As an integral part of the global “dual-carbon” strategy, enhancing energy efficiency and reducing emissions within the building sector has become paramount. My research focuses on unlocking the immense potential of electric vehicle (EV) batteries as distributed mobile energy storage units integrated with building energy systems. Specifically, I investigate the Vehicle-to-Building (V2B) technology, where electric vehicle batteries serve not only transportation needs but also actively participate in building energy management by charging during low-cost periods and strategically discharging to power the building or reduce grid dependence during high-cost periods. This study rigorously analyzes the application value and development potential of V2B technology under diverse and evolving electricity price policies using a sophisticated optimization framework. The core objective is to quantify the economic benefits achievable through optimized bidirectional energy flow between buildings and fleets of electric vehicles, considering real-world constraints and future cost and policy trajectories. My findings provide critical insights for accelerating the adoption of V2B as a cornerstone technology for zero-carbon buildings and a more resilient power grid.
1. Introduction: The V2B Imperative
The push towards carbon neutrality necessitates radical transformations in how buildings consume and manage energy. Key trends include the significant increase in the penetration of renewable energy sources (RES) within building systems and the accelerated electrification of building services. However, the inherent intermittency and uncertainty of RES, such as solar photovoltaics (PV), pose substantial challenges for matching real-time energy supply with building demand. Concurrently, the rapid global proliferation of electric vehicles presents a transformative opportunity. Each electric vehicle embodies a substantial mobile battery storage unit. When aggregated, fleets of electric vehicles parked at commercial or residential buildings represent a vast, largely untapped resource for energy flexibility – termed electric vehicle-to-building (V2B) technology.
V2B leverages bidirectional charging infrastructure, allowing energy stored in electric vehicle batteries to be discharged back into the building’s electrical system when needed. This capability enables several critical functions:
- Peak Shaving: Reducing the building’s peak grid electricity demand during high-price periods by utilizing stored EV energy.
- Valley Filling: Increasing grid electricity consumption during low-price, low-demand periods to charge EV batteries.
- RES Integration Enhancement: Storing surplus local RES generation (e.g., midday solar PV peak) in EV batteries for later use within the building when RES generation is low or demand is high, thereby increasing self-consumption and reducing grid reliance.
- Cost Reduction: Capitalizing on electricity price arbitrage opportunities inherent in time-of-use (TOU) tariffs by charging EVs when electricity is cheap and discharging (or reducing building grid draw) when electricity is expensive.
While prior research has explored EV scheduling within microgrids and building clusters, demonstrating benefits like improved economics and RES utilization [17, 18], integrated PV-V2B systems for peak load reduction [19], and even incorporating carbon trading mechanisms [20], a significant gap remains. Most existing studies simulate V2B based on the static electricity price structure of a specific region. In reality, electricity pricing policies vary dramatically across different provinces and cities within large economies like China. Furthermore, the electricity landscape is dynamic, influenced by the rapid evolution of power systems (especially increased RES integration) and advancing electric vehicle and charging technologies. Future electricity prices and EV characteristics are certain to undergo significant changes, rendering conclusions drawn from single, static price scenarios potentially obsolete.
Therefore, my research addresses this critical gap. I employ a comprehensive Mixed-Integer Linear Programming (MILP) optimization model to rigorously evaluate the economic and operational performance of V2B technology integrated with a typical office building. This analysis is conducted under both current representative electricity price policies and plausible future scenarios characterized by increased price volatility and reduced technology costs. My goal is to provide robust, forward-looking insights into the economic viability and strategic deployment of V2B technology for achieving zero-carbon building objectives.
2. Building Microgrid System Architecture
The foundation of my study is a modeled office building microgrid system designed to facilitate and analyze V2B interactions. The system architecture, conceptualized in this research, comprises three primary components integrated under a central Energy Management System (EMS):
- The Office Building: Represents the primary electricity consumer, with demand profiles generated through detailed simulation.
- Photovoltaic (PV) Generation System: Installed on the building rooftop, providing on-site renewable energy.
- V2B Charging Infrastructure: Comprising bidirectional charging piles that enable controlled energy flow between the grid, the building, and connected electric vehicle batteries.
Electricity supply originates from two sources: the main power grid and the on-site PV system. The system operates under a fundamental principle: the building load and electric vehicle charging demands are prioritized to be met by the instantaneous PV generation whenever possible. Only when PV output is insufficient to meet the combined building and EV charging load does the system draw power from the grid. Crucially, the system adopts a “no export to grid” policy. Any surplus PV generation that cannot be immediately consumed by the building or used for charging electric vehicles is considered dissipated energy (Ech(t)Ech(t)) – it is not fed back into the utility grid. This policy reflects common constraints or economic disincentives for small-scale distributed generation export in many regions.
The electric vehicles connect to the system via the V2B charging piles. These piles are the physical and control interface enabling bidirectional energy transfer:
- Charging: EVs can charge from either the grid (during low-price periods) or directly from surplus PV generation.
