Storage: An Indispensable Ingredient in Future Energy
- Written by Francesco A. Amoroso and Gregorio Cappuccino
Energy storage can contribute to the smart grid by facilitating integration of renewable sources and provision of important ancillary services. At the same time, energy storage cannot really be exploited to its fullest potential without a smart grid infrastructure capable of managing bidirectional energy flows. Because of that reciprocal relationship, energy storage and the smart grid represent an indispensable synergy.
It is generally appreciated that electrical energy storage systems can greatly enhance the efficiency and usefulness of intermittent energy sources, such as solar and wind energy, whether or not turbines and solar panels are on or off the grid. Equally important, but perhaps less widely recognized, storage can also be leveraged to furnish ancillary services for the distribution grid, such as frequency regulation, spinning reserves and voltage control—services that will be needed more crucially than ever as the smart grid becomes a widespread reality.
It is hardly surprising that energy storage systems are receiving increased attention nowadays not only at grid scale but also at the commercial and residential levels. Community energy storage systems consisting of relatively small packages of batteries (typically around 25 kWh) are cropping up in neighborhoods and will become increasingly commonplace. The batteries in electric vehicles (EVs) and hybrid-electric cars represent another form of distributed storage with big scope for expansion. As EVs are more widely adopted because of concerns about greenhouse gases and oil dependence, energy stored in their batteries can be injected into the grid as needed to perform the aforementioned ancillary services.
For example, consider a 100-vehicle fleet of a small community with an average energy capacity of 20 KWh, in which 30 percent of usable charge remains in the battery pack after a day of usage. Using a conservative assumption that approximately 40 percent of vehicles are not available to perform vehicle-to-grid services at any given time, the total available energy globally will be 360 KWh. This is equivalent to the aggregate power of more than 1MW for a 15 minute time window. The energy reserve would offer a valuable addition to peak power capacity of the grid. Even more importantly, the energy reserve would be well-suited for grid regulation due to the possibility of a real-time response to power commands regulated by utilities.
Thus, energy storage could be a very important resource for the actual implementation of a smart grid. But for this to be realized, grids need to have the specific smarts in order to manage energy storage systems effectively.
Consider the expected impact on the distribution grid from charging an ever increasing number of car batteries. Typical EVs are equipped with batteries whose energy capacities are in the range 10-50 KWh. Battery charging will tend to be concentrated in specific periods of the day—such as when cars are parked at work during normal working hours, or at home at night—and that will put a significant additional energy load on the distribution grid. If EVs represented a quarter of a community's total vehicle fleet, peak electrical demand might be 30 percent higher, according to the Joint Research Centre of the European Commission.
Therefore, if battery chargers are managed as conventional household appliances that draw energy from the grid whenever they are plugged in, localized overload conditions may result—with degradation of service and damage to utility and customer equipment—if there is no kind of coordination. Though individual consumers might delay charging cars to economize on their personal costs, this would not ameliorate the basic dilemma of aggregate demand overtaxing the distribution grid.
One possible consequence of such overloads could be an increase of line losses. Even with just 10 percent penetration of EVs in some typical medium-voltage distribution grids, the lower limit allowed for line voltage might be breached.
Obviously, the urgency of devising countermeasures to these kinds of problems will depend on how rapidly EV use will actually increase. Currently, governments in nearly all advanced industrial countries are making transport electrification a fundamental priority though forecasts differ from country to country as to how fast EVs will penetrate their markets.
Regardless of how quickly EVs are adopted in the short term, we should start thinking now about how distribution grids need to evolve. We must figure out how grids can become capable of accommodating a large number of EVs. Once we start giving the grid smart features to manage charging, EV deployment will accelerate. Whatever form distributed electrical storage takes in the future, some kind of centralized coordination will be required to optimize energy usage and charging processes, where multiple batteries are connected to the same source—whether some part of the grid or a renewable generator.
Thus, if we want to guarantee penetration of both EVs and distributed electrical energy sources, it appears that we should move from the concept of a classical passive grid which just supplies energy to a new paradigm of a smart grid interacting with battery chargers to "modulate" charging processes.
At present, charging processes are implicitly sorted on the basis of the plug-in time of the battery and are performed at a fixed charging rate imposed by the battery charger. We should pursue the development of that system architecture into a more intelligent one that is capable of reacting in real time to system conditions. This smarter distribution grid will:
- Sort the charging processes on the basis of smart priority policies that provide an understanding of which batteries need to be charged most urgently.
- Dynamically vary the charging rate of batteries plugged into the grid (i.e. the power delivered by the grid to each battery), taking account of current grid load conditions to avoid overloads.
Our research suggests that with adoption of various smart variable-rate charging strategies a smarter grid will support at least 15 percent more charging requests compared to "dumb" fixed-rate strategies implementable with the classical grid structure. However, achieving such results will require a lot of effort in defining smart metering, communications and grid control features, as well as standardizing hardware and software components. Especially for this reason, we strongly believe that both today's engineers and engineering students should be carefully sensitized to the concept of a smart grid.