Building Energy Consumption Control Based on BIM and Machine Learning

To solve the problem of building energy consumption (EC) and promote the development of green buildings in China, this paper simulates and predicts EC of building based on BIM and machine learning (ML), so as to provide optimization strategies for building energy conservation. Firstly, this paper uses designbuilder (DB) to simulate the building energy consumption of the design test. Then Support vector machine (SVM) is introduced to fit the functional relationship between energy consumption influencing factors and EC of building, and a EC prediction model is established. Finally, through range analysis, the importance ranking and optimal scheme of six energy consumption influencing factors are obtained. Taking a teaching building in Chengdu as an example, the accuracy of the prediction model is verified, which provides a theoretical basis for the optimal design of building energy conservation.