China’s Mobile EV Charging Trucks Pivot to Grid Arbitrage Amid Demand Uncertainty
By late 2025, China’s expressways are set to host a new generation of mobile charging infrastructure—not just to plug in stranded electric vehicles (EVs), but to actively trade electricity with the grid. A groundbreaking study published this month reveals how truck-mounted mobile charging stations (TMCS) are being optimized not only as emergency roadside power banks but as dynamic energy assets capable of switching between serving drivers and arbitraging wholesale electricity markets—sometimes within the same day.
This dual-role strategy, developed by researchers at Tianjin University and Lanzhou Jiaotong University, leverages a sophisticated decision-making framework known as Information Gap Decision Theory (IGDT) to navigate the volatile terrain of uncertain EV demand and fluctuating grid constraints. Unlike traditional fixed charging stations, which remain idle during off-peak hours, TMCS units can relocate, recharge during low-price periods, discharge during high-price windows, and still respond to surges in roadside charging needs—all while maximizing operator revenue and maintaining service quality.
The innovation arrives at a critical juncture. China added nearly 100% more public chargers in 2022 alone, yet fixed infrastructure still struggles with high capital costs, long permitting timelines, and inflexibility during holiday travel spikes or unexpected grid outages. In response, national policy now mandates a “fixed-plus-mobile” charging network along major highways by 2025. Companies like NIO have already pledged to deploy 120 high-capacity TMCS units across China’s northwest and northeast corridors by year-end. Tesla and Volkswagen have demonstrated similar mobile platforms in North America and Europe, but China’s approach is distinct in its integration with grid markets and algorithmic dispatch.
At the heart of this new paradigm is a scheduling model that treats uncertainty not as a problem to eliminate—but as a variable to manage based on risk appetite. The research team, led by He Kecheng, Jia Hongjie, and Mu Yunfei, developed two complementary IGDT-based strategies: a robust model for conservative operators who prioritize guaranteed minimum returns, and an opportunistic model for aggressive players willing to bet on favorable demand or price swings.
In practical terms, this means a single TMCS might spend midnight to 3 a.m. drawing power from the grid at $0.03/kWh, store it in its 3 MWh battery pack, then sell it back at 1 p.m. for $0.12/kWh during peak pricing—unless, of course, real-time traffic data shows a sudden cluster of EVs queuing at a nearby fixed station. In that case, the unit reroutes, bypasses arbitrage, and delivers fast-charging services at a premium. The system continuously weighs these trade-offs using predictive models of driver behavior, road congestion, and locational marginal pricing.
Field simulations on a 465-kilometer ring expressway—with five interchanges, 18 fixed stations, and four TMCS units—demonstrated the model’s resilience. Under scenarios where EV demand dropped by 39% due to weather or economic factors, the robust IGDT strategy still secured 20% higher profits than conventional deterministic scheduling by shifting focus to grid arbitrage. Conversely, when demand surged by 94%, the opportunistic model boosted operator revenue by 12.6% while increasing charging service fulfillment from 73% to 91%, effectively decongesting overwhelmed fixed stations.
Critically, the system does not require precise probability forecasts—a major advantage in emerging markets where EV adoption patterns remain erratic. Traditional stochastic optimization methods demand accurate historical distributions of charging behavior, which are often unavailable or rapidly outdated. IGDT sidesteps this by defining an “information gap”—a radius of uncertainty around predicted values—and then optimizing for either worst-case protection (robust) or best-case exploitation (opportunistic). This makes the approach particularly suited for China’s fast-evolving EV ecosystem, where monthly sales can swing by 30% and new highway corridors open quarterly.
The economic implications are substantial. TMCS operators typically face three cost centers: battery degradation, labor/maintenance, and vehicle mileage. By intelligently alternating between high-margin arbitrage and high-utility roadside service, the IGDT model reduces unnecessary battery cycling and minimizes deadheading (empty travel). In one test case, a TMCS unit stationed near a major logistics hub earned 44% of its daily profit from grid trading during midday lulls, then switched to serving delivery EVs during evening shift changes—without returning to base.
Regulatory tailwinds are accelerating adoption. China’s “dual carbon” goals—peaking emissions by 2030 and achieving carbon neutrality by 2060—have made transport electrification non-negotiable. The National Development and Reform Commission now classifies mobile charging as “strategic flexible infrastructure,” eligible for subsidies and priority grid interconnection. Meanwhile, provincial power markets are piloting day-ahead pricing mechanisms that create clear arbitrage windows—exactly the conditions TMCS units need to thrive.
Industry players are taking note. NIO’s “Power North” initiative, targeting remote regions with sparse grid capacity, relies heavily on solar-charged TMCS units that operate off-grid. State Grid Corporation, China’s largest utility, is testing TMCS fleets as mobile grid-balancing resources during summer peaks. Even oil majors like Sinopec are converting gas station forecourts into hybrid hubs where TMCS units park during low-traffic hours to trade electricity.
Yet challenges remain. Battery costs, though falling, still dominate TMCS economics—especially when units cycle multiple times per day. Current models assume 12 fast-charging ports per truck, but real-world deployments often cap at 6–8 to preserve range and payload. Moreover, while the IGDT framework handles demand and grid uncertainty well, it doesn’t yet incorporate weather disruptions, cyberattacks, or policy shocks—factors that could widen the information gap beyond modeled thresholds.
Looking ahead, the researchers suggest integrating TMCS into virtual power plants (VPPs), where aggregated mobile units could bid into ancillary service markets for frequency regulation or black-start capability. As China’s spot electricity markets mature, such capabilities could turn TMCS from cost centers into profit engines. Pilot programs in Guangdong and Jiangsu are already exploring this, with early results showing TMCS participation can reduce local grid congestion by up to 18% during holiday travel surges.
For global observers, China’s TMCS strategy offers a blueprint for managing EV infrastructure in volatile environments. Unlike the U.S. or EU, where mobile charging remains largely reactive (e.g., event support or disaster relief), China is embedding mobility into the core economics of charging networks. The result is a system that doesn’t just respond to uncertainty—it profits from it.
As He Kecheng, lead author and lecturer at Lanzhou Jiaotong University, notes: “The future of EV infrastructure isn’t just about more chargers—it’s about smarter, more adaptive assets that can wear multiple hats. A TMCS isn’t a backup plan. It’s a strategic node in the energy-transport nexus.”
With over 20 million EVs expected on Chinese roads by 2026, the race is on to build not just capacity, but intelligence. Mobile charging trucks, once seen as stopgap solutions, may well become the agile backbone of a decarbonized transport future.
Author Affiliations:
He Kecheng¹,², Jia Hongjie¹, Mu Yunfei¹, Yu Xiaodan¹, Xiao Qian¹, Dong Xiaohong³
¹Key Laboratory of Smart Grid of Ministry of Education (Tianjin University), Tianjin 300072, China
²School of New Energy and Power Engineering, Lanzhou Jiaotong University, Lanzhou 730070, China
³State Key Laboratory of Reliability and Intelligence of Electrical Equipment (Hebei University of Technology), Tianjin 300131, China
Journal: IEEE Transactions on Intelligent Transportation Systems
DOI: 10.1109/TITS.2024.3387652