Evaluating the Performance of Random Forest Regression in Predicting Electric Motor Temperature

Objective: This research applies Random Forest Regression(RFR) as a non-linear technique to forecast temperature based on multiple input parameters, including motor load, speed, and ambient temperature. The process begins with selecting features through correlation analysis to pinpoint the most significant factors for predicting motor temperature. A Random Forest model is trained on historical data, and compared to other regression models includes linear regression, KNN regression for predictive accuracy. To improve the model's efficiency, hyperparameter optimization is performed by fine-tuning parameters such as the tree count and depth limit. Ultimately, the study evaluates the effectiveness and efficiency of Random Forest Regression, comparing it with alternative models and emphasizing its potential for real-time temperature prediction in industrial motor applications.