by Russell Lefevre, IEEE Fellow
A recent report by The Shpigler Group released by the Utilities Telecom Council indicates energy usage attributable to electric vehicles (EVs) could rise 1,700 percent by 2030.
Although studies indicate an increase of this amount will not necessarily impact the generation and transmission elements of the electric grid severely, serious difficulties will be introduced on the distribution element, especially on pole- and pad-mounted transformers that deliver electricity to users.
As many IEEE researchers and others have shown, this will put a premium on energy efficiency in charging EV batteries. The utility industry is aware of the issue and is developing approaches to meet the demand.
A transformer steps down the voltage from the transmission element to the normal household voltage, which is 120 volts. A transformer typically services a few households in a neighborhood. EV sales will lead to geographic clusters of environmentally conscious, affluent customers. Thus, local transformers servicing these neighborhoods will feel the impact. This indicates results will be scenario-dependent.
Saifur Rahman, an IEEE Fellow and editor-in-chief of “IEEE Transactions on Sustainable Energy,” and his colleagues have examined several scenarios. One shows the impact on a typical 25-kVA transformer servicing three households. Two EVs charging at 6 p.m. can stress the transformer, Rahman said. Adding a third EV results in transformer overload, and adding a clothes dryer shows severe overload.
Southern California Edison (SCE)’s watch list includes star-studded Santa Monica, where the probability of more than one EV being serviced by a transformer is high. As sales of EVs become more common, SCE will observe neighborhoods where electricity demand shows a significant increase to determine the necessity to upgrade or replace transformers.
The Rahman simulations indicate potential problems because of overloads during a 24-hour span. Matthew J. Rutherford and Vahid Yousefzadeh studied the loss of life of a distribution transformer under different load conditions. The results indicate staggering the battery-charging profile of EVs in a neighborhood can improve significantly the loss of life compared with simultaneous off-peak charging. Uncontrolled charging could add to maintenance requirements and likely lead to early transformer replacement. Hilshey and others estimated the acceleration of transformer aging because of EV charging. Their results showed the loss of life as a function of the number of EVs on the transformer and the ambient temperature. A significant increase in loss of life was demonstrated for six vehicles on a 15-kVA transformer for Los Angeles and Burlington, Vt., with the Los Angeles loss much larger because of the higher ambient temperature. Rahman and Rutherford show that staggering the EV battery-charging profile in a neighborhood results in significant improvement in transformer loading.
A potential problem could arise as electricity rates designed to discourage charging during the daytime may result in a night-charging challenge. Most utilities employ transformers designed to cool overnight. With the potential increased load, sustained excess current eventually could cook a transformer’s copper wiring, cause a short and black out the local loads it serves.
The impact of EVs on the grid has been studied for several countries expected to have significant EVs deployed soon. Studies of the French and Belgian electric grids have led to similar results. These and other studies show utilities must address concerns, especially about their distribution systems.
The EV Charging Infrastructure USA conference held early 2011 in San Francisco indicates the seriousness to which the utility industry assigns these issues. The purpose was to evaluate strategies, technologies and regulatory frameworks for developing commercially viable EV charging infrastructure—multistakeholder business models for making EV charging profitable. During the conference U.S. utilities collaborated with stakeholders to find solutions to make EV charging infrastructure commercially viable. The conference addressed business models, impact on the grid, infrastructure upgrades and in-home charging infrastructure.
Speakers from at least 12 utilities examined from a business perspective the issues researchers had indicated. Especially pertinent was a session that included Pacific Gas and Electric Co. (PG&E) Director of Integrated Demand Side Management Saul Zambrano called “Understanding What Impact Electric Vehicles Will Have on Grid Systems, Power Distribution & Load Profiles to Determine What Can be Done to Manage Their Impact.”
PG&E assumes customers will prefer to use 240V/30A charging to shorten recharge times. This corresponds to the Society of Automotive Engineers International’s (SAE’s) Standard Level 2, a North American standard for electrical connectors for EVs maintained by the SAE. When this level of charging is employed, it is comparable to an average peak summer load of a single home, for example, in Fresno, Calif. The impact is the same as if a new home were placed onto an existing transformer. The peak load will occur between 5 and 7 p.m. The assumptions leading to this statement have been used by researchers who expect EV owners to plug in their chargers when they arrive home after a commute.
The PG&E study looked at the likelihood of having to upgrade transformers within a given time. The metric was the percent of time an upgrade is needed in the first one to five years. The study noted a significant difference between air conditioning areas and nonair conditioning areas. For Berkeley, Calif., a nonair conditioning area, the percentage is 11 percent for off peak and 54 percent for on peak. For Fresno, an air conditioning area, the percentage is zero percent for off peak and 57 percent for on peak. Nonair conditioning areas will need upgrades at peak and off peak, and air conditioning areas will need upgrades at peak.
The next part of the study examined how to alleviate the stress at the substation. Time-of-use (TOU) rates alone will be insufficient. Even with TOU rates and not allowing charging at peak times, the stress is still too high. Some form of demand response is needed. This implies that the smart grid implementation will be required to avoid major expense to upgrade the transformers and other substation equipment.
The Sacramento Municipal Utility District reached similar conclusions. Its summary also noted the problems associated with vehicle clustering and that smart charging can be a significant tool to alleviate the impact. Nevertheless, more research is needed to develop optimal solutions.
RWE Power, Europe’s third-leading power utility and headquartered in Germany, has determined electric mobility is coming with enormous speed. Its studies indicate simplified solutions will achieve cost savings in the short term and might slow development. The preferred option is to deploy smart charging from the start. This way, smart grid will grow with e-mobility and will reduce stress on the grid.
Increasing EV adoption presents opportunities and threats for properly sustaining energy efficiency in the smart grid. Utility companies, as well as all EV ecosystem players, must continue working together to advance EV infrastructure challenges.
Russell Lefevre, an IEEE Fellow and recognized IEEE smart grid technical expert, has a Bachelor of Science and Master of Science in Physics from the University of North Dakota and a doctorate in electrical engineering from the University of California, Santa Barbara. He is chairman of the IEEE Steering Committee on Electric Vehicles, an adjunct professor of physics and electrical engineering at the University of North Dakota, a fellow of the American Association for the Advancement of Science (AAAS) and past president of IEEE-USA and the IEEE Aerospace and Electronic Systems Society.