China’s EV Charging Stations Cut Grid Dependence with Smart Solar-Storage Integration
By integrating solar power and battery storage, China is redefining the economics of electric vehicle (EV) charging infrastructure—turning once-grid-dependent stations into semi-autonomous energy hubs that slash peak-hour electricity purchases, reduce user wait times, and boost overall system efficiency. A new study from researchers at the Electric Power Research Institute of State Grid Fujian Electric Power Company Co., Ltd. demonstrates how optimized capacity planning for photovoltaic (PV) and energy storage systems can lower total social costs by over 12% while maintaining high service quality.
The findings arrive at a pivotal moment. As China’s EV fleet surges past 20 million vehicles—nearly 60% of the global total—the strain on urban power grids intensifies. Traditional fast-charging stations, especially those equipped with 180-kW or higher chargers, draw massive loads during daytime peak hours, triggering costly demand charges and grid congestion. But the integration of on-site renewables and storage offers a compelling alternative: not just greener charging, but smarter, more profitable operations.
At the heart of this shift is a novel optimization model that treats the charging station not merely as a service point, but as a dynamic micro-energy system. Developed by Li Zhicheng, Chen Dawei, and Zhang Weijun, the model minimizes total social cost—a composite metric encompassing fixed investment, PV and storage capital and maintenance expenses, grid electricity purchases, and user waiting time—while respecting hard constraints on land use, transformer capacity, queue length, and equipment utilization.
Unlike prior studies that focus solely on investor-side economics or technical feasibility, this approach explicitly balances infrastructure costs with user experience. “It’s not enough to install solar panels and batteries,” explains Li Zhicheng, lead author and senior engineer specializing in integrated energy systems. “You must configure them in a way that aligns with real-world EV arrival patterns, local solar irradiance, and time-of-use electricity pricing. Otherwise, you risk underutilization or service degradation.”
The team’s methodology begins with a high-fidelity simulation of daily EV charging demand. Using Monte Carlo sampling, they model vehicle arrivals and state-of-charge (SOC) distributions based on empirical data: drivers typically arrive with low battery levels and expect to charge up to 95% SOC. Each vehicle’s charging duration is calculated from battery capacity (29 kWh in the baseline case), charger power (180 kW), and system efficiency (90%). This generates a realistic 24-hour charging load curve—peaking during midday and evening commutes.
Overlaying this demand profile with a typical PV generation curve—zero at night, ramping up at sunrise, peaking near noon, and tapering off by sunset—the researchers then apply a rule-based operational strategy tied to China’s three-tier time-of-use tariff structure: off-peak (11 p.m.–7 a.m. at $0.05/kWh), mid-peak (7–10 a.m., 3–6 p.m., 9–11 p.m. at $0.12/kWh), and peak (10 a.m.–3 p.m., 6–9 p.m. at $0.19/kWh).
During off-peak hours, with no solar generation, the station draws cheap grid power to serve EVs and charge its battery. In mid-peak periods, solar output first meets EV demand; any surplus charges the battery or, if full, is sold back to the grid at 80% of the retail rate. During expensive peak hours, the system prioritizes solar and battery discharge to avoid grid purchases—only resorting to the grid when local resources are exhausted.
This strategy is embedded within a constrained optimization framework solved via a quantum-enhanced particle swarm algorithm—a metaheuristic known for robust global search capabilities in high-dimensional spaces. The decision variables include the number of chargers (5–20 units), PV capacity (0–800 kW), and storage capacity (0–1,000 kWh). The algorithm iteratively evaluates configurations against the total cost objective and feasibility constraints.
Four scenarios were tested. Scenario 1: baseline station with no PV or storage. Scenario 2: 760 kW of solar only. Scenario 3: 1,000 kWh battery only. Scenario 4: combined 760 kW PV + 1,000 kWh storage—the optimal configuration identified by the model.
The results are striking. In Scenario 1, the station draws all power from the grid, incurring $13.18 in hourly electricity costs (converted from ¥932.12/day). User waiting time averages 22 minutes due to limited chargers, violating the 20-minute service threshold. Total social cost: $14.64/hour.
