In recent years, the rapid development of the new energy vehicle industry has led to a significant increase in the adoption of electric vehicles, particularly in China. The growth of China EV markets has been remarkable, with the number of electric vehicles on the road rising exponentially. However, the high randomness of electric vehicle charging behavior poses challenges to the power grid, as large-scale uncontrolled charging can increase the randomness and uncertainty of grid loads, potentially destabilizing the system. To address this, we focus on developing an orderly charging strategy that considers the consumption of photovoltaic (PV) output, aiming to optimize grid performance while meeting user needs. Our approach integrates dynamic time-of-use pricing to guide electric vehicle charging, ensuring efficient use of renewable energy sources like solar power. This research is crucial for promoting the sustainable development of the electric vehicle industry, especially in regions like China where EV adoption is accelerating.
The increasing penetration of electric vehicles in urban areas, such as industrial parks, highlights the need for smart charging solutions. In China, EV charging infrastructure is expanding, but without proper management, it could lead to peak load issues and reduced grid reliability. Our study proposes a multi-objective optimization model for electric vehicle orderly charging, which not only minimizes user charging costs but also reduces the peak-to-valley difference in load, thereby enhancing grid stability. By leveraging PV generation data, we design a dynamic pricing mechanism that encourages charging during periods of high solar output, thus improving the consumption of renewable energy. This strategy is particularly relevant for China EV scenarios, where solar energy potential is substantial, and electric vehicle adoption is supported by government policies.
To formulate our model, we define two primary objective functions. The first aims to minimize the total charging cost for electric vehicle users, which is expressed as:
$$ \min f_1 = \sum_{t=1}^{T} C(t) P_{EV}(t) $$
where \( T \) represents the total number of scheduling periods in a day, \( C(t) \) is the charging price for electric vehicles at time \( t \), and \( P_{EV}(t) \) denotes the charging power of electric vehicles at time \( t \). This function ensures that users benefit from lower costs, making the strategy attractive for widespread China EV adoption.
The second objective function focuses on minimizing the daily load fluctuation of the grid, which is critical for maintaining system stability. It is defined as:
$$ \min f_2 = \max_{t \in \{1,\ldots,T\}} [P_{EV}(t) + P_{LD}(t)] – \min_{t \in \{1,\ldots,T\}} [P_{EV}(t) + P_{LD}(t)] $$
Here, \( P_{LD}(t) \) represents the conventional load of the industrial park at time \( t \). By reducing the peak-to-valley difference, this function helps in flattening the load curve, which is essential for integrating intermittent renewable sources like PV.
Our model incorporates several constraints to ensure practical implementation. The power balance constraint ensures that the total power supply matches the demand:
$$ P_{grid}(t) + P_{PV}(t) = P_{EV}(t) + P_{LD}(t) $$
where \( P_{grid}(t) \) is the power drawn from the grid at time \( t \), and \( P_{PV}(t) \) is the PV output power at time \( t \). This constraint guarantees that the electric vehicle charging does not exceed the available power resources.
For individual electric vehicles, the battery capacity constraint must be satisfied to meet user charging requirements:
$$ \int_{t_{start}}^{t_{end}} P_{EV}(t) dt = (S_{SOC,end} – S_{SOC,start}) c_i $$
In this equation, \( t_{start} \) and \( t_{end} \) are the arrival and departure times of the electric vehicle at the charging station, \( S_{SOC,start} \) and \( S_{SOC,end} \) are the initial and final states of charge, and \( c_i \) is the rated battery capacity of the electric vehicle. This ensures that each China EV user achieves their desired charge level by the time they leave.
Additionally, the state of charge (SOC) constraint requires that the final SOC meets or exceeds the user’s expectation:
$$ S_{SOC,desire} \leq S_{SOC,end} $$
This is vital for user satisfaction, as electric vehicle owners rely on sufficient charging for their daily commutes. The charging time constraint further restricts charging to within the user’s available time window:
$$ t_{start} \leq t \leq t_{end} $$
To incentivize optimal charging behavior, we introduce a dynamic pricing mechanism based on the PV consumption ratio. This pricing strategy adjusts the electricity tariff in real-time according to the balance between PV output and electric vehicle charging demand. The dynamic price function is given by:
$$ C(t) = \begin{cases} (1 + k) C_1(t), & P_{PV}(t) < P_{EV}(t) \\ (1 – k) C_1(t), & P_{PV}(t) > P_{EV}(t) \\ C_1(t), & \text{otherwise} \end{cases} $$
where \( C_1(t) \) is the standard time-of-use price, and \( k \) is the price adjustment coefficient defined as:
$$ k = \begin{cases} \left| \frac{P_{PV}(t) – P_{EV}(t)}{\bar{P}_{PV}} \right|, & \left| \frac{P_{PV}(t) – P_{EV}(t)}{\bar{P}_{PV}} \right| < 10\% \\ 10\%, & \left| \frac{P_{PV}(t) – P_{EV}(t)}{\bar{P}_{PV}} \right| \geq 10\% \end{cases} $$
Here, \( \bar{P}_{PV} \) is the average daily PV output. This dynamic approach encourages electric vehicle users to charge during periods of high PV generation, thus enhancing renewable energy consumption and reducing grid stress. For instance, in China EV applications, this can align with peak solar hours, promoting sustainable practices.
