Source-linked AI summary
Economic Evaluation of V2G-Enabled Fast Charging Stations Under Endogenous EV Adoption Dynamics
Mingjian Tuo, Jie Zhou, Yao Yan, Cunzhi Zhao, Long Wang, Mulan Zhang
TL;DR
The paper addresses planning that overlooks infrastructure-driven EV adoption and the computational burden of tracking individual vehicle states. It combines endogenous adoption with aggregate V2G modeling in collaborative FCS planning, reporting different adoption and V2G outcomes across strategies.
Problem
Many studies assume exogenous EV demand despite infrastructure affecting adoption, while tracking microscopic vehicle states at scale is computationally intractable for macroscopic planning.
Source: §1 · p. 2Method
The framework models infrastructure-dependent EV adoption and aggregates fleet V2G flexibility through a virtual-battery causal inequality.
Source: §1 · p. 2, §2.1 · p. 2, §2.3 · p. 4Results
25.36% EV penetration and 91.18% V2G potential under Strategy I compare with 19.88% EV adoption under Strategy III.
Source: §4.2 · p. 6Takeaways & Limitations
The comparison indicates that deploying V2G without endogenous traffic feedback leads to inefficient infrastructure utilization.
Source: §4.2 · p. 6The conclusion leaves large-scale heuristic decomposition and out-of-sample robustness verification to future research.
Source: §5 · p. 8
Abstract
from arXiv · showhide
Building fast charging stations (FCSs) is crucial for transportation electrification, but there exists an indirect network effect: while the increasing number of electric vehicles (EVs) decides the FCS capacity expansion, the spatial locations of these facilities strongly influence drivers' willingness to adopt EVs. Ignoring this interaction can lead to bad capital investments and exacerbate power grid vulnerabilities during tidal traffic peaks. Therefore, we explicitly model the EV adoption dynamics as decision-dependent uncertainties (DDUs) in a new multi-period collaborative planning framework. This framework evaluates the economic viability of V2G-enabled FCSs across both transportation and distribution networks. To simplify the complex calculation, we introduce an aggregated fleet virtual battery model to catch macroscopic vehicle-to-grid (V2G) flexibility. This successfully circumvents the dimension curse inherent in tracking microscopic state-of-charge. To further guarantee calculation speed, the nonlinear infrastructure exposure is transformed into a mixed-integer program by using Special Ordered Set type 2 (SOS2) variables and Second-Order Cone Programming (SOCP) relaxations for grid limits. Finally, numerical studies on a coupled Sioux Falls and IEEE 33-bus testbed prove that our framework achieves superior expected social welfare. Also, macroscopic V2G aggregation is highlighted for its capability to mitigate distribution grid congestion penalties.
1 Introduction
The introduction identifies a gap in planning that treats EV demand as exogenous, despite infrastructure influencing adoption and charging creating grid pressures. It proposes endogenous adoption modeling and macroscopic V2G aggregation for collaborative FCS planning.
- Most current studies assume exogenous EV demand, omitting infrastructure-driven adoption feedback that may lead to poor investments and greater grid vulnerability during traffic peaks.Source: §1 · p. 2
- Tracking microscopic state of charge for thousands of moving vehicles makes macroscopic planning computationally intractable.Source: §1 · p. 2
- A data-driven piecewise-linear mapping represents decision-dependent adoption trajectories in a form suited to mixed-integer linear programming.Source: §1 · p. 2
- An aggregated fleet virtual battery captures large-scale V2G arbitrage while avoiding the combinatorial explosion of tracking microscopic state of charge.Source: §1 · p. 2
2 System Modeling
System modeling links infrastructure-dependent EV adoption to traffic patterns and distribution-grid operations, while representing fleet V2G flexibility without tracking individual vehicle states.
