In the context of global efforts toward carbon neutrality, the building sector plays a pivotal role in energy conservation and emission reduction within power systems. Office buildings, in particular, account for a significant portion of urban energy consumption and carbon emissions, making their transition to low-carbon operations a critical priority. The integration of renewable energy sources, such as photovoltaic (PV) systems, presents both opportunities and challenges due to the inherent intermittency and unpredictability of solar power. This mismatch between PV generation and building energy demand often leads to inefficiencies, necessitating flexible load management solutions. Among these, bidirectional DC EV charging stations have emerged as a promising technology, enabling dynamic charging and discharging of electric vehicles (EVs) to enhance system flexibility, reduce operational costs, and lower carbon footprints. In this study, I explore the application of an orderly charging and discharging strategy for bidirectional DC EV charging stations in office buildings, evaluating its impact on economic and environmental performance through various operational scenarios.
The rapid adoption of EVs worldwide has introduced new dynamics into building energy systems. EVs not only represent a growing load but also a potential distributed energy resource when equipped with bidirectional charging capabilities. A bidirectional DC EV charging station allows for power flow in both directions, enabling EVs to supply electricity back to the building or grid during peak demand periods. This functionality transforms EVs into mobile storage units, facilitating peak shaving, valley filling, and improved renewable energy utilization. However, the effective implementation of such strategies requires sophisticated control mechanisms that consider real-time signals like time-of-use electricity prices and carbon emission factors. In this analysis, I develop a multi-source system model that integrates PV generation, battery storage, and bidirectional DC EV charging stations, assessing their coordinated operation under different optimization objectives.
The system topology for this study is based on a typical office building in Beijing, which has undergone partial DCification. The DC bus operates at 750 V, directly connecting DC loads, such as air conditioning systems, and bidirectional DC EV charging stations. An AC/DC converter interfaces the system with the external AC grid, allowing for energy exchange. In future phases, PV panels and battery storage systems are planned for integration into the DC bus. The core of this setup is the bidirectional DC EV charging station, which adjusts its power output based on system demands, supporting objectives like cost minimization and carbon reduction. The energy balance within the system is governed by the following equation:
$$ P_{EV} + P_{load} + P_g + (P_{PV} + P_b) = 0 $$
where \( P_{EV} \) is the total power of the EV charging stations (kW), \( P_{load} \) is the building’s DC load power (kW), \( P_g \) is the power exchange with the AC grid (kW), \( P_{PV} \) is the PV generation power (kW), and \( P_b \) is the battery storage power (kW). This equation ensures that the sum of all power flows equals zero, maintaining system stability.
Key components of the system are modeled to capture their dynamic behavior. For the bidirectional DC EV charging station, the EV charging process is described by a set of equations that account for energy constraints and power limits. The state of charge (SOC) of an EV battery at time \( t \) is given by:
$$ E_{EV}(t) = E_{EV,arr} + \sum_{t=T_{arr}}^{t} P_{EV}(t) \cdot \Delta t $$
where \( E_{EV}(t) \) is the real-time energy of the EV (kWh), \( E_{EV,arr} \) is the initial energy upon arrival (kWh), \( P_{EV}(t) \) is the charging or discharging power (kW), and \( \Delta t \) is the time step. The power output is constrained within permissible limits:
$$ P_{dis,max} \leq P_{EV}(t) \leq P_{ch,max} $$
with \( P_{ch,max} \) and \( P_{dis,max} \) being the maximum charging and discharging powers, respectively. The energy level must satisfy:
$$ E_{ch,min}(t) \leq E_{EV}(t) \leq E_{EV,max} $$
where \( E_{EV,max} \) is the maximum battery capacity, and \( E_{ch,min}(t) \) is the minimum required energy at time \( t \), calculated as:
$$ E_{ch,min}(t) = \max[(t – T_{dep}) \cdot P_{ch,max} + E_{EV,dep}, E_{EV,min}] $$
Here, \( T_{dep} \) is the departure time, and \( E_{EV,dep} \) is the desired energy level upon departure. Additionally, \( P_{EV}(t) = 0 \) when the EV is not connected (\( t \notin (T_{arr}, T_{dep}) \)), and \( E_{EV}(T_{dep}) = E_{EV,dep} \) ensures the charging demand is met. These constraints ensure user convenience while allowing flexibility in power调度.
