Data-Driven load management of stand-alone residential buildings including renewable resources, energy storage system, and electric vehicle

Demand side management (DSM) plays a critical role in dealing with power system reliability as one of the main challenges of generating energy in stand-alone residential buildings. This paper introduces a new effective approach of DSM based on predetermined hourly generation and time-varying tariffs to enhance the reliability and quality of a stand-alone energy system. This research focus on two main novelties: 1) analyzing data via clustering method to improve the efficiency and accuracy of DSM, and 2) a state-of-the-art modeling framework DSM for a stand-alone residential building. Looking at first novelty, the well-known linkage-ward clustering method is used for defining time-varying tariffs and in the second novelty the mixed integer linear programming (MILP) formulates the power system components, load management and objective function considering user's priority and convenience to improve the reliability of a stand-alone energy system. The proposed method outperforms DSM in hybrid power systems consisting of photovoltaic, wind turbines, and energy storage systems. To illustrate the efficiency of this strategy, the proposed DSM is applied to a stand-alone building located in Sistan and Baluchistan province of Iran as an experimental case. Considering four different case studies that include a battery energy storage system and electric vehicle, the results reveal that the power system reliability and energy supplying quality are significantly improved by utilizing the aforementioned DSM.