- Discharging: When economically advantageous or necessary for building supply, EVs can discharge stored energy back into the building microgrid to offset grid consumption during high-price periods or supplement supply when PV is unavailable.
The EMS acts as the central intelligence. It receives inputs including:
- Real-time and forecasted building electricity demand (Eload(t)Eload(t))
- Real-time and forecasted PV generation (EPV(t)EPV(t))
- Electricity prices (Pbuy(t)Pbuy(t))
- EV connection status, arrival/departure times, initial State of Charge (SOC), battery capacity, and user constraints (e.g., required departure SOC)
- Charging pile status and capabilities
Based on these inputs and the optimization objectives (primarily minimizing total daily operating cost), the EMS computes and executes the optimal charging and discharging schedules (EEV,cha(i,t)EEV,cha(i,t), EEV,dis(i,t)EEV,dis(i,t)) for each connected electric vehicle, as well as the optimal grid purchase schedule (Ebuy(t)Ebuy(t)). This centralized control maximizes the system-wide benefits of V2B integration.
3. Electric Vehicle Modeling for V2B
Accurately representing the behavior and characteristics of the electric vehicle fleet is essential for realistic V2B simulation and optimization. My model captures four key stochastic aspects of EV usage patterns relevant to building interaction:
3.1. Time Model (Arrival & Departure)
The availability of electric vehicles at the building for V2B participation is dictated by their arrival and departure times. Based on statistical analysis from authoritative sources like the China New Energy Vehicle Big Data Annual Report 2024 [21], I model the arrival and departure times of electric vehicles using Normal distributions. This reflects the typical commuting patterns observed in metropolitan areas.
The probability density function for arrival/departure time tt is given by:f(t)=1σ2πexp(−(t−μ)22σ2)f(t)=σ2π1exp(−2σ2(t−μ)2)
where:
- μμ represents the mean arrival or departure time. For a typical office building scenario, arrival mean (μarrμarr) is set to 8:00 AM (08:00) and departure mean (μdepμdep) is set to 5:00 PM (17:00).
- σσ represents the standard deviation of arrival/departure times, set to 0.41 hours (approximately 25 minutes) based on observed data variability [21, 22].
3.2. Initial State of Charge (SOC) Model
The energy available in an electric vehicle battery upon arrival significantly impacts its potential for discharging into the building or its need for charging. To capture realistic heterogeneity among users, I categorize EV users into two groups based on their commuting distance and corresponding energy consumption:
- Long Commute Users: Representing users who live farther from the office. They typically start their commute with a higher initial SOC but also consume more energy during the trip to work. Their initial SOC upon arrival (SOCarr,longSOCarr,long) is modeled using a Normal distribution:SOCarr,long∼N(μ=80%,σ=5%)SOCarr,long∼N(μ=80%,σ=5%)
- Short Commute Users: Representing users who live closer. They typically arrive with a lower initial SOC and have less immediate charging demand solely for transportation needs. Their initial SOC upon arrival (SOCarr,shortSOCarr,short) is modeled using a Normal distribution:SOCarr,short∼N(μ=70%,σ=5%)SOCarr,short∼N(μ=70%,σ=5%)
This distinction acknowledges that users with longer commutes arrive with more usable energy potentially available for V2B discharge, while those with shorter commutes might arrive needing more charging [22].
3.3. Battery Capacity Model
The energy storage capacity of individual electric vehicles directly influences the magnitude of potential V2B energy exchange. Surveying popular EV models available in the market, I observed that battery capacities predominantly fall within the 60-80 kWh range. To reflect this diversity, the battery capacity (I(i)I(i)) for each simulated electric vehicle is modeled using a Uniform distribution:I(i)∼U(60,80)kWhI(i)∼U(60,80)kWh
This ensures the model considers a realistic mix of EV sizes and capabilities within the fleet interacting with the building.
4. V2B Optimization Scheduling Model
The core of my analysis is a Mixed-Integer Linear Programming (MILP) model formulated to determine the optimal charging and discharging schedule for the electric vehicle fleet integrated with the office building microgrid, minimizing the total daily operational cost. The model is implemented and solved using the CPLEX solver within MATLAB, with a temporal resolution of 1 hour (Δt=1Δt=1 h). The optimization process integrates building load, PV generation, EV characteristics, and dynamic electricity prices to compute optimal setpoints for grid purchases, EV charging/discharging, and potential PV dissipation over a 24-hour period.
4.1. Objective Function
The primary objective is to minimize the total daily operating cost (CdCd) of the integrated building-EV system:J=min(Cd)J=min(Cd)Cd=Cd,B,ele+Cd,EV,bat+Cd,EV,eleCd=Cd,B,ele+Cd,EV,bat+Cd,EV,ele
where:
- JJ: Objective function value (Total Daily Cost, CNY).
- CdCd: Total daily operating cost (CNY).
- Cd,B,eleCd,B,ele: Daily electricity cost for the building’s grid consumption (CNY).
- Cd,EV,batCd,EV,bat: Daily battery degradation cost compensation for EV users participating in V2B (CNY).
