Enhancing Thermal Comfort in Buildings with Machine Learning-Based Overheating Prediction

The goal of this project is to improve the application of machine learning techniques in the summertime prediction of thermal comfort in residential structures (for both present and future weather situations). Using DesignBuilder’s integrated simulation engine and simulated data, the research creates strong prediction models with Random Forest and XGBoost algorithms. Essential factors like building orientation, window-to-floor ratios, U-values, and operating temperatures were examined using exploratory data analysis, feature engineering, and thorough data preparation. Mean Absolute Error (MAE) and R-squared values were applied for the accurate and effective validation of the models. The findings demonstrate significant potential for early-stage decision making on building designs, for reducing risk of overheating and opening the door to more sustainable and comfortable living spaces. Future research endeavors aim to enlarge the dataset, explore different Machine learning modeling techniques, and enhance the models’ capability to predict and mitigate overheating in different building kinds and climatic conditions.

Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.