Project 21:
Effect of Electric Vehicles on Power System Expansion and Operation

Objective

Examine the effects of electric vehicles on electric power system design and operation. This work includes using an existing Hawaii developed model that will be validated against an established utility-scale model. The work will evaluate the benefits of optimally-timed EV charging and the requirements and costs of electric grid infrastructure to serve different types of vehicle fleets. It will also assist in evaluating the effects on battery life of battery duty cycles used in grid-to-vehicle and vehicle-to-grid applications.

Brief Description

This project will examine the effects of electric vehicles (EVs) on Hawaii's electricity systems and their operation and expansion. The tasks include four interrelated parts: (1) benchmarking an open-source power system planning model previously developed by University of Hawaii researchers against an industry-standard production cost model; (2) evaluating the benefits of scheduling EV charging at optimal times each day; (3) calculating the technical requirements and costs of electric grid infrastructure to serve different types of vehicle fleets; and (4) estimating battery duty cycles for grid-to-vehicle and vehicle-to-grid applications.

Research Results

Implementation of Research

Part 1. SWITCH is an open-source power system planning model previously developed by researchers at the University of Hawaii. It consists of a lightweight, hourly production cost model nested within a multi-decade capacity expansion model. SWITCH is designed to choose optimal plans for expansion and operation of power systems based on the hourly behavior of renewable resources, demand response, storage and conventional power plants. This work will benchmark SWITCH's production-cost modeling capabilities against results from GE MAPS, an industry-standard production-cost model, when considering a future Hawaii power system with large shares of renewable power and electric vehicles.

Part 2. The second part of this work will use both the capacity planning and production cost capabilities of SWITCH to calculate the economic benefits of charging EVs at optimal times. These benefits include obtaining a larger share of power from renewables and high-efficiency baseload generators, and reducing the need for new generation, transmission and storage capacity.

Part 3. The third part of this work will compare the effects of three different kinds of vehicle fleet on the power system: primarily traditional internal combustion engine vehicles (ICE), primarily pure-electric vehicles (EVs) or primarily plug-in hybrid electric vehicles. For each type of fleet, the research will use SWITCH to identify the grid infrastructure needed, the average cost of power for non-vehicle uses, and the average cost of providing transport. Various effects are possible and will be included in this assessment: for example, the extra load from EVs and PHEVs may force the system to adopt lower-quality and higher-cost renewable resources, raising costs compared to the ICE case; alternatively, well-timed charging of EVs and PHEVs could help with integrating large-scale renewable energy, reducing costs. The limited all-electric range of PHEVs may mean they shift less transport energy onto renewable sources than EVs do; on the other hand, PHEVs may reduce the need to build conventional power plants, since they can revert to gasoline-only mode on days when wind and solar power are scarce.

Part 4. The fourth part of this work will report duty-cycles for vehicle batteries used for inter-hour load shifting (charging at optimal times during their plugged-in period). This work will be an input for battery degradation analysis reported in other projects.

Completed Work

Part 1 - Benchmarking SWITCH vs. GE MAPS

• Populated the SWITCH model with Hawaii-specific data
• Calculated hourly production profiles for existing and potential renewable power projects using data from the Oahu Wind Integration and Transmission Study
• estimated roof area available for installing solar photovoltaic (PV) systems in each census block (based on Google Maps data), and calculated population-weighted hourly output from distributed PV systems,
• identified land available for central-station solar power plants throughout Oahu and create profiles of potential projects (cost, size and hourly production)
• identiied individual sites for potential wind turbines, cluster wind turbines into wind resource areas and calculate hourly production for each area
• Tabulated heat rate and fuel data for existing and potential thermal power plants (low-sulfur fuel-oil, distillate, LNG, biodiesel, waste-to-energy and coal; based on HECO IRP documents), based on HECO IRP filings
• Prepared fuel cost trajectories based on HECO IRP documents
• Prepared future hourly load profiles based on HECO FERC Form 714 filings and IRP documents
• Prepared a profile of a generic grid-connected storage resource based on sodium-sulfur batteries
• Completed initial verification that SWITCH performs as expected when the cost of fuels or portfolio of technologies is adjusted.
• Integrated more detailed data on individual thermal power plants
• Rewrote SWITCH in Python language using open source modeling and solver frameworks; published new version for free access by other users
• used SWITCH to reproduce analysis previously done with GE MAPS
• compared results between SWITCH and GE MAPS

Part 2 - Economic benefits of optimizing timing of EV charging

• Model technological options for future Hawaii power system
• integrate HECO's near-term plans for changes to the power system into SWITCH base case (LNG conversion and adding quick-ramping generators)
• add spinning reserve requirements to SWITCH as a function of the amount of wind and solar power in operation
• Develop EV charging profiles
• develop profiles of EV charging availability, energy requirements and feasible penetration of EVs, based on National Household Travel Survey (2009);
• use EV profiles to develop business-as-usual charging scenarios for EV-dominated or PHEV-dominated fleets, assuming vehicles recharge to 100% as soon as they are parked at their main charging site
• Add iterative solution method to SWITCH for high-complexity demand systems

Future Work

Part 1 - Benchmarking SWITCH vs. GE MAPS

• report benchmarking results

Part 2 - Economic benefits of optimizing timing of EV charging

• Run SWITCH with EV's charged under business-as-usual or optimal control by system operator; compare optimal design and operation of power system, and the resulting costs

Part 3 - Comparing the cost of ICE and EV fleets

• Run SWITCH with fleets dominated by EVs or ICEs. Compare optimal design and operation of power system, and the resulting costs. These costs include power system upgrades and operating costs. Compare vehicle operating costs.

Part 4 - Duty cycles for EVs providing grid support

• Use SWITCH to identify duty cycles for EVs when providing hour-to-hour demand response. Provide the demand-response profiles to other members of the team for degradation analysis.

Impacts/Benefits of Implementation

The increasingly popular SWITCH power system planning model has been rewritten to eliminate any dependencies on proprietary frameworks. SWITCH is now available as free and open-source software for all users. This allows any user to conduct state-of-the-art research on electric power system expansion and adaptation, based on hourly behavior of EVs and renewable energy.

Data for the SWITCH-Hawaii version of the model are also available for any users who want to study integration of EVs and renewable energy in Hawaii. Some members of the University of Hawaii Dept. of Economics are now using it to study the benefits of dynamic electricity pricing in Hawaii, working with Dr. Fripp.

Numerous additional capabilities have been added to the SWITCH modeling framework (spinning reserves; part-load heat rates; fuel markets; battery storage; modeling of arbitrary, high-complexity demand functions).

SWITCH has been used to provide technically grounded guidance to the Hawaiian Electric Company, regulators and other stakeholders as they make plans to meet the state's new 100% renewable electricity target by 2045.

Novel techniques have been developed to represent the charging requirements and flexibility of the EV fleet, based on first principles and nationally representative transportation surveys. Profiles of business-as-usual charging and fleet flexibility are now available for other users. Additional techniques have been developed to integrate these exceptionally rich profiles efficiently into power system production cost models and capacity expansion models.

Reports

M. Fripp; Development of SWITCH-Hawaii Model: Loads and Renewable Resources, Report Number: HI-13-16, August 2016.

P. Das, M. Fripp; Savings and Peak Reduction Due to Optimally-Timed Charging of Electric Vehicles on the Oahu Power System, Report Number: HI-14-17, Dec 2015

M. Fripp; Intermodel Comparison Between Switch 2.0 and GE MAPS: Evaluating a New Tool for Integrated Modeling of Electric Vehicles and High-Renewable Power Systems, Report Number: HI-20-17, July 2018.