For the battery storage system, the SOC is defined as:
$$ SOC_b(t) = \frac{E_b(t)}{Cap_b} $$
where \( E_b(t) \) is the available energy (kWh) and \( Cap_b \) is the nominal capacity (kWh). The SOC must remain within safe limits:
$$ SOC_{b,min} \leq SOC_b(t) \leq SOC_{b,max} $$
and the power output is bounded by:
$$ -P_{b,max} \leq P_b \leq P_{b,max} $$
where \( P_{b,max} \) is the maximum charging/discharging power. The efficiency of the battery, denoted as \( \eta_b \), is typically around 95%, accounting for energy losses during operation.
The control strategy for the system involves optimizing the operation of the bidirectional DC EV charging station and other devices based on dynamic grid signals. The EV charging station’s reference power \( P_{EV,ref} \) is determined by minimizing an objective function that incorporates economic or environmental goals. For instance, in cost optimization, the function aims to reduce the total electricity cost:
$$ \min \sum_{t \in T} (P_{g,im} \cdot c_{ele}) $$
where \( P_{g,im} \) is the power imported from the grid (kW), and \( c_{ele} \) is the time-of-use electricity price (USD/kWh). Similarly, for carbon emission reduction, the objective is to minimize the carbon footprint:
$$ \min \sum_{t \in T} (P_{g,im} \cdot C_r) $$
where \( C_r \) is the carbon emission factor (kg CO₂/kWh). The battery’s reference power \( P_{b,ref} \) is derived to maximize PV self-consumption:
$$ P_{b,ref} = \begin{cases}
\max(-P_{EV} – P_{load} – P_{PV}, -P_{b,max}) & \text{discharging} \\
\min(-P_{EV} – P_{load} – P_{PV}, P_{b,max}) & \text{charging}
\end{cases} $$
This ensures that the battery charges when there is excess PV generation and discharges during deficits, enhancing renewable energy utilization.
To evaluate system performance, I use several metrics. The levelized cost of electricity (LCOE) assesses economic efficiency:
$$ E_{ele} = \frac{\sum_{t \in T} (P_{g,im} \cdot c_{ele})}{W_{load} + W_{EV}} $$
where \( W_{load} \) and \( W_{EV} \) are the total energy consumption of the building load and EVs (kWh), respectively. The carbon emission per unit of electricity consumed is calculated as:
$$ E_{ce} = \frac{\sum_{t \in T} (P_{g,im} \cdot C_r)}{W_{load} + W_{EV}} $$
When PV and storage are integrated, the self-sufficiency ratio (SSR) measures the proportion of energy supplied locally:
$$ SSR = 1 – \frac{W_{g,im}}{W_{load} + W_{EV}} $$
where \( W_{g,im} \) is the total grid imports (kWh). A higher SSR indicates better renewable energy integration and reduced grid dependence.
In testing the bidirectional DC EV charging station’s power control capabilities, I conducted experiments to verify its responsiveness. For steady-state performance, power commands ranging from 10 kW to -10 kW in 2 kW increments were issued every 10 seconds. The results showed that the EV charging station accurately tracked the setpoints with deviations under 0.5 kW, demonstrating reliable control. In transient tests, a step change from -10 kW to 10 kW (a 20 kW swing) was applied, and the station achieved the target power within 0.2 seconds, highlighting its rapid response and stability. These findings confirm that the bidirectional DC EV charging station can effectively follow system commands, making it suitable for dynamic energy management.
