Development of a domestic hot water demand prediction model based on a bottom-up approach for residential energy management systems

An energy management system (EMS) that controls energy equipment in residential dwellings contributes to energy conservation and cost reduction. A hot energy demand prediction method is necessary for the EMS to provide the best possible performance. A prediction method using a bottom-up approach based on end-use analysis with detailed energy consumption measurement was developed. One of the advantages of the proposed prediction method is that it provides good accuracy without needing large computational resources. Since the method is based on end-use analysis, the EMS can revise the prediction and controls when necessary. The performance of the developed prediction method was evaluated by comparing it with results produced by support vector regression. An operational simulation was performed to evaluate performance in a more practical environment. The simulation indicated that the proposed method showed the best performance.