EVI-InMotion: Electric Vehicle Infrastructure – In Motion Tool

NREL's Electric Vehicle Infrastructure – In Motion (EVI-InMotion) tool is used for planning, optimizing, and analyzing the feasibility of charging electric vehicles (EVs) while driving on electric roads.

This versatile tool is applicable to the full range of vehicle classes—light-, medium-, and heavy-duty—as well as various in-motion charging technologies—inductive, capacitive, and conductive. EVI-InMotion uses high-resolution travel data along with detailed information about existing road networks to optimize system design parameters and evaluate performance.

Key Capabilities

Optimizing dynamic charger system parameters for a given vehicle and drive cycle, with a focus on electric road coverage, charger power, charger location, charger length at each location, onboard battery size, and the number of power receivers on the EV
Evaluating the impact of dynamic chargers on EV efficiency, driving range, and charging time
Performing feasibility analyses for large-scale deployments—at the city or state level—of dynamic chargers on public roadways
Exploring the impact of dynamic chargers on light- and medium-duty EV intercity and intracity travel as well as heavy-duty EV local, regional, and long-haul travel.

How It Works

EVI-InMotion incorporates several models and algorithms.

Models

Vehicle Powertrain Model – Predicts EV battery power, energy, and state of charge while considering vehicle system parameters, driving behavior, and regenerative braking. Based on NREL’s Future Automotive Systems Technology Simulator.

Cost Model – Accounts for charger components including power converters, materials, and installation costs as well as vehicle components such as batteries and onboard electronics.

Dynamic Inductive Charger Power Model – Estimates the power an EV receives via the electric road. One segment of charger is modeled for power as a function of charger dimensions and alignment. Several segments are stacked together according to the desired electric road length.

Algorithms

Automatic Placement Algorithm – Allocates the number and location of charger segments. Drawing on road and travel data, the algorithm places the system based on existing road networks in the United States.

Automatic Search Algorithm – Solves a single- or multi-objective planning optimization problem, identifying the best combination of charger system parameters (e.g., power level, roadway coverage, location, length at each location, etc.) and vehicle system parameters (e.g., battery size and number of power receivers on the EV). Such optimization aims to minimize overall system costs—specifically, charger and vehicle component costs—while satisfying driving range and charging time requirements.

Applications

NREL developed two versions of EVI-InMotion—one for fixed-route applications and another for dynamic charging on public roadways.

Fixed-Route Applications

In fixed-route applications, such as with circular and on-demand shuttle services and transit buses, EVI-InMotion employs representative drive cycles to optimize system design.

Illustration showing two interlinked boxed areas, including an optimization layer at the top and an energy estimation layer at the bottom. The optimization layer features three connected sub-topics, including 1) evaluate fitness and constraints, 2) planning optimization algorithms, and 3) generate variables. A list under the planning optimization layer includes charge sustaining, state-of-charge limits, battery charging rate limit, and system cost model. The energy estimation layer features four connected sub-topics, including 1) travel data, 2) vehicle powertrain model, 3) charger power model, and 4) energy state-of-charge estimation. The travel data area links to the vehicle powertrain model (with speed and grade) as well as the charger power model (with route). The vehicle powertrain model and the charger power model connect to energy and state-of-charge estimation via a formula that includes “PE” at the positive (associated with the vehicle powertrain model) and “PC” at the negative (associated with the charger power model), resulting in Pbat as it connects to the energy state-of-charge area. Meanwhile, the energy state-of-charge area connects to the evaluate fitness and constraints area; and the generate variables area connects to the vehicle powertrain model (with battery capacity) as well as the charger power model (with track positions, track segments, and rated power). — Illustration showing the EVI-InMotion process flow for studying charging systems for fixed-route applications.

Dynamic Charging on Public Roadways

EVI-InMotion is also used to inform the design of large-scale dynamic charger deployments. Based on a real-world road network, EVI-InMotion draws on a large set of drive cycles to evaluate system performance. Optimized electric roadway system design is based on average energy models for vehicles and chargers along with performance and feasibility analyses for a given region.

Illustration showing connectivity between various areas, including an original road network connecting (with road class, latitude, and longitude) to an allocation algorithm. Vehicle travel data also connects (with speed, route, road grade, and annual average daily traffic) to the allocation algorithm. Meanwhile, the allocation algorithm connects to an electrified road network, which connects (with locations) to a charger power model in an energy estimation layer. The energy estimation layer contains three components, including 1) charger power model, 2) energy state-of-charge estimation, and 3) vehicle powertrain model. The energy state-of-charge estimation area connects to the management and optimization algorithm, which connects to three areas, including the vehicle powertrain model (with battery capacity, charger power model (with length and rated power), and the allocation algorithm (with roadway coverage, number of dynamic/wireless power transfer per 300 miles). — Illustration showing the EVI-InMotion process flow for studying large-scale charging systems for primary roadways in a city or state.

Implementing high-power, low-roadway-coverage dynamic wireless charging on primary roadways could potentially provide charge-sustaining operation for intercity EV travel at less cost than low-power, full-roadway-coverage charging. Using EVI-InMotion, researchers investigated the optimal characteristics of dynamic chargers (e.g., power level, roadway coverage, and locations) and EVs (e.g., battery size and number of receivers) for the Atlanta metro area. Results indicate that high-power, low-roadway-coverage dynamic wireless charging has the potential to enable charge-sustaining operation for light- and heavy-duty EVs while reducing battery capacity and stationary charging infrastructure requirements. Text version

Publications

The following publications provide detailed information about using EVI-InMotion for planning, optimizing, and analyzing the feasibility of electric roads. View NREL’s full collection of related publications.

In-Route Inductive Versus Stationary Conductive Charging for Shared Automated Electric Vehicles: A University Shuttle Service, Applied Energy (2021)

Planning Optimization for Inductively Charged On-Demand Automated Electric Shuttles Project at Greenville, South Carolina, IEEE Transactions on Industry Applications (2020)

Planning of In-Motion Electric Vehicle Charging on Freeways, IEEE Smart Grid Newsletter (2020)

SMART Mobility Advanced Fueling Infrastructure Capstone Report, U.S. Department of Energy Office of Energy Efficiency and Renewable Energy Technical Report (2020)

System Design and Optimization of In-Route Wireless Charging Infrastructure for Shared Automated Electric Vehicles, IEEE Access (2019)