Optimal Demand Response Bidding Scheduling of a Residential Customer Considering Load Resource Dynamics based on Tenant Behavior

Abstract

This study proposes optimal day-ahead economic demand response (DR) bidding strategies and distributed energy resource (DER) management in a residential building considering external environment, dynamic characteristics of load resources, and tenant behavior. First, this study introduces the structure of the proposed home-to-grid (H2G) operation model. The photovoltaic generation system (PV) is applied to reduce the electricity demand in the building. In addition, customer-owned electric vehicle (EV) battery is participated in DR bids through optimal charging and discharging. Among the load demand of the building, air conditioner (AC) and luminaire (LM) loads are included in DR bidding scheduling. More specifically, weather information is analyzed by autoregressive integrated moving average model with exogenous inputs (ARIMAX) for PV generation forecasting. Sub-metering data of a residential customer is investigated by hidden Markov model (HMM) to find out tenant presence profile for EV and load demand scheduling. The problem is solved using mixed integer linear programming (MILP) to minimize the operation cost. DR bidding potential is calculated through the optimal resource scheduling on a certain day of winter season, and based on the results, DR aggregation amount in the experimental area is estimated.