Electric-drive vehicles for peak power in Japan

Abstract

Electric-drive vehicles (EDVs), whether based on batteries, engine-electric hybrid, or fuel cells, could make major contributions to the electric utility supply system. Computer-controlled power connections from parked EDVs would provide grid power from on-board storage or generators. Kempton and Letendre conclude that, in the United States, battery EDVs can be cost-effective as a source of peak power (Kempton and Letendre, 1997) or as spinning reserves (1999). This option is even better matched to urban Japan, where vehicles are typically parked throughout peak electrical demand periods. Using Ministry of International Trade and Industry (MITI) forecasts for the number of zero emission vehicles in 2010, we estimate the maximum potential power from EDVs in the Kanto region (which includes Tokyo) at 15.5 GW, 25% of Kanto's 1998 peak demand. This paper calculates the cost to provide power from five current EDVs — both battery and hybrid vehicles — and compares those costs to current purchase rates for independent power producers (IPPs) in Japan. Battery characteristics are calculated from current manufacturer-provided data as well as the California Air Resources Board (CARB) projections. Given current vehicle battery costs and current utility purchase rates, no vehicles would be cost-effective peak power resources. Given CARB projections for batteries, the Nissan Altra is cost-effective as a utility power source. Using projected IPP purchase rates for peak power and CARB battery projections, the Nissan Altra and Toyota RAV4L EV are cost-effective. The net present value to the electric grid could be near 300,000 yen ($US 2500) per vehicle. If utilities take advantage of this opportunity to purchase peak power from vehicles, it would make the electric grid more efficient, enlarge the market for EDVs, lower urban air pollution, and facilitate future introduction of renewable energy.

Introduction

Electric-drive vehicles (EDVs) have gained attention in the past few years due to growing public concerns about urban air pollution and other environmental and resource problems. In the United States, the California Air Resource Board (CARB) instituted an ambitious requirement for 10% of the new cars sold in the state to be zero-emission vehicles (ZEVs) by the year 2003, versions of which have been adopted by other states (Nadis and Mackenzie, 1993, p. 72). This boosted the development of EDVs and accordingly many auto-manufacturers have developed their own EDV models to meet this goal. EDVs include three primary types: battery-based, hybrid, and fuel cell. Battery and fuel cell vehicles are considered ZEVs, while hybrids are low-emission or ultra-low emission vehicles.

While an increase in battery-based EDVs is expected to increase electricity sales, extra generation capacity is not needed if the EDVs are recharged at times of low demand, such as overnight. In fact, as we argue here, there is a potential to reduce the peak load if EDVs are grid-connected to allow discharging of the electricity stored in their batteries, or running their on-board generators, during times of peak demand. This approach was suggested by Kempton and Letendre, 1997, Kempton and Letendre, 1999 who calculated the economic value of discharging battery electric vehicles as a peak power source and as spinning reserves. Kempton and Letendre conclude that in the US, under the right conditions, it can be cost-effective for the utility as well as for the vehicle owner. That is, the value to the utility of tapping stored electricity is greater than the total costs to the vehicle owner: two-way electrical connections, purchased energy, losses in charging and discharging, and the cost of wear from additional cycles on the battery. Kempton and Letendre (1997) also outline the design of a controller which would allow the utility to tap power when needed, limited by constraints set by the driver (for example, “I must have enough charge by 7 a.m. tomorrow to drive 20 km”).

In Japan, energy security issues are even more serious than in the US mainly due to its scarce resources. Japan has one of the lowest energy intensities (energy-use per unit of GNP) in the industrialized world, but its aggregate energy-use is still rising. While the government seeks nuclear power to solve both energy scarcity and greenhouse gas mitigation, the national debate on nuclear energy has been rapidly intensifying. The government is also engaged in several renewable energy programs, such as rooftop solar photovoltaic (PV) generation. Storage in EDVs improves the economics and performance of both nuclear and solar: Nuclear is best run at constant output, so storage helps even out the peaks and troughs of demand, while renewable energy fluctuates with sunlight or wind, so storage helps even out supply. Storage also helps solar match supply to demand peak, since the solar peak is a couple of hours earlier than the typical load peak. Therefore, EDVs would seem to be a promising way to add storage to the electric system, especially for a country like Japan, since automobiles are not used as frequently as in other industrialized countries, and since urban automobiles are typically idle through and past peak hours. Consequently, there are more vehicles available to be discharged during peaks, and each one can allow deeper discharge.

This paper applies the methods of Kempton and Letendre (1997) to evaluate the economic potential of EDVs for the Kanto region of Japan. The Kanto region houses major cities such as Tokyo, Yokohama, and Chiba. Although not analyzed here, we believe that EDVs can also be useful as peak power sources in the Kansai region (including the cities of Osaka, Kobe, and Kyoto), Chubu region (including Nagoya), and parts of other regions that have major cities with comprehensive public transportation systems as well.

The utility serving the Kanto region, the Tokyo Electric Power Company (TEPCO), has faced significant challenge in meeting its peak demand every year.¹ In fact, TEPCO's annual load factor is low, below 60% (TEPCO, 1999, p. 29), mainly due to the enormous demand for space conditioning during the summer and winter. The policies we suggest would tap EDVs to reduce peak generation need, and thus reduce the need for further investment in peak generation capacity.