Demand Response is a collection of mechanisms used to balance supply and demand in the Smart Grid by utilizing the varied capabilities and properties of devices attached to the electrical grid, typically owned by commercial, industrial, and residential entities. Ideally, they pose no inconvenience to their owners and operate in such a way as to not even be noticed, generating a small stream of income as compensation for their use. Other times, they represent minor inconveniences, representing a sacrifice for the continued stability of the grid as a whole. After all, consuming less electricity is better than overloading the grid, causing a blackout, and not being able to consume any electricity, in addition to infrastructural damage and the inherent dangers of a blackout.
Dynamic Demand enabled devices sense the frequency of the electricity grid, a direct representation of the balance of supply and demand in the grid. Then, when demand on the grid outpaces supply, they temporarily reduce their demand. Demand Response mechanisms go further, responding to signals from the Smart Grid and reducing, rescheduling, or increasing their demand for electricity as the situation calls for it. Together, they ease the strain put on electrical grids by both reducing peak levels of demand and mitigating the effects of surges.
Two benefits arise from these mechanisms, an immediate economic benefit and a more nuanced infrastructural benefit that itself carries additional economic benefits. The immediate economic benefit is directly tied to the avoidance of using peaking power plants during periods of peak demand. Additionally, since surges in demand require the use of peaking power plants due to their fast response times, then, if surges can be slowed down, load following power plants can be used instead. The result of a reduction in peaking power plant usage is simple – cheaper production of electricity leads to lower prices for consumers.
The simplest example of Demand Response technology is the temporary dimming of fluorescent lighting (PDF link) in stores and offices. Considering the amount of fluorescent lights drawing electricity during the daytime (20% of peak electrical loads in commercial buildings) and the simplicity with which they can be dimmed, even a slight reduction in the current provided to those circuits leads to a significant reduction in demand. Similarly, refrigerators can slightly decrease their cooling, barely raising their internal temperature (e.g., by half a degree) for an hour without negatively affecting the food stored within. Collectively, as the number of refrigerators is enormous, this noticeably decreases the load placed on the grid. The same principle applies to various other temperature control technologies such as air conditioning and heating.
Such mechanisms would only be implemented where slight changes in temperature would be, at best, unnoticeable or otherwise at most a minor inconvenience. Laboratories, computer farms, and other facilities where climate control is a major consideration would obviously not be included in this sort of Demand Response.
Additional significant benefits can be seen by rescheduling the use of various devices. Large water heaters are prime candidates, as they use large amounts of electricity and their load can be moved to off-peak hours relatively easily. It’s frequently better from an economic, environmental, and infrastructural perspective to expend more electricity during off-peak hours than a smaller amount during peak demand.
Other rescheduling options are the recharging of electronic devices, whether as small as phones or as large as plug-in electric vehicles, until later in the night rather than in the evening when many people come home from work.
Demand Response also draws upon electrical storage already present in the grid and encourages its continued development in the future. By presenting an opportunity for arbitrage between low off-peak prices at night and high prices during each day’s peak, the grid effectively combines many distributed independent operators of electrical storage and treats them as a single load balancing network. With the growing popularity of electric vehicles, each of them representing potential energy storage, the capacity of this network is set to expand dramatically.
Lastly, distributed small-scale generation can be called upon to fill the gap between supply and demand, similarly joined together into a virtual power plant by the Smart Grid.
Such a network can be used to smooth out the variability from intermittent sources of energy, increasing their viability and decreasing the overall cost of electrical production. In fact, many solar power plants already use energy storage methods to smooth out their own electrical output and this is simply applying the same idea in a more distributed manner.
In terms of continuously maintaining the balance of supply and demand, the ability to reduce demand by a certain amount is as useful as the ability to increase supply by that same amount, with the added benefit of not having to build, maintain, and run additional large power plants, leading to both infrastructural and environmental benefits.
While not always the case, decreasing peak demand is frequently advantageous from an economic, environmental, and infrastructural perspective even if doing so increases the overall demand for electricity.
Demand Response significantly affects the economics of power generation. Taking Ontario as an example, the highest demand placed on the grid in 2006 was 27,005 MW. That same year, demand rose over 25,000 MW for only 32 hours, representing less than 0.004% of the entire year. Essentially, 2,000 MW of reserve capacity was invested in, designed, built, maintained, and fueled only to be used for a total of 32 hours over the course of an entire year. Other grids throughout North America experience similar usage profiles, with only 80-100 hours accounting for roughly 8-12 percent of the maximum demand.
If demand during those hours could be lowered even slightly, significant savings would be realized from reduced reserve capacity costs (construction, maintenance) and then again from not actually utilizing that capacity (expensive fuel, low efficiency). In fact, these savings have been calculated (PDF link) to recoup a significant percentage of the initial expense of deploying the necessary technology. In addition, demand response and the infrastructure behind it reduce ongoing operating costs and improve consumer satisfaction, adding further savings to help shift the economics into the feasible range.
Overall, Demand Response, while initially expensive to implement, offers a wide variety of options to both consumers and providers of electricity. Furthermore, even if one were to ignore all of the benefits with regard to customer satisfaction and grid stability, the savings offered by Demand Response in the long run are sufficient to warrant its implementation.
Click to see an ANSI press release discussing recent developments in the Smart Grid standardization effort.
This is the fourth article in a series about Smart Grids, each of which can be read independently.
The first article is an introduction to the concept of a Smart Grid.
The second article explains the generation of electricity.
The third article covers Smart Grid features and Technology.
Also, a previously published and updated list of standards for Smart Grid Interoperability.