Adding only solar (Scenario 2) reduces grid purchases by 72%, cutting electricity costs to $3.62/hour. However, without storage, midday solar surplus can’t be shifted to evening peaks, limiting savings. Total cost drops to $11.72/hour—a 20% improvement—but waiting time remains high due to unchanged charger count.
Battery-only integration (Scenario 3) enables load shifting: charging cheaply overnight and discharging during peaks. Grid purchases fall by 60%, yet the lack of solar means the battery must be cycled daily using grid power, increasing wear and operational complexity. Total cost: $13.03/hour—only marginally better than baseline.
But the hybrid system (Scenario 4) unlocks synergistic benefits. Solar meets ~60% of daytime demand directly; excess charges the battery. The battery then discharges during evening peaks, reducing grid reliance by 82% compared to baseline. Electricity costs plummet to $3.03/hour. Crucially, the optimization also recommends increasing chargers from 5 to 6 units—raising equipment utilization to 78% while cutting average queue time to just 11 minutes, well below the 20-minute limit.
Total social cost in Scenario 4: $11.56/hour—12.4% lower than baseline and the lowest among all options. The model confirms this is the global optimum: further increases in PV or storage raise capital costs faster than they reduce electricity expenses.
From an investor perspective, the payback is compelling. Though upfront costs for PV ($7,000/kW) and storage ($2,000/kWh including power electronics) are significant, the 10-year net present value analysis—using a 5% discount rate—shows positive returns in regions with high peak tariffs and strong solar resources. Maintenance is minimal: 1% annual O&M for PV, and battery replacement every 7 years.
For grid operators, the implications are equally profound. Widespread adoption of such optimized stations could defer costly substation upgrades in dense urban corridors. During heatwaves or cold snaps—when both EV charging and HVAC loads spike—distributed solar-storage assets can act as virtual peaker plants, enhancing grid resilience.
Critically, the study validates that user experience isn’t sacrificed for efficiency. By embedding queuing theory (M/G/s model) directly into the cost function, the model ensures service quality remains high. This dual focus—economic and experiential—is what distinguishes it from purely engineering-driven approaches.
Industry experts note that China’s policy environment is uniquely conducive to this model. Generous feed-in tariffs, streamlined interconnection rules, and municipal mandates for green infrastructure in public charging networks create fertile ground for deployment. Companies like NIO, XPeng, and State Grid’s own EV subsidiary are already piloting solar-canopied charging plazas in Fujian, Guangdong, and Zhejiang.
Yet challenges remain. Land scarcity in megacities limits rooftop PV area. Battery degradation under frequent cycling affects long-term economics. And regulatory uncertainty around grid export compensation persists in some provinces.
Still, the trajectory is clear. As battery prices continue to fall—down 89% since 2010—and bifacial solar panels boost yield by 10–15%, the business case strengthens. Future iterations may incorporate vehicle-to-grid (V2G) capabilities, turning parked EVs into mobile storage units.
For global automakers and energy firms watching China’s EV ecosystem, the message is unambiguous: the next competitive frontier isn’t just in batteries or autonomous driving—it’s in intelligent, integrated charging infrastructure that harmonizes generation, storage, demand, and user behavior.
“This isn’t just about saving money,” says Zhang Weijun, co-author and expert in power system automation. “It’s about building a charging network that’s resilient, user-friendly, and aligned with national decarbonization goals. The technology exists. Now it’s about smart deployment.”
As cities from Los Angeles to Berlin grapple with how to electrify transport without overloading grids, China’s data-driven, holistic approach offers a replicable blueprint—one where every charging station becomes a node in a cleaner, more agile energy future.
Li Zhicheng, Chen Dawei, Zhang Weijun, Electric Power Research Institute of State Grid Fujian Electric Power Company Co., Ltd., Distributed Energy, Vol. 9, No. 6, Dec. 2024, DOI: 10.16513/j.2096-2185.DE.2409610