- Cd,EV,eleCd,EV,ele: Daily net electricity cost/revenue associated with EV charging and discharging (CNY).
4.1.1. Building Electricity Cost (Cd,B,eleCd,B,ele)
This cost component depends on the amount of electricity purchased from the grid and the prevailing time-of-use (TOU) electricity price at each time step:Cd,B,ele=∑t=124Pbuy(t)⋅Ebuy(t)Cd,B,ele=t=1∑24Pbuy(t)⋅Ebuy(t)
where:
- Pbuy(t)Pbuy(t): Electricity purchase price from the grid at time tt (CNY/kWh).
- Ebuy(t)Ebuy(t): Amount of electricity purchased from the grid at time tt (kWh).
4.1.2. EV Battery Degradation Cost (Cd,EV,batCd,EV,bat)
Participating in V2B involves additional charge-discharge cycles compared to normal charging-only operation, accelerating battery wear. To compensate EV users for this additional degradation and incentivize V2B participation, my model explicitly includes a battery degradation cost:Cd,EV,bat=∑i=1NPEV,bat⋅I(i)Rrated(i)⋅R(i)Cd,EV,bat=i=1∑NRrated(i)PEV,bat⋅I(i)⋅R(i)R(i)=∑t=tarr(i)tdep(i)(EEV,cha(i,t)+EEV,dis(i,t))2×D×I(i)−(SOCEV(i,tdep)−SOCEV(i,tarr))R(i)=2×D×I(i)∑t=tarr(i)tdep(i)(EEV,cha(i,t)+EEV,dis(i,t))−(SOCEV(i,tdep)−SOCEV(i,tarr))
where:
- NN: Maximum number of electric vehicles considered.
- PEV,batPEV,bat: Electric vehicle battery price (CNY/kWh). Based on market data, this is set to 958 CNY/kWh.
- I(i)I(i): Battery capacity of the ithith EV (kWh).
- Rrated(i)Rrated(i): Rated cycle life of the EV battery (cycles). Assuming an 80% Depth of Discharge (DoD), this is set to 2000 cycles [23].
- R(i)R(i): Additional battery cycle degradation incurred by the ithith EV due to V2B participation compared to baseline charging (cycles). This formulation accounts for the total energy throughput (charge + discharge) during the connection period, adjusted for the net change in SOC and normalized by the battery’s energy capacity at the specified DoD (D).
- tarr(i)tarr(i), tdep(i)tdep(i): Arrival and departure time of the ithith EV.
- EEV,cha(i,t)EEV,cha(i,t), EEV,dis(i,t)EEV,dis(i,t): Charging and discharging energy of the ithith EV at time tt (kWh).
- SOCEV(i,tarr)SOCEV(i,tarr), SOCEV(i,tdep)SOCEV(i,tdep): State of Charge of the ithith EV at arrival and departure times.
- DD: Depth of Discharge (DoD), representing the maximum percentage of battery capacity used during a regular cycle, set to 80% [23].
4.1.3. EV Electricity Cost/Revenue (Cd,EV,eleCd,EV,ele)
This component captures the financial transactions associated with EV charging and discharging. EV users pay for the electricity they consume (charging) but receive compensation for the electricity they provide back to the building (discharging). The net cost/revenue is:Cd,EV,ele=∑t=124∑i=1N(PEV,dis(t)⋅EEV,dis(i,t)−PEV,cha(t)⋅EEV,cha(i,t))Cd,EV,ele=t=1∑24i=1∑N(PEV,dis(t)⋅EEV,dis(i,t)−PEV,cha(t)⋅EEV,cha(i,t))
where:
- PEV,cha(t)PEV,cha(t): Price paid by the EV user for charging energy at time tt (CNY/kWh).
- PEV,dis(t)PEV,dis(t): Compensation price paid to the EV user for discharging energy at time tt (CNY/kWh).
For simplicity and reflecting common practice in such studies, both PEV,cha(t)PEV,cha(t) and PEV,dis(t)PEV,dis(t) are set to a constant value of 1.2 CNY/kWh [20], representing a baseline service fee/compensation structure. (Note: More complex pricing schemes, like those linked directly to TOU grid prices, can be incorporated within this framework).
4.2. Constraints
The optimization model is subject to several physical and operational constraints to ensure feasibility, safety, and user satisfaction:
4.2.1. Energy Balance Constraint
The fundamental requirement for stable microgrid operation is instantaneous power balance at every time step tt:Ebuy(t)+EPV(t)+∑i=1NEEV,dis(i,t)=Eload(t)+Ediss(t)+∑i=1NEEV,cha(i,t)Ebuy(t)+EPV(t)+i=1∑NEEV,dis(i,t)=Eload(t)+Ediss(t)+i=1∑NEEV,cha(i,t)
This equation states that the total energy flowing into the microgrid (grid purchase Ebuy(t)Ebuy(t), PV generation EPV(t)EPV(t), EV discharge ∑EEV,dis(i,t)∑EEV,dis(i,t)) must equal the total energy flowing out (building consumption Eload(t)Eload(t), dissipated PV surplus Ediss(t)Ediss(t), EV charging ∑EEV,cha(i,t)∑EEV,cha(i,t)).
