Economic and Technical Feasibility of V2G in Non-Road Mobile Machinery sector

Technical Summary: Economic and Technical Feasibility of V2G in the Non-Road Mobile Machinery Sector

Problem Statement
The paper addresses the potential for integrating Vehicle-to-Grid (V2G) technology into the Non-Road Mobile Machinery (NRMM) sector, specifically focusing on electric tractors within a rental service model. While V2G is well-explored in passenger vehicles and electric bus fleets, its application to NRMM remains under-investigated. The authors posit that NRMM fleets offer unique advantages over private EVs, including predictable usage schedules and high idle times, which mitigate common barriers like range anxiety and parking uncertainty. The core problem is to determine the economic and technical feasibility of utilizing these often-idling assets to provide grid services, optimize energy costs through price-aware charging, and reduce peak grid loads, while simultaneously sizing the necessary energy infrastructure (PV, stationary storage, and charging equipment).

Methodology
The study employs a novel two-layer simulation and optimization framework to evaluate a fleet of ten electric tractors in a rental dealership setting in Aldenhoven, Germany.

• Inner Layer (Operational Optimization): This layer simulates the daily and weekly operation of the fleet using a Linear Programming (LP) solver within MATLAB. The objective is to minimize yearly electricity costs by optimizing the charging and discharging schedules of the tractors and a Battery Energy Storage System (BESS). The model incorporates:

  • Usage Profiles: Three distinct weekly rental use-cases (long-term weekly rentals, daily rentals, and weekend rentals) ensuring tractors depart with 100% State of Charge (SoC) and return with 5% SoC.
  • Energy Components: A photovoltaic (PV) system, a BESS, bidirectional Electric Vehicle Supply Equipment (EVSE), and the dealership building load.
  • Cost Function: Minimizes the sum of energy purchase costs (based on dynamic day-ahead prices) minus revenue from energy feed-in (set at 90% of the buy price to account for losses and discourage inefficient trading), subject to power limits and SoC constraints.
  • Simplifications: The model assumes 100% component efficiency, perfect knowledge of future prices and vehicle schedules (perfect foresight), and neglects component aging during the simulation.

• Outer Layer (Infrastructure Sizing via Bayesian Optimization): A Bayesian Optimization (BO) algorithm iteratively searches the multi-dimensional parameter space to find the optimal sizing of the energy infrastructure. The BO optimizes:

  • Peak PV power capacity.
  • BESS capacity.
  • Maximum grid connection power.
  • EVSE configuration (power rating and unidirectional vs. bidirectional capability) for each use-case.
  • The BO minimizes the total cost ($c_{Tot}$), which is the sum of the cumulative electricity costs from the inner layer and the capital costs of the components (based on lifetime depreciation). An error model within the BO penalizes configurations that violate the maximum grid power limit, ensuring feasible solutions.

Key Contributions

1. Novel Use Case: The paper introduces V2G integration specifically for NRMM rental fleets, highlighting the sector's suitability due to predictable scheduling and centralized charging.
2. Integrated Optimization Framework: It proposes a methodology combining an Energy Management System (EMS) for operational strategy with Bayesian Optimization for infrastructure sizing. This approach avoids the computational expense of training neural networks or exhaustive Design of Experiments (DoE) while effectively navigating complex, non-linear parameter spaces.
3. Economic Analysis of Rental Models: The study quantifies the financial potential of pooled electric NRMM fleets, demonstrating how coordinated V2G strategies can generate revenue and reduce costs across different ownership structures (rental vs. private).

Results
The simulation, conducted over 51 weeks with 100 BO iterations, yielded the following findings:

• Optimal Configuration: The BO identified a configuration consisting of a 133.9 kWp PV system, a 99.0 kWh BESS, a 69.4 kW grid connection, and 11 kW bidirectional chargers for all tractors.
• Cost Performance: The optimized system achieved an average electricity price of €0.068/kWh. Over the simulated year, the dealership realized a net profit of €695 from energy trading with the grid (revenue from selling minus cost of buying), despite the capital investment in bidirectional hardware.
• Operational Behavior: The EMS successfully utilized the tractors (specifically those in Use-Case 3) as temporary storage, charging them during low-price/PV-production hours and discharging them during peak price hours, similar to the BESS. The system maintained the required 100% SoC for all departures.
• Regulatory Caveat: The authors note that the economic viability relies on the assumption that fixed grid fees and renewable levies (which currently apply to electricity consumption but not feed-in) will be reformed. Without the removal of this "double taxation," the economic benefits of V2G could be nullified.

Significance and Claims
The paper claims that electrified NRMM is not only environmentally beneficial but also economically viable when combined with V2G and smart energy management. The study demonstrates that the predictability of rental fleets makes them ideal candidates for grid interaction, potentially offering similar benefits to electric bus fleets.

However, the authors maintain a modest tone regarding their findings. They explicitly state that the study is limited by:

• The lack of real-world operational data for electric NRMM.
• Simplified models that ignore component aging, conversion losses, and regulatory uncertainties.
• The assumption of perfect foresight regarding vehicle schedules and energy prices.

Consequently, the paper concludes that while the methodology and preliminary results are promising, further research is required to validate these findings with real-world data, refine the models for accuracy, and address the regulatory challenges—specifically the taxation of bidirectional energy flows—that currently hinder the widespread adoption of V2G in this sector.