This study proposes a novel structural health monitoring (SHM) approach for detecting and localizing damage in composite materials. The developed system integrates spatial coordinate-based localization with ply-level damage identification. Sensor data are collected from the composite structure and processed through an ensemble of machine learning models designed to detect and localize damage with high precision. Three numerical case studies are presented to evaluate and compare the performance of different machine learning models. Additionally, a two-stage framework is introduced to improve robustness, allowing different models to be trained on distinct datasets and specialize in damage detection at various ply levels. The proposed SHM system demonstrates significant potential for real-time damage monitoring and localization, accurate remaining service life prediction, and integration with emerging smart material technologies..