4.2.2. Grid Purchase Constraint
Energy purchased from the grid cannot be negative (no net export under the “no export” policy):Ebuy(t)≥0Ebuy(t)≥0
4.2.3. EV Operational Constraints
These constraints govern the charging and discharging behavior of each electric vehicle:
- Discharge Power Limit:0≤EEV,dis(i,t)≤αEV,dis(i,t)⋅Pmaxdis⋅Δt0≤EEV,dis(i,t)≤αEV,dis(i,t)⋅Pmaxdis⋅Δt
- Charge Power Limit:0≤EEV,cha(i,t)≤αEV,cha(i,t)⋅Pmaxch⋅Δt0≤EEV,cha(i,t)≤αEV,cha(i,t)⋅Pmaxch⋅Δt
- Mutual Exclusivity (No Simultaneous Charge/Discharge):αEV,dis(i,t)+αEV,cha(i,t)≤1αEV,dis(i,t)+αEV,cha(i,t)≤1
where:
* αEV,dis(i,t)αEV,dis(i,t), αEV,cha(i,t)αEV,cha(i,t): Binary decision variables (0 or 1) indicating if EV ii is discharging or charging at time tt.
* PmaxchPmaxch, PmaxdisPmaxdis: Maximum charge and discharge power rate of the EV (kW). Reflecting prevalent slow charging infrastructure and to mitigate potential battery stress, Pmaxch=Pmaxdis=7Pmaxch=Pmaxdis=7 kW.
* ΔtΔt: Time step duration (1 hour).
4.2.4. EV State of Charge (SOC) Constraints
- Departure SOC Requirement: Ensures the electric vehicle has sufficient energy for the user’s next trip:SOCEV(i,tdep)≥SOCEV,endSOCEV(i,tdep)≥SOCEV,endwhere SOCEV,endSOCEV,end is the minimum required SOC at departure, set to 80%.
- SOC Dynamics: Tracks the SOC of each EV battery over time:SOCEV(i,t)=SOCEV(i,t−1)−EEV,dis(i,t)/ηdis+EEV,cha(i,t)⋅ηchI(i)SOCEV(i,t)=SOCEV(i,t−1)−I(i)EEV,dis(i,t)/ηdis+EEV,cha(i,t)⋅ηchwhere:
- SOCEV(i,t)SOCEV(i,t), SOCEV(i,t−1)SOCEV(i,t−1): SOC of EV ii at time tt and t−1t−1.
- ηchηch, ηdisηdis: Charging and discharging efficiency coefficients (assumed constant, e.g., ηch=ηdis=0.95ηch=ηdis=0.95).
- SOC Safety Limits: Prevents battery over-charging or excessive depletion:SOCmin≤SOCEV(i,t)≤SOCmaxSOCmin≤SOCEV(i,t)≤SOCmaxwhere SOCmin=10%SOCmin=10% and SOCmax=90%SOCmax=90% of battery capacity.
- Initial SOC: Defined upon arrival:SOCEV(i,tarr)=SOCEV(i,0)SOCEV(i,tarr)=SOCEV(i,0)where SOCEV(i,0)SOCEV(i,0) is drawn from the appropriate initial SOC distribution (long or short commute).
5. Case Study: Beijing Office Building
To apply and validate the V2B optimization model, I selected a representative office building located in Beijing as the case study.
5.1. Building and PV System Modeling
- Building: A 4-story office building with a total floor area of 8,702 m² (2,175.5 m² per floor). The building is equipped with standard HVAC, ventilation, lighting, and office equipment systems.
- Building Load Simulation: The building’s hourly electricity demand profile (Eload(t)Eload(t)) was simulated over a full year using DeST software [24], a validated building energy simulation tool widely used in research and practice. Inputs included building geometry, construction materials, occupancy schedules, equipment schedules, internal loads, and Beijing’s typical meteorological year (TMY) weather data. Annual simulation results showed a total building electricity consumption of 401,800 kWh.
- PV System: A rooftop PV system comprising 1,000 monocrystalline silicon panels with a total installed capacity of 400 kW was modeled. The hourly PV generation output (EPV(t)EPV(t)) was calculated using established models [25, 26]:PPV(t)=NPV⋅fPV⋅PSTC⋅G(t)GSTC⋅[1+k⋅(Tcell(t)−TSTC)]PPV(t)=NPV⋅fPV⋅PSTC⋅GSTCG(t)⋅[1+k⋅(Tcell(t)−TSTC)]Tcell(t)=Tamb(t)+[(TNOCT−20)⋅G(t)800]Tcell(t)=Tamb(t)+[(TNOCT−20)⋅800G(t)]where:
- PPV(t)PPV(t): PV system output power at time tt (kW).
- NPVNPV: Installed PV area (m²).