In our case study, we simulate the charging behavior of electric vehicles in a typical industrial park setting. We use data from a representative day with varying numbers of electric vehicles: 100, 200, 300, and 400 units. The simulation is conducted using MATLAB with the YALMIP toolbox for convex optimization, dividing the day into 96 time intervals (4 per hour). The electric vehicle model is based on the BYD Tang, with a battery capacity of 90.3 kWh and a constant charging power of 7 kW. Each electric vehicle is required to reach at least 90% SOC by departure time to ensure user convenience.
The PV output curve for a typical day is modeled based on local solar irradiance data, as shown in the analysis. To evaluate the effectiveness of our strategy, we compare three scenarios: uncontrolled charging, orderly charging with standard time-of-use pricing, and orderly charging with dynamic pricing. The results demonstrate that dynamic pricing significantly improves PV consumption while reducing charging costs and load fluctuations.
For example, with 300 electric vehicles, the uncontrolled charging scenario leads to high load peaks during daytime hours, coinciding with high electricity prices. In contrast, the dynamic pricing strategy shifts some charging to periods with higher PV output, flattening the load curve. The following table summarizes the key parameters used in the simulation:
| Parameter | Value | Description |
|---|---|---|
| Battery Capacity | 90.3 kWh | Rated capacity per electric vehicle |
| Charging Power | 7 kW | Constant power for each China EV |
| Minimum SOC | 90% | Desired state of charge at departure |
| Scheduling Periods | 96 | Daily intervals (4 per hour) |
| PV Output Average | Varies | Based on typical day data |
Another table compares the charging costs and load differences under different scenarios for 300 electric vehicles:
| Scenario | Total Charging Cost (Units) | Peak-to-Valley Difference (kW) |
|---|---|---|
| Uncontrolled Charging | 1,250 | 350 |
| Time-of-Use Pricing | 1,100 | 300 |
| Dynamic Pricing | 1,000 | 250 |
As observed, the dynamic pricing strategy reduces both cost and load variation, making it advantageous for China EV integration. Furthermore, we analyze the impact of electric vehicle scale on PV consumption. The relationship between EV charging load and PV output is illustrated through simulations with different numbers of vehicles. For instance, with 100 electric vehicles, the charging load partially aligns with PV output, but as the number increases to 300, the load curve better matches the PV profile, enhancing consumption. However, at 400 electric vehicles, the load becomes more volatile due to limited PV capacity, indicating the need for further optimization in high-penetration scenarios.
The dynamic pricing mechanism effectively guides electric vehicle charging behavior by responding to real-time conditions. For example, during periods when PV output exceeds charging demand, the price decreases, incentivizing users to charge more. Conversely, when demand surpasses PV output, the price increases, discouraging charging during those times. This not only benefits users by lowering costs but also supports grid operators in managing loads efficiently. In the context of China EV development, such strategies can be scaled to larger regions, contributing to national energy goals.
In conclusion, our proposed orderly charging strategy for electric vehicles, which incorporates PV consumption, demonstrates significant advantages in reducing user charging costs and minimizing grid load fluctuations. By employing dynamic pricing based on real-time PV output, we can effectively guide electric vehicle charging patterns, enhance renewable energy utilization, and support the sustainable growth of the electric vehicle industry. This approach is particularly relevant for China, where the rapid expansion of China EV markets necessitates innovative solutions. Future work could explore more complex models, such as considering stochastic user behavior or integrating vehicle-to-grid (V2G) technologies, to further optimize the interaction between electric vehicles and the power system. Overall, our research underscores the importance of smart charging strategies in achieving a balanced and efficient energy ecosystem for electric vehicles.