loading page

Machine Learning Based Coordination for IDMT Overcurrent and Earth Fault Relays
  • Sultan Uddin Khan ,
  • Md Shazzad Hossain ,
  • Ibrahim Sultan
Sultan Uddin Khan
North Carolina A&T State University, North Carolina A&T State University

Corresponding Author:[email protected]

Author Profile
Md Shazzad Hossain
Author Profile
Ibrahim Sultan
Author Profile


As machine learning continues to transform numerous industries, its application to smart grid systems has emerged as an intriguing area of investigation. Coordination of Inverse Definite Minimum Time (IDMT) overcurrent and earth fault relays is an essential aspect of smart grid operation, as it is crucial for ensuring grid stability and safety. In this study, we evaluate the relay coordination efficacy of a number of widely used machine learning models, including Linear Regression, Decision Tree, Random Forest, Support Vector Regression, and Gradient Boosting. Our research shows the gradient boosting model achieved the highest R2-score and the lowest Mean Square Error (MSE) and Mean Average Error (MAE) values in our simulation scenario demonstrating the superior efficacy of the gradient boosting model in accurately performing relay coordination, thereby providing a promising alternative to the conventional, calculation-intensive method.