- fPVfPV: PV derating factor (accounting for losses), set to 0.9.
- PSTCPSTC: PV module rated power under Standard Test Conditions (STC) (kW/m²).
- G(t)G(t): Actual solar irradiance on the tilted plane (W/m²).
- GSTCGSTC: Solar irradiance at STC (1000 W/m²).
- kk: PV power temperature coefficient (-0.47 %/°C).
- Tcell(t)Tcell(t): PV cell temperature at time tt (°C).
- TSTCTSTC: Cell temperature at STC (25 °C).
- Tamb(t)Tamb(t): Ambient temperature at time tt (°C).
- TNOCTTNOCT: Nominal Operating Cell Temperature (47 °C).
Using Beijing TMY data, the annual PV generation was simulated to be 395,000 kWh.
Table 1: Building and PV System Specifications
| Parameter | Value | Unit |
|---|---|---|
| Building Floor Area | 8,702 | m² |
| Annual Building Load | 401,800 | kWh |
| PV Installed Capacity | 400 | kW |
| Annual PV Generation | 395,000 | kWh |
| EV Fleet Size | 20 | Vehicles |
| EV Max Charge/Discharge Rate | 7 | kW |
The annual profiles revealed a crucial challenge: while the total annual PV generation (395,000 kWh) closely matched the total annual building load (401,800 kWh), significant temporal mismatches occurred on hourly and daily timescales. Periods of high PV generation (midday) often coincided with periods of lower building demand, leading to potential curtailment under the “no export” policy. Conversely, periods of high building demand (mornings, evenings) often occurred when PV generation was low or zero, necessitating grid purchases. This mismatch underscores the value proposition for energy storage like V2B to shift energy temporally.
5.2. Electricity Price Policy Setting
China employs diverse Time-of-Use (TOU) electricity tariffs to manage grid demand. I analyzed two prevalent and structurally different TOU tariff types representing current and evolving pricing structures:
- Policy 1 (Standard TOU – 4 Periods): Widely adopted (e.g., Beijing, Shanghai, Guangdong, etc.), featuring Peak, Shoulder, Off-Peak, and Deep Off-Peak periods (if applicable). This policy emphasizes cost differentiation during high grid stress periods and incentives for nighttime consumption. The specific periods vary seasonally.
- Policy 2 (Advanced TOU – 5 Periods): Emerging in regions with high RES penetration (e.g., Shandong, Jiangsu, Shaanxi), this policy introduces a Super Off-Peak period, typically during midday hours (e.g., 11:00-14:00) coinciding with peak solar PV generation. This aims to incentivize consumption during high RES output, reducing curtailment and better integrating renewables. This policy generally exhibits larger price differentials between periods.
Based on published tariffs from Beijing (Policy 1) and Shandong (Policy 2), representative prices and seasonal period definitions were established for the simulation:
*Table 2: Electricity Price Policies (CNY/kWh)*
| Tariff Policy | Super Peak | Peak | Shoulder | Off-Peak | Super Off-Peak | Price Ratio (Ref: Shoulder=1.00) |
|---|---|---|---|---|---|---|
| Policy 1 | 1.295 | 1.152 | 0.835 | 0.558 | N/A | 1.55 : 1.38 : 1.00 : 0.67 |
| Policy 2 | 1.252 | 1.100 | 0.742 | 0.385 | 0.283 | 1.69 : 1.48 : 1.00 : 0.52 : 0.38 |
Table 3: Representative Seasonal Period Definitions
| Season (Months) | Policy 1 Periods | Policy 2 Periods |
|---|---|---|
| Summer (Jul-Aug) | Peak: 10:00-15:00, 18:00-21:00; Shoulder: 07:00-10:00, 15:00-18:00, 21:00-23:00; Off-Peak: 23:00-07:00 | Super Peak: 10:00-11:00, 19:00-21:00; Peak: 11:00-14:00, 17:00-19:00; Shoulder: 08:00-10:00, 15:00-17:00; Off-Peak: 22:00-08:00; Super Off-Peak: 11:00-14:00 |
| Winter (Jan, Dec) | Peak: 08:00-11:00, 16:00-21:00; Shoulder: 06:00-08:00, 11:00-16:00, 21:00-22:00; Off-Peak: 22:00-06:00 | Similar structure adjustments reflecting winter demand patterns, maintaining Super Off-Peak at midday. |
| Transition (Other) | Periods adjusted based on typical load patterns. | Periods adjusted based on typical load patterns, maintaining Super Off-Peak at midday. |
Policy 2, with its larger price spreads (e.g., Super Off-Peak at 0.283 CNY/kWh vs. Peak at 1.100 CNY/kWh = 0.817 CNY/kWh difference) and inclusion of a midday low-price window aligned with solar generation, is considered more representative of future tariff structures aimed at facilitating RES integration. Therefore, Policy 2 forms the basis for exploring future scenarios involving increased price volatility and reduced technology costs.
