Concentration of CO2 in the Atmosphere

The Virtual Power Plant

Renewable power for 24/7 energy production

By George Harvey and Tori Wiechers

On a hot summer weekday afternoon, when factories and air conditioners are running full speed, demand for electric power can be very high. On a temperate night in the spring, demand can be very low. Demand goes through daily and seasonal variation.

The dirty little secret of conventional base-load power plants, such as conventional fossil fuel and nuclear facilities, is that they cannot make sudden or short-term changes in the output of their generators to match such changes in demand. Their boilers take so long to bring up to temperature that if they were turned down at night they would not be ready to meet peak demand loads in the following afternoon. Longer-term changes can be adjusted for by turning units off or on. If they produce more power than the grid requires during low demand times, they need some way to offload it. Otherwise something will overload, and possibly burn up.

Base-load power operates with a complicated and expensive set of back-up generation sources, in a highly variable wholesale market. If the demand exceeds what base-load plants can provide, peaking power plants are turned on, providing for the shortfall, though at a high price. If there is need to offload power, various mechanisms kick in to use it up. The wholesale market provides incentives to use electricity at low demand times by adjusting prices continuously. The lowest power prices are actually negative, and this is not uncommon. The highest are about $10,000 per megawatt-hour (MWh), which translates to about $10 per kWh, and is over a hundred times normal prices. (For an example, see bit.ly/Australian-market)

Renewable power systems add a new and exciting dimension to this mix. There are many types of them, each type with its own set of characteristics. Some are intermittent, but some of these are highly predictable. Some are variable in output, and some are as continuous as traditional base-load power plants. They include at least four different kinds of solar power, several of wind power, over six kinds of hydro power that do not require dams, different kinds of biomass and bio-digesters, and more. We have lots of options.

Adding to the complexity, we have ways of using power that can help equalize supply and demand. At low demand times a pumped storage plant can buy inexpensive electricity to pump water to the top of a mountain, and at high demand times, the plant generates electricity from the water as it is released back down the mountain. Such plants have been around for decades. But now, smart battery systems and electric vehicles can be charged when demand and prices are low. Low-priced power can also be used to produce synthetic fuels such as hydrogen that are used when power prices are high. Some buildings have ice makers that operate at low demand times, and the ice is later used for air conditioning. And there are many other ways we can shift power use to low-demand times.

All these systems can be controlled with computers operating in essence as virtual power plants. Combining information about weather conditions, resource characteristics, supply availability from a variety of sources, and load demands, computers can be used to balance the grid to eliminate wasted power generation. Changes in supply and demand can be managed in seconds or less, whereas previously, a power source that could be turned on in a quarter hour was considered fast. This greatly reduces supply and price problems.

Thousands of virtual power plants are already in use, most commonly in Germany, with the largest having “outputs” similar to a coal-burning generator. They manage supply of renewable power, making it more reliable and less expensive at the wholesale level than power from coal or nuclear plants, because it matches demand.

Leave a Reply

You can use these HTML tags

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>