2.1 Endogenous EV Adoption Dynamics (DDU)
- EV adoption follows logistic diffusion with an infrastructure-dependent saturation limit, and an SOS2 piecewise-linear mapping keeps the decision-dependent model tractable and polyhedral.Source: §2.1 · pp. 2–3
2.2 Spatiotemporal Traffic Flow and Virtual Queuing
- Destination-based tidal probabilities project adoption onto hourly traffic; a static topology mask bounded by driving range and virtual queues buffer temporal congestion, with terminal clearance preventing deferred demand.Source: §2.2 · p. 3
2.3 Active Distribution Network and Macroscopic V2G
- Distribution operations use a penalty-minimization OPF with DistFlow branch modeling and an SOCP relaxation of branch-power limits.Source: §2.3 · pp. 3–4
- A macroscopic causal inequality bounds cumulative fleet V2G discharge by cumulative charging energy plus the initial fleet-energy cushion, ensuring aggregate thermodynamic consistency.Source: §2.3 · p. 4
3 Problem Formulation
The planning objective weighs annualized infrastructure investment and operational reliability against charging-service benefits and traffic and grid penalties.
- The master objective minimizes equivalent annualized investment and expected operational penalties while accounting for charging-service revenue and traffic and grid violations.Source: §3 · pp. 4–5
4 Case Study
The coupled Sioux Falls–IEEE 33-bus case compares four planning strategies and attributes the value of joint V2G and endogenous adoption planning to adoption gains and V2G arbitrage. The results also show a trade-off between grid congestion and EV penetration.
4.1 System Setup and Parameters
- The case study couples the 24-node Sioux Falls transportation network with the IEEE 33-bus active distribution system over six years and three investment stages.The grid VOLL is 25,000 $/MWh, the unserved EV traffic penalty is 500 $/MWh, and each fast-charging module is 120 kW.Source: §4.1 · p. 5
4.2 Economic Value of Endogenous DDU Modeling
- DDU strategies reach 25.36% and 26.69% EV adoption, while open-loop Strategy IV reaches 15.37% with the lowest CAPEX, 45.37 M$; Strategy I CAPEX is 57.37 M$.The authors report that endogenously stimulated demand offsets Strategy I's higher initial outlay through long-term social service benefits.Source: Table 1 · p. 6, §4.2 · p. 6
- V2G strategies drive grid congestion indices to about 215–217 p.u./MWh; Strategy I combines 25.36% EV penetration with 91.18% V2G potential, while Strategy III has 19.88% adoption.The comparison links V2G without endogenous traffic feedback to inefficient infrastructure utilization.Source: §4.2 · p. 6, Table 1 · p. 6
4.3 Value Breakdown of V2G and Closed-Loop Planning
- About 10% absolute market penetration increase is attributed to endogenous interaction capturing extra revenue from an infrastructure-stimulated EV fleet.The waterfall chart decomposes the welfare improvement from baseline Strategy IV to global-optimum Strategy I.Source: §4.3 · p. 7, Figure 2 · p. 7
- The closed-loop V2G value uses 91.18% V2G energy capacity as an aggregated virtual battery and applies spatiotemporal arbitrage against 57.37 M$ CAPEX.The authors describe the active distribution network as helping maximize overall social welfare rather than serving only as a physical constraint.Source: §4.3 · p. 7, Figure 2 · p. 7
5 Conclusion
The conclusion presents a collaborative V2G-enabled fast-charging planning framework driven by endogenous EV adoption, using a macroscopic virtual-battery causal inequality to avoid microscopic tracking burdens. Numerical simulations validate its feasibility and superiority, while large-scale heuristic decomposition and out-of-sample robustness remain future work.
5 Conclusion
- The framework combines endogenous EV adoption planning with a macroscopic virtual-battery causal inequality to bypass the combinatorial explosion of microscopic vehicle tracking.The authors say it preserves exact V2G arbitrage boundary conditions while guaranteeing computational tractability.Source: §5 · p. 7
- Numerical simulations validate the framework's superiority and feasibility and report accelerated regional EV adoption and neutralized uncoordinated grid congestion.Source: §5 · p. 7
- Future work will explore large-scale heuristic decomposition and out-of-sample robustness verification in coupled power-traffic systems.Source: §5 · p. 8