5.3. Future Scenario Definition
To assess the resilience and future potential of V2B under evolving market conditions, I defined scenarios based on Policy 2:
- Increased Price Volatility: Reflecting potential future grid dynamics (higher RES share, demand shifts), the peak-to-valley price difference within Policy 2 was systematically increased by +10%, +20%, and +30%. This was implemented by proportionally increasing the prices of Super Peak and Peak periods while proportionally decreasing the prices of Off-Peak and Super Off-Peak periods relative to the Shoulder price, which was held constant. For example, a +10% increase in the price ratio spread changes the ratios from 1.69:1.48:1.00:0.52:0.38 to approximately 1.76:1.53:1.00:0.47:0.32.
- Reduced Charging Infrastructure Cost: Reflecting anticipated technological learning curves and economies of scale, the capital cost of bidirectional charging piles was reduced by -10%, -20%, and -30% from the current baseline of 20,000 CNY per 7 kW unit.
5.4. Simulation Methodology
Annual simulations were performed using a typical meteorological year (TMY) dataset for Beijing. To manage computational complexity while capturing seasonal and daily variability, the full year of building load, PV generation, and weather data was clustered into 16 representative typical days using the K-means algorithm. Annual results (total cost, energy metrics) were then calculated by weighting the results from each typical day by the number of days it represented in the full year. The EV fleet size was set to 20 vehicles based on findings suggesting optimal economic benefits for this building scale [27]. Key performance indicators analyzed include:
- Annual Economic Cost: Total cost comprising building grid electricity cost, EV battery degradation cost, EV electricity transaction cost, and annualized investment cost for V2B charging piles.
- Self-Supply Rate (SSR): Percentage of the building’s total electricity demand met by on-site PV generation (including direct use and storage via V2B).
- Self-Consumption Rate (SCR): Percentage of the total on-site PV generation consumed directly by the building or stored in EVs (minimizing dissipation EdissEdiss).
- Grid Electricity Consumption: Total annual electricity purchased from the grid.
- PV Utilization: Total annual PV generation consumed by the building or EVs.
6. Results and Discussion
6.1. V2B Performance Under Different Current Price Policies
The baseline scenario (“Original”) represents the building microgrid without any electric vehicle integration. Integrating 20 EVs with V2B capability (“V2B”) under optimized control using the MILP model yields significant benefits under both price policies.
Table 4: Annual Performance Comparison Under Current Price Policies
| Performance Indicator | Policy 1 (Original) | Policy 1 (V2B) | Policy 2 (Original) | Policy 2 (V2B) | Unit |
|---|---|---|---|---|---|
| Annual Economic Cost (per m²) | 16.61 | 10.47 | 14.67 | 8.29 | CNY/(m²·a) |
| Economic Cost Saving (per m²) | – | 6.14 | – | 6.38 | CNY/(m²·a) |
| Economic Cost Saving (%) | – | 36.96% | – | 43.53% | % |
| Self-Consumption Rate (SCR) | 68.59% | 91.25% | 68.59% | 91.06% | % |
| Self-Supply Rate (SSR) | 58.71% | 64.27% | 58.71% | 64.12% | % |
| Building Grid Consumption (per m²) | 46.12 | 45.77 | 46.12 | 45.69 | kWh/(m²·a) |
| PV Energy Utilized (Building + EV) (per m²) | 46.12 | 53.94 | 46.12 | 53.86 | kWh/(m²·a) |
| Total System Electricity Consumption (per m²) | 46.12 | 54.31 | 46.12 | 54.23 | kWh/(m²·a) |
Key Findings:
- Significant Cost Reduction: V2B integration delivers substantial economic savings under both TOU policies: 6.14 CNY/m²/yr (36.96%) under Policy 1 and 6.38 CNY/m²/yr (43.53%) under Policy 2. The slightly higher savings under Policy 2 are attributed to its larger price spreads, enabling more profitable energy arbitrage by the electric vehicle fleet.
- Enhanced RES Integration: Both SCR and SSR increase significantly with V2B. SCR jumps from 68.59% to over 91% under both policies, indicating that almost all generated PV is now consumed on-site or stored, drastically reducing wasted solar energy. SSR increases from 58.71% to around 64.1%, meaning a larger portion of the building’s total demand is met by on-site PV, thanks to the temporal shifting capability provided by EV batteries.
- Managed Energy Increase: Integrating 20 EVs naturally increases the total system’s electricity consumption by approximately 8.17-8.19 kWh/m²/yr (reflecting EV charging needs). However, V2B optimization ensures that 7.74-7.82 kWh/m²/yr of this additional demand is met by utilizing surplus PV generation that would otherwise have been dissipated. Consequently, the net increase in grid electricity purchases is minimal: only 0.35-0.43 kWh/m²/yr. This demonstrates V2B’s ability to satisfy EV charging requirements primarily using otherwise curtailed solar energy, minimizing the burden on the grid.
- Policy Advantage: Policy 2’s structure, particularly the Super Off-Peak period coinciding with high PV output, provides a marginally better environment for V2B economics (higher cost saving percentage) compared to Policy 1.
