Co-scheduling of HVAC Control, EV Charging and Battery Usage for Building Energy Efﬁciency Tianshu Wei Electrical and Computer Engineering University of California, Riverside twei002@ucr.edu Qi Zhu Electrical and Computer Engineering University of California, Riverside qzhu@ece.ucr.edu Mehdi Maasoumy C3 Energy Redwood City, CA mehdi.maasoumy@c3energy.com Abstract—Building stock consumes 40% of primary energy consumption in the United States. Among various types of energy loads in buildings, HVAC (heating, ventilation, and air conditioning) and EV (electric vehicle) charging are two of the most important ones and have distinct characteristics. HVAC system accounts for 50% of the building energy consumption and typically operates throughout the day, while EV charging is an emerging major energy load that is hard to predict and may cause spikes in energy demand. To maximize building energy efﬁciency and grid stability, it is important to address both types of energy loads in a holistic framework. Furthermore, on the supply side, the utilization of multiple energy sources such as grid electricity, solar, wind, and battery storage provides more opportunities for energy efﬁciency, and should be considered together with the scheduling of energy loads. In this paper, we present a novel model predictive control (MPC) based algorithm to co-schedule HVAC control, EV scheduling and battery usage for reducing the total building energy consumption and the peak energy demand, while maintaining the temperature within the comfort zone for building occupants and meeting the deadlines for EV charging. Experiment results demonstrate the effectiveness of our approach under a variety of demand, supply and environment constraints. I. I NTRODUCTION The building stock, including commercial and residential buildings, accounts for nearly 40% of the U.S. primary energy consumption, 40% of the greenhouse gas emissions, and 70% of the electricity use [1]. It is therefore critical to improve building energy efﬁciency for the nation’s energy and environ- ment security. There are a variety of energy loads in buildings, including HVAC (heating, ventilation and air conditioning), lighting, appliances, and emerging loads such as EV (electric vehicle) charging. Among these loads, HVAC system con- sumes around 50% of the total building energy consumption. To satisfy the room temperature and air ﬂow requirements of a building, the HVAC system typically operates throughout the day and its energy demand may change drastically based on the dynamic physical environment (e.g., outside air temperature and sun radiation) and building occupancy activities. If not controlled properly, the HVAC system may cause very high energy demand during peak hours. In addition, EV charging has become a signiﬁcant energy load in many commercial and residential buildings, and will likely continue growing rapidly. Depending on the building types and occupancy behavior, the charatecristics of EV charging demand may vary signiﬁcantly. In residential buildings, EV charging is often conducted at night; while in commercial buildings with installed charging stations for tenants, the EV charging demand concentrates at daytime and may coincide with the HVAC demand to cause spikes during peak hours. The peak demand and total energy consumption may be reduced through intelligent scheduling of HVAC control (by turning on and off air conditioning and changing air ﬂow volume at different times) and EV charging (by charging EVs at different times and with different power levels). In the litera- ture, there has been extensive work on efﬁcient HVAC control, such as the system models and algorithms proposed in [2]– [9]. In [2], a nonlinear model of the overall cooling system is proposed, including chillers, cooling towers and thermal storage banks, and an MPC scheme for minimizing energy consumption is developed. In [3], a system model is proposed that is bilinear in inputs, states and weather parameters, and a form of sequential linear programming (SLP) is developed for solving the control optimization problem. In [6], a building thermal behavior is modeled as RC networks and validated against historical data, and a tracking linear-quadratic regulator (LQR) is proposed for HVAC control. The work in [8] uses the similar building model as in [6], and proposes a set of HVAC control algorithms that address the sensing data inaccuracy using unscented or extended Kalman ﬁlters. There are also approaches proposed for optimizing EV charging [10]–[12] to reduce the peak demand and total energy cost, with utilization of renewable energy sources including solar and wind. Despite these approaches for HVAC control and for EV charging, little work exists for addressing the two types of energy loads in a holistic fashion. We believe that to maximize building energy efﬁciency, it is essential to model the HVAC demand and EV charging demand in an integrated formulation and develop methods for co-scheduling the two types of demands, especially in the cases where HVAC demand and EV charging demand coincide. On the energy supply side, utilizing multiple energy sources such as grid electricity, battery storage, and renewable sources provides more opportunities for reducing the peak demand and total energy cost. The demand side control (e.g., HVAC control and EV charging scheduling) depends on the availability of the various energy sources, and the supply side energy sources scheduling (i.e., deciding which source to use and for how much at different times) requires the knowledge of the demand. Therefore, it is important to co-schedule the energy demands with supply sources for maximizing building energy efﬁciency. In our earlier work [13], we propose an algorithm for co-scheduling HVAC control and battery usage and demonstrate its effectiveness in reducing energy cost.