Classification of Inter-Turn Insulation Faults in Three-Phase Induction Motors and Optimum Detection Using GJO-GBDT Method

This paper proposes a hybrid technique to optimally detect and classify inter-turn insulation faults in three-phase induction motors (IMs). The proposed strategy is the novel integration of Golden Jackal Optimization and Gradient Boosting Decision Tree and is termed as GJO-GBDT system. Using the proposed simulation platform, the client-characterized framework is described in order to gather the required error training dataset. Using the GJO, the number of features is decreased and the most significant characteristics from the dataset are selected. The motor's health is determined using the GBDT method. The primary objective is to detect faults and improve the motor's life. MATLAB is used to implement the proposed technique, and its performance is compared to the existing approach. Compared to the existing techniques under 100 trials, the proposed strategy hasalower Root Mean Square Error (RMSE) of 12.2, Mean Absolute Percentage Error (MAPE) of 2, Mean Bias Error (MBE) of 1.6, and consumption time of 4.2 min.