Hierarchical microgrid energy management in an o ﬃ ce building

(cid:129) A hierarchical microgrid energy management method in an o ﬃ ce building is proposed. (cid:129) An o ﬃ ce building is modelled as a virtual energy storage system (VESS). (cid:129) A V2B control strategy is developed to dispatch the EVs as a ﬂ exible resource. (cid:129) The VESS and the EVs are coordinated and dispatched in two di ﬀ erent time scales. A two-stage hierarchical Microgrid energy management method in an o ﬃ ce building is proposed, which con- siders uncertainties from renewable generation, electric load demand, outdoor temperature and solar radiation. In stage 1, a day-ahead optimal economic dispatch method is proposed to minimize the daily Microgrid oper- ating cost, with the virtual energy storage system being dispatched as a ﬂ exible resource. In stage 2, a two-layer intra-hour adjustment methodology is proposed to smooth the power exchanges at the point of common coupling by coordinating the virtual energy storage system and the electric vehicles at two di ﬀ erent time scales. A Vehicle-to-Building control strategy was developed to dispatch the electric vehicles as a ﬂ exible resource. Numerical studies demonstrated that the proposed method is able to reduce the daily operating cost at the day-ahead dispatch stage and smooth the ﬂ uctuations of the electric power exchanges at the intra-hour adjustment stage.


Introduction
Increasing attention is being paid to technologies in renewable energy and energy efficiency improvement due to the rapid growth of global energy use and environmental deterioration [1,2]. According to the International Energy Agency, energy consumption of buildings occupies about 32% of the global energy use and they are responsible for about 30% of the total end-use and energy-related CO 2 emissions [3]. As a result, a number of regions and countries have taken specific initiatives to facilitate a high penetration of renewable generation and the low energy consumption technologies in their building sectors, including the European Union [4], the United States [5] and China [6][7][8][9]. In China, the building sector currently accounts for 27.6% of the total energy use and it is estimated to reach 35% by 2020 [6,7]. The Chinese government has paid attention to the retrofits and renovations of the existing buildings, and provided financial support for the energy management in large public buildings [8,9]. Therefore, as the major power consumers at demand side, buildings represent a great potential contributor for reducing the energy consumption and relieving power imbalance of the electric grid.
Aiming to facilitate a high penetration of renewable generation and the low energy consumption technologies at the demand side, there is significant development of low-carbon buildings integrated with renewable generation [10]. However, renewable generation is usually intermittent, uncertain and uncontrollable, which induces power mismatches between power demand and supply for low-carbon buildings [10].
Microgrids provide an opportunity and a desirable infrastructure for facilitating integration of intermittent renewable generation in lowcarbon buildings [11]. Microgrids can increase the penetration of intermittent renewable generation and provide an economical energy supply for the low-carbon buildings by utilizing advanced energy management technologies and intelligent communication technologies [12]. The fluctuations of the electric power exchanges at the point of common coupling (PCC) of a Microgrid can also be smoothed by coordinating and optimizing the operation of various energy sources and energy loads of the Microgrid [13,14]. Therefore, Microgrid energy management in buildings is attracting more and more attentions in recent years.
Studies have been carried out to investigate the Microgrid energy management methods in a commercial building. The operational performance of a Microgrid in a building in Hong Kong was studied considering operating cost and environmental constraints [10]. A multiobjective dispatch model was proposed in [15] to minimize the daily operating cost and the pollutants emission. An electric chiller (EC) was used as the cooling system of a building in [16] and the electric power consumption of the EC was dispatched using a nonlinear programming method for cost saving. In [17], the electric power consumption of an EC was dispatched in the dynamic economic dispatch process with the discrete EC operating constraints. As a flexible resource, the integration of electric vehicles (EVs) to the building is creating new opportunities for the Microgrid energy management [18,19]. EVs have a certain flexibility to shift their electricity consumption in time and facilitate the integration of intermittent renewable generation [20,21]. A Microgrid energy management method in an office building was proposed in [22] to reduce the impact of EV charging on the external grid with different charging strategies being considered. In [23], a Vehicle-to-Building (V2B) operational model was proposed for the EVs to reduce the total energy cost of a Microgrid in a building.
The existing research work has made good contributions to the Microgrid energy management in a building. However, the flexibility of the building with heat inertia hasn't been fully explored in the Microgrid energy management. As the major power consumer of the Microgrid, a building can perform as a distributed thermal storage to Parameters and constants C ph , C se real-time electricity purchasing/selling prices ($/kWh) P el electric load of the Microgrid (kW) P PV electric power generated by photovoltaic system (kW) U wall , U win heat transfer coefficient of the wall/window of the building [W/(m 2 ·K)] F wall,j the area of the total wall surface at the jth wall orientation (m 2 ) F win,j the area of the total window surface at the jth wall orientation (m 2 ); It is assumed that the total window surfaces are distributed in the south, west, north and east orientations of the walls in a building uniformly T out outdoor temperature (°C) τ win the glass transmission coefficient of the windows SC the shading coefficient of the windows α w absorbance coefficient of the external surface of the wall Qi n internal heat gains from people, appliances and lighting (kW) ρ, C, V the density (kg/m 3 ), specific heat capacity [J/(kg·°C)] and volume of the air (m 3 ) in the building R se j , the external surface heat resistance for convection and radiation of the external wall j (m 2 ·K/W) I T j , the total solar radiation on the walls/windows surface at the j-wall orientation (kW/m 2 ) EER EC energy efficiency ratio of the EC ρ PV maintenance cost of the photovoltaic system ($/kWh) ρ EC maintenance cost of the EC ($/kWh) t in, j , t out, j plug-in/plug-out time of EV j , the rated charging and discharging power of EV j at time t (kW) the lower and upper power output limits of the EV j at time t (kW) SOC j,d,home the expected SOC when EV j leaves home C e the EV energy consumption per kilometer (kWh/km) D EV's daily travelling distance (km) Cap the battery capacity of M1 EV (kW) Cap min the minimum battery capacity of M1 EV (kW) Cap max the maximum battery capacity of M1 EV (kW) η j,c , η j,d the charging and discharging efficiency of EV j SOC j V the upper limit of the SOC of EV j SOC j V the lower limit of the SOC of EV j Q EC the upper limit of the cooling power output of the EC (kW) T in , T in the upper and lower limits of the indoor temperature setpoints of the building (°C) P ex , P ex the upper and lower limits of electric power exchange with the external grid of the building (kW) Variables P ex electric power exchange with the external grid (kW) P EC electric power consumption by the EC (kW) QĖ C cooling power generated by the EC (kW) Qẇ all heat transfer through the external walls (kW) Qẇ in heat transfer across the windows (kW) Qṡ w heat contribution due to the solar radiation on the opaque surface of the external walls (kW) Qṡ g the whole solar radiation transmitted across the windows (kW) Qċ l building , cooling load of the Microgrid with VESS being dispatched (kW) ′ Q̇c l building , cooling load of the Microgrid without VESS being dispatched (kW) SOC in, j the initial SOC of EV j SOC j,d,office the minimum expected SOC at t out of EV j D h-w the EV's travelling distance from home to work (km) P j t V , the real-time power output of EV j at time t (kW), with negative (positive) value representing the charging (discharging) process P t V2B, the real-time power output of the V2B system at time t (kW) P t tar V2B, the target power output of the V2B system at time t (kW) P t upper V2B, the upper boundary of the power output of the V2B system at time t (kW) P t lower V2B, the lower boundary of the power output of the V2B system at time t (kW) T in indoor temperature set-point (°C)