Smart Charging: How Time-Based Pricing Shapes EV Behavior in China’s Grid
As electric vehicles (EVs) surge across Chinese cities, their impact on the power grid has become a critical topic for energy planners and policymakers. While EVs represent a vital component of the nation’s carbon neutrality ambitions, their uncontrolled charging patterns pose a growing threat to grid stability. A new study from researchers at Anhui Vocational and Technical College and State Grid Anhui Marketing Service Center offers a compelling solution: leveraging dynamic pricing to turn EVs from grid stressors into flexible assets.
Published in the Journal of Changchun Institute of Technology (Natural Science Edition), the research led by Assistant Professor Xue Xinbai introduces a sophisticated demand-side response (DSR) strategy that aligns EV charging behavior with grid needs through optimized time-of-use (TOU) electricity tariffs. Rather than treating EVs as passive loads, the study reframes them as intelligent participants in a smarter, more resilient power system—one where economic incentives guide when and how vehicles charge.
The challenge is clear. As China accelerates its transition to electric mobility, the sheer volume of EVs plugging into the grid could amplify existing load imbalances. Peak electricity demand already strains distribution networks, particularly during summer afternoons or winter mornings when heating and cooling loads converge with household usage. Adding thousands of EVs charging simultaneously—especially during these high-demand windows—risks deepening peak loads, increasing infrastructure costs, and threatening reliability.
Yet, this challenge also presents an opportunity. Unlike traditional appliances, EVs offer temporal flexibility. Most private EVs are parked for extended periods, often overnight, creating a vast, distributed energy storage potential. The key lies in harnessing this flexibility through smart market mechanisms. This is where Xue Xinbai’s research makes a significant contribution.
At the heart of the study is a user response model grounded in the concept of price elasticity of demand. In simple terms, this measures how sensitive consumers are to changes in electricity prices. If prices rise during peak hours, will users shift their consumption to off-peak times? The model goes beyond a single elasticity value by employing a price elasticity matrix—a 24×24 grid that captures how a price change in one hour affects electricity demand in every other hour of the day. This nuanced approach recognizes that human behavior is complex: a price hike at 7 PM doesn’t just reduce demand at 7 PM; it may also increase demand at 10 PM or 6 AM, depending on user habits, work schedules, and lifestyle patterns.
The researchers applied this model to real-world data from Hefei, a major city in eastern China. Using actual daily load profiles from spring, summer, autumn, and winter of 2021, they simulated how different TOU pricing structures could reshape the city’s electricity demand curve. The optimization goal was straightforward: minimize the difference between peak and off-peak loads—a metric known as the peak-to-valley load difference (PPVL). A flatter load curve means more efficient grid operation, reduced need for costly peaking power plants, and enhanced system reliability.
To find the optimal pricing structure, the team employed a particle swarm optimization (PSO) algorithm. This computational method mimics the social behavior of birds flocking or fish schooling to efficiently search for the best solution in a complex landscape. Each “particle” in the simulation represents a potential TOU pricing scheme, with its position defined by the price set for each of the 24 hours. The algorithm iteratively adjusts these prices, evaluating how each configuration affects the PPVL, until it converges on the most effective strategy.
The results were striking. Across all four seasons, the optimized TOU tariffs successfully flattened the load curve. In winter, the peak-to-valley difference dropped by nearly 70%, from 1,353 MW to just 408 MW. Autumn saw a 53% reduction, spring 48%, and even summer—typically the most challenging season due to high cooling loads—achieved a 37% improvement. These numbers underscore the power of economic signals to influence collective energy behavior.
But the study didn’t stop at modeling residential and commercial demand. It extended the analysis to private EVs, which are increasingly central to urban energy dynamics. Using statistical models based on real-world driving data, the researchers simulated the charging behavior of approximately 47,000 private EVs in Hefei. Two scenarios were compared: one where EV owners charge immediately upon returning home (a common real-world behavior), and another where they respond to the optimized TOU tariffs by charging during the cheapest hours.
The contrast was dramatic. Under the “charge immediately” strategy, EV load clusters around evening hours—precisely when the grid is already under stress. But when users respond to price signals, charging shifts decisively to off-peak periods, typically between midnight and 6 AM. This strategic load transfer not only reduces strain on the grid but also delivers tangible economic benefits to EV owners. The study found that responsive charging could cut electricity costs for EV users by 48% to 52% across seasons. In summer, the average daily charging cost dropped from 7.43 yuan to just 3.53 yuan per vehicle—savings that could accelerate EV adoption by improving total cost of ownership.