6.2. V2B Scheduling Strategy Analysis
To understand how V2B achieves these benefits, I analyzed optimized schedules for two contrasting typical days selected from the K-means clusters: one with low PV generation (Poor SSR = 25.12%, representative of winter) and one with abundant PV generation (Good SSR = 83.83%, representative of sunny spring/fall). Analysis focused on Policy 1 and Policy 2 under December and May price settings, respectively.
6.2.1. Day with Low PV Generation (Poor SSR)
- Policy 1 (Small Price Spreads): Optimization results show EV charging concentrated during periods of relatively lower prices or available PV: late morning/early afternoon (13:00-17:00, Shoulder price) and briefly during midday if PV is available (14:00-15:00). Crucially, no discharging occurs. The price difference between Peak (approx 1.15 CNY/kWh) and Off-Peak (0.56 CNY/kWh) periods is about 0.59 CNY/kWh. However, the compensation required to cover battery degradation for discharging is estimated around 0.5 CNY/kWh. Given the small arbitrage margin and the inherent losses in the charging/discharging cycle, discharging during Peak hours was not economically justified under this price structure. Optimized charging simply avoids the highest Peak prices.
- Policy 2 (Large Price Spreads): Charging occurs during the Super Off-Peak period (11:00-14:00, 0.283 CNY/kWh) and Shoulder periods (e.g., 10:00-11:00, 15:00). More importantly, significant discharging occurs during the early morning Peak period (07:00-09:00, 1.100 CNY/kWh). The price difference between Peak and Super Off-Peak is substantial (0.817 CNY/kWh). The analysis showed that discharging 209.05 kWh during Peak hours (avoiding cost at 1.100 CNY/kWh = 229.96 CNY savings) and charging that energy back during Super Off-Peak hours (cost at 0.283 CNY/kWh = 59.18 CNY) yielded a gross arbitrage profit of 170.78 CNY. After accounting for the battery degradation cost compensation of approximately 100.34 CNY, a net profit of 70.44 CNY was achieved. This clear economic incentive drives the V2B discharge behavior under Policy 2.
Conclusion for Low PV Days: When PV generation is insufficient and building load relies heavily on the grid, V2B strategy is highly sensitive to TOU price spreads. Small spreads (Policy 1) favor optimized charging only. Large spreads (Policy 2) make strategic discharging during high-price periods economically viable and beneficial.
6.2.2. Day with Abundant PV Generation (Good SSR)
- Policy 1: Charging occurs during early morning Off-Peak (06:00-08:00) and during the midday PV peak (12:00-15:00). Discharging occurs during the late morning Shoulder/Peak period (08:00-11:00) when PV generation might be ramping up but not yet sufficient to cover the full building load, and grid prices are relatively high. This strategy shifts 70.95 kWh of load from the high-price morning period to the midday period characterized by abundant PV and lower Shoulder prices. The grid cost saving was 46.53 CNY, battery degradation cost was 34.08 CNY, resulting in a net economic benefit of 12.45 CNY. This primarily utilizes the temporal flexibility of EV charging to absorb surplus midday PV and displace grid purchases during higher-priced periods.
- Policy 2: Charging is concentrated during the early morning Off-Peak (06:00-08:00) and the midday Super Off-Peak/PV peak period (11:00-15:00). Discharging occurs during the mid-morning Peak period (08:00-10:00). This strategy shifted 55.69 kWh, saving 46.20 CNY in grid costs with a battery degradation cost of 26.74 CNY, yielding a net benefit of 19.46 CNY.
Conclusion for High PV Days: When PV generation is abundant, V2B optimization consistently utilizes EV charging to absorb surplus midday solar energy. It also strategically discharges during high-price periods within the day (typically mornings) if the price spread justifies the battery wear, effectively time-shifting solar energy and reducing grid costs. The net economic gain arises from avoiding high grid prices during discharge times by using stored PV energy, even after accounting for battery degradation. Policy 2 again yielded a higher net benefit per kWh shifted due to its larger price differentials.
6.3. Impact of Future Price Volatility (Increased Peak-Valley Spread)
Building upon Policy 2 as the base case, increasing the peak-to-valley price difference significantly enhances the economic benefit of V2B.
Table 5: Impact of Increased Price Spread on V2B Economics (Policy 2 Base)
| Price Spread Increase | Annual Cost Saving (per m²) | Change vs. Base Policy 2 V2B | Saving Increase (%) |
|---|---|---|---|
| Base Policy 2 V2B | 6.39 | – | – |
| +10% | 7.35 | +0.96 | +15.11% |
| +20% | 7.56 | +1.17 | +18.11% |
| +30% | 7.63 | +1.24 | +19.37% |
Key Findings:
- Enhanced Savings: Larger price spreads directly increase the profitability of energy arbitrage via V2B. A 10% increase in spread boosts savings by 15.11%, a 20% increase by 18.11%, and a 30% increase by 19.37% compared to the base Policy 2 V2B case.
