The heat load modelling for an air-conditioned room using Buckingham-Pi theorem

Abstract The objective of this study is to develop and compare two simplified heat load models for air-conditioned rooms. The heat load models are intended for use in a demand response (DR) program in future work, to solve critical electricity peaks from air-conditioners used. Buckingham-Pi theorem was applied to develop the heat load model in this study. Moreover, another heat load model using sensitivity analysis and multiple linear regression was developed in this study. Then, the result of the heat load model using Buckingham-Pi theorem was compared to the heat load model using sensitivity analysis and multiple linear regression to identify the advantage of both heat load models. In this study, heat load model using the Buckingham-Pi theorem was successfully developed. A heat load model with three independent variables was recommended for heat load prediction. The results showed that the heat load model gave a good prediction of the heat load with an RMSE of 4.82%. However, the RMSE of predicting the heat load model using sensitivity analysis and multiple linear regression is 0.55% using a model with ten independent variables. The comparison shows that the heat load model using Buckingham-Pi theorem has a good agreement and simplifies the model better than using sensitivity analysis and multiple linear regression.