What makes this research particularly robust is its grounding in real data and practical assumptions. The team didn’t rely on theoretical load curves; they used actual Hefei load data. They didn’t assume perfect user compliance; they modeled realistic driving and parking patterns, including the fact that most EVs return home between 5 PM and 8 PM and remain parked for six hours or more. They also accounted for battery constraints, ensuring that charging strategies respect minimum state-of-charge (SOC) levels to avoid damaging batteries.
Moreover, the study explores the sensitivity of results to policy design. When the researchers widened the TOU price range—from a peak-to-off-peak ratio of 3:1 to 5:1—the peak shaving effect improved significantly. In summer, the PPVL reduction jumped from 37% to nearly 70%. This finding has direct policy implications: stronger price signals yield stronger user responses. However, it also raises questions about equity and consumer acceptance. Extremely high peak prices could burden low-income households or those unable to shift their usage. The optimal balance between effectiveness and fairness remains a key area for future research.
Another strength of the paper is its forward-looking perspective. The authors acknowledge that their model is based on day-ahead pricing, where tariffs are announced the previous day. As power markets evolve and real-time pricing becomes more feasible, the potential for even finer control increases. With smart meters, connected vehicles, and advanced energy management systems, EVs could respond dynamically to real-time grid conditions, providing ancillary services like frequency regulation or voltage support. This vision aligns with the broader trend toward transactive energy systems, where distributed resources actively participate in energy markets.
The implications of this work extend beyond Hefei. Urban centers across China—and indeed, across the world—face similar challenges as EV adoption grows. Cities like Beijing, Shanghai, and Shenzhen are already experimenting with TOU pricing, but often with static, inflexible structures that don’t adapt to seasonal or daily variations in load. This research demonstrates the value of data-driven, adaptive pricing that evolves with changing grid conditions.
For utilities, the findings offer a roadmap for managing EV integration without massive infrastructure upgrades. By incentivizing off-peak charging, utilities can defer or avoid investments in new transformers, substations, and transmission lines. For grid operators, a flatter load curve improves voltage stability and reduces losses. For policymakers, the study supports the case for integrating DSR into national energy strategies, particularly as China seeks to balance its renewable energy expansion—wind and solar, which are intermittent—with growing electricity demand.
For EV owners, the message is equally clear: smart charging isn’t just good for the grid—it’s good for their wallets. As vehicle-to-grid (V2G) technology matures, the potential for two-way energy flow could turn EVs into revenue-generating assets. Owners might not only save on charging costs but also earn money by selling stored energy back to the grid during peak hours. While V2G remains in early stages, the foundation for such systems is being laid by today’s DSR programs.
The study also highlights the importance of user engagement. For TOU pricing to work, consumers must be informed, willing, and able to respond. This requires not just price signals but also user-friendly interfaces, real-time feedback, and education. Automakers and charging network operators have a critical role to play in integrating pricing information into vehicle infotainment systems and mobile apps, making it easy for drivers to schedule charging at optimal times.
Looking ahead, several research directions emerge. First, the interaction between TOU pricing and renewable energy generation deserves deeper exploration. In regions with high solar penetration, midday electricity prices might drop significantly, creating new charging opportunities. Second, the impact of extreme weather events—heatwaves or cold snaps—on user responsiveness should be studied, as these conditions can alter both load patterns and user behavior. Third, the long-term evolution of user elasticity needs monitoring; as more consumers adopt smart charging, the overall sensitivity to price may change.
Additionally, the role of aggregators—companies that pool the charging loads of many EVs to bid into energy markets—could be examined. These entities could act as intermediaries, simplifying participation for individual users while amplifying the grid benefits of coordinated charging.
In conclusion, Xue Xinbai and her colleagues have provided a timely and rigorous analysis of how economic incentives can shape the future of electric mobility. Their work demonstrates that with the right pricing signals, EVs can be a force for grid stability rather than disruption. By turning charging behavior into a controllable variable, cities can accommodate millions of EVs without compromising reliability or affordability.
As China continues its rapid electrification, studies like this offer a blueprint for a smarter, more sustainable energy future—one where every EV plugged in becomes a node in a responsive, resilient, and efficient power network. The road to carbon neutrality isn’t just about replacing internal combustion engines with electric motors; it’s about reimagining how energy is produced, delivered, and consumed. This research is a significant step in that direction.
Xue Xinbai, Anhui Vocational and Technical College; Hu Lvlong, State Grid Anhui Marketing Service Center; Journal of Changchun Institute of Technology (Natural Science Edition), doi:10.3969/j.issn.1009-8984.2024.02.008