- Diminishing Marginal Returns: While savings increase with larger spreads, the rate of increase slows down. The jump from +20% to +30% spread yields only an additional 1.26% points in saving increase compared to the jump from +10% to +20% (3.00% points). This suggests a diminishing marginal economic return to ever-increasing price volatility from the V2B perspective. There is an optimal range of price differentiation beyond which further increases yield smaller relative benefits, important for future tariff design.
6.4. Impact of Reduced Charging Pile Investment Cost
Reducing the capital cost of bidirectional charging piles, a major barrier to V2B deployment, significantly improves the overall economic case.
Table 6: Impact of Reduced Charging Pile Cost on V2B Economics
| Charging Pile Cost Reduction | Annual Cost Saving Increase (%) |
|---|---|
| -10% | +4.8% |
| -20% | +9.6% |
| -30% | +14.4% |
Key Finding:
- Linear Cost-Benefit Improvement: The analysis reveals a clear linear relationship: every 10% reduction in the capital cost of bidirectional charging piles leads to a 4.8% improvement in the annual cost savings achieved by V2B. A 30% cost reduction improves the saving by 14.4%. This underscores the critical importance of technological advancements and economies of scale in driving down the cost of V2B hardware for widespread adoption.
7. Conclusions
My comprehensive analysis of V2B technology integrated with a typical office building under diverse current and future electricity price scenarios yields the following key conclusions:
- Significant and Robust Economic Benefit: V2B technology delivers substantial economic value under all analyzed electricity price structures. Optimized scheduling of electric vehicle charging and discharging reduced total annual system costs by 36.96% under the prevalent “Peak-Shoulder-Off-Peak” tariff (Policy 1) and by 43.53% under the more advanced “Peak-Shoulder-Off-Peak-Super Off-Peak” tariff (Policy 2) featuring larger price spreads and a midday low-price window. This demonstrates the core economic driver: leveraging electric vehicle batteries for price arbitrage and enhanced utilization of on-site PV.
- Tariff Structure Dictates V2B Strategy: The optimal operating strategy for the electric vehicle fleet is highly sensitive to the specific TOU tariff structure and its price differentials. Tariffs with smaller peak-valley spreads (like Policy 1) primarily benefit from optimized charging schedules (shifting charging away from Peak periods). Tariffs with larger spreads (like Policy 2) unlock the full potential of bidirectional V2B, making strategic discharging during high-price periods economically viable and significantly more profitable. The presence of a Super Off-Peak period aligned with solar noon is particularly advantageous.
- Increased Price Volatility Amplifies Benefits (with Diminishing Returns): Future scenarios involving increased electricity price volatility (larger peak-valley spreads) further enhance the economic attractiveness of V2B. A 10% increase in the peak-to-valley price difference within Policy 2 boosted V2B savings by 15.11%. A 20% increase boosted savings by 18.11%, and a 30% increase boosted savings by 19.37%. However, the analysis clearly shows diminishing marginal returns to increasing price spreads, suggesting an optimal range for maximizing V2B value from a system perspective.
- Charging Infrastructure Cost is Critical: The high upfront cost of bidirectional charging piles is a major barrier. My sensitivity analysis confirms that reductions in this cost directly and linearly improve V2B’s economic viability. Every 10% reduction in charging pile cost leads to a 4.8% increase in annual cost savings. Technological advancements and mass production driving cost reductions are therefore essential for widespread V2B deployment.
- Enhanced Renewable Integration: Beyond economics, V2B significantly improves the utilization of on-site renewable energy. The Self-Consumption Rate (SCR) increased from 68.59% to over 91% under both price policies, drastically reducing solar PV curtailment. The Self-Supply Rate (SSR) also increased, demonstrating that V2B helps meet more of the building’s demand with its own solar generation through time-shifting. Crucially, V2B integration allowed the additional electricity demand from charging 20 EVs to be met primarily (over 94%) by utilizing otherwise curtailed PV energy, minimizing the grid impact.
- V2B as a Zero-Carbon Enabler: The ability of V2B to reduce operational costs, decrease grid dependence, and, most importantly, maximize the utilization of intermittent on-site renewable generation like solar PV positions it as a critical enabling technology for achieving zero-carbon building goals. By transforming electric vehicles from passive loads into active grid-supporting resources, V2B contributes to building energy resilience and the broader stability of the electricity grid during the transition to high renewable penetration.
In summary, my research provides compelling evidence for the economic viability and strategic value of V2B technology in commercial buildings. While benefits are observable under current tariffs, they become substantially more pronounced under tariffs designed with larger price spreads to reflect grid costs and incentivize flexibility, and under future scenarios with reduced technology costs. Policymakers, utilities, building owners, and electric vehicle charging infrastructure providers should prioritize measures that facilitate the deployment of bidirectional charging and encourage electricity pricing structures that properly value the grid services provided by aggregated electric vehicle batteries. V2B represents a powerful synergy between the clean transportation revolution and the decarbonization of the built environment.