Integrating AIoT and Machine Learning for Enhanced Transformer Overload Power Protection in Sustainable Power Systems
摘要
This research introduces an advanced Transformer Overload Protection System, integrating Artificial Intelligence of Things (AIoT) and Machine Learning (ML) technologies. Employing Simulink simulations and a comprehensive dataset reflecting real-world scenarios, the system demonstrated commendable adaptability, as evidenced by current variations, system responses, and anomaly detections. Evaluation metrics underscored its high accuracy, precision, recall, F1 score, sensitivity, and swift response times. The methodology encompassed a mathematical model, AIoT integration, and ML algorithm implementation, establishing a robust framework. This study not only validates the system’s efficacy but also sets the stage for future advancements in transformer protection mechanisms, emphasizing the potential for real-world applications and continuous improvements.