Prosumer-centric demand side management for minimizing electricity bills in a DC residential PV-battery system: An Australian household case study

Abstract

Technological advancements lead to the development of a home energy management system (HEMS) to perform demand side management (DSM) in residential houses. The current study proposes a DSM strategy to minimize electricity bills in a DC residential house considering the predictions of electrical load consumption and solar photovoltaic (PV) power based on long short-term memory networks. A simulation study compares the proposed DSM strategy against four alternate scenarios, the result shows that the proposed DSM strategy has the lowest annual energy cost increment (5.81%) compared to the benchmark scenario where it is assumed to have the ideal predictions of the load consumption and solar PV power. It also shows that the predictions of the load consumption and solar PV power are essential for economically sensible DSM, however further analysis reveals that the prediction errors do not correlate with the differences in daily energy costs (or daily peak powers) resulting from the proposed DSM strategy and benchmark scenario. Finally, the proposed DSM strategy is experimentally validated on a small-scale hardware-in-the-loop test platform to demonstrate the feasibility of its practical implementation in a HEMS.