Hybrid data driven/thermal simulation model for comfort assessment
This addresses the challenge of data scarcity for thermal comfort assessment, but it is incremental as it builds on existing simulation and machine learning techniques.
The paper tackled the problem of predicting thermal comfort by proposing a hybrid method that combines real and simulated data, achieving an F1 score of 0.999 with a random forest model.
Machine learning models improve the speed and quality of physical models. However, they require a large amount of data, which is often difficult and costly to acquire. Predicting thermal comfort, for example, requires a controlled environment, with participants presenting various characteristics (age, gender, ...). This paper proposes a method for hybridizing real data with simulated data for thermal comfort prediction. The simulations are performed using Modelica Language. A benchmarking study is realized to compare different machine learning methods. Obtained results look promising with an F1 score of 0.999 obtained using the random forest model.