To analyze the operational benefits, I compared the proposed smart charging strategy with a reference case of constant-power charging. In the economic optimization scenario, the time-of-use electricity prices vary throughout the day, as summarized in Table 1. The smart strategy dynamically adjusts the EV charging station’s power based on these prices, shifting loads to off-peak periods and even discharging during peak hours to reduce grid purchases. For example, during peak price intervals (e.g., 10:00-13:00), the EV charging station discharges to support building loads, while charging occurs in lower-price periods. This approach reduced the LCOE from 0.715 USD/kWh in the reference case to 0.615 USD/kWh, a 14.0% improvement. Similarly, in the carbon optimization scenario, the EV charging station responds to dynamic carbon emission factors, charging when emissions are low and discharging during high-emission periods. This led to a decrease in the carbon emission per unit electricity from 0.985 kg CO₂/kWh to 0.846 kg CO₂/kWh, also a 14% reduction. These results underscore the versatility of the bidirectional DC EV charging station in achieving both economic and environmental goals.
| Period | Price (USD/kWh) |
|---|---|
| Off-peak: 23:00-7:00 | 0.408 |
| Mid-peak: 7:00-10:00, 13:00-17:00, 22:00-23:00 | 0.652 |
| Peak: 10:00-13:00, 17:00-18:00, 21:00-22:00 | 0.895 |
| Super-peak: 18:00-21:00 | 1.025 |
When PV and battery storage are incorporated into the system, the bidirectional DC EV charging station’s role becomes even more critical. In a simulated typical day, the smart control strategy coordinates the EV charging station with PV generation and storage operations. For instance, during early morning hours when PV output is zero, the battery discharges to meet loads. As EVs connect, the EV charging station modulates its power to align with PV fluctuations, charging when surplus solar energy is available and discharging during deficits. This coordination boosted the SSR from 67.4% in the reference case to 82.6%, meaning over 80% of the energy was supplied locally. Consequently, the LCOE dropped sharply from 0.213 USD/kWh to 0.111 USD/kWh, a 47.9% reduction. The peak grid import power was also suppressed, alleviating stress on the external grid. These outcomes highlight the synergistic effects of integrating a bidirectional DC EV charging station with renewables and storage.
| Component | Parameter | Value |
|---|---|---|
| EV Charging Station | \( P_{ch,max} \) | 10 kW |
| \( P_{dis,max} \) | -10 kW | |
| PV System | Installed Capacity | 12 kW |
| Battery Storage | \( Cap_b \) | 6 kWh |
| \( SOC_{b,max} \) | 0.95 | |
| \( SOC_{b,min} \) | 0.10 | |
| Charge/Discharge Rate | 0.5 C | |
| Battery Efficiency | \( \eta_b \) | 95% |
The effectiveness of the bidirectional DC EV charging station strategy is further evident in its ability to handle real-world variability. For example, in days with high PV intermittency, the EV charging station compensates by adjusting discharge rates, ensuring continuous load supply. The mathematical formulation for the EV charging station’s power dispatch can be extended to include uncertainty terms, such as stochastic PV output or EV arrival patterns. However, in this study, deterministic models suffice to demonstrate core benefits. The bidirectional DC EV charging station not only supports grid stability but also empowers buildings to become prosumers, actively participating in energy markets. Future work could explore machine learning algorithms for predictive control, further enhancing the bidirectional DC EV charging station’s adaptability.
In conclusion, the integration of bidirectional DC EV charging stations in office buildings offers substantial advantages in terms of cost savings, carbon emission reductions, and renewable energy utilization. Through orderly charging and discharging strategies, the EV charging station acts as a flexible resource, responding to dynamic grid signals and coordinating with PV and storage systems. The results indicate that in standalone scenarios, the smart strategy reduces operational costs and emissions by approximately 14%, while in integrated systems, it boosts self-sufficiency by over 15% and cuts costs by nearly 48%. These findings validate the practical potential of bidirectional DC EV charging stations in advancing building energy systems toward sustainability and resilience. As EV adoption grows, leveraging such technologies will be crucial for achieving carbon neutrality goals in the built environment.