Building Heating and Cooling Load Prediction Using Ensemble Machine Learning Model.

Area of Science:

• Building energy analysis
• Sustainable architecture
• Smart city infrastructure

Background:

• Building energy consumption is a critical factor in sustainable development and smart city initiatives.
• Data-driven methods, leveraging IoT sensor data, are essential for accurate energy consumption modeling and forecasting.
• Current prediction models often lack accuracy and fail to comprehensively analyze building attributes.

Purpose of the Study:

• To develop a high-performance data-driven ensemble model for predicting building heating and cooling loads.
• To investigate the impact of various building features on energy load predictions.
• To improve the accuracy of energy consumption forecasting for sustainable building design.

Main Methods:

• A data-driven approach utilizing an ensemble model for energy load prediction.
• Investigation of building features including glazing area, orientation, height, relative compactness, roof area, surface area, and wall area.
• Experimental validation of the proposed model's predictive capabilities.

Main Results:

• The proposed ensemble model achieved R2 scores of 0.999 for heating load and 0.997 for cooling load prediction.
• Relative compactness, surface area, and wall area were identified as significant features influencing heating and cooling loads.
• The model demonstrated superior performance compared to existing state-of-the-art methods.

Conclusions:

• The developed ensemble model offers highly accurate predictions for building heating and cooling loads.
• Accurate load prediction aids engineers in designing more energy-efficient buildings for smart homes.
• This research contributes to advancing sustainable building design through precise energy consumption forecasting.