|Fig. 1: Load management involves smoothing out energy demand so it can be more easily met. This is done via peak clipping, load shifting, and valley filling.|
Electric load management, which is often called simply load management, refers to the systems in place that match electricity supplies with demand. A steady supply of power is generally quite straightforward to produce with a typical coal, gas or nuclear plant - simply fire up the generator and make sure you have a steady supply of fuel. However, the demand is not steady: there is more demand during dinnertime, for example, and on hot afternoons when the air conditioners are on. A power company must be able to supply power at all times, however, so they are motivated to shift large electrical loads from high-demand peak times to low-demand off-peak times. The means by which they do this are collectively called load management. Load management generally falls into one of three categories: load clipping, load shifting, and valley filling, which are shown in Fig. 1.  Most of the strategies described here are load clipping or load shifting strategies. Practically speaking, this usually means raising energy prices during peak usage times and on high-volume users. Raising rates during peak hours is possible because peak usage is quite predictable, with modern predictions typically within 1% .
Fig. 2 shows how this comes into play in practice. Demand is low at night, as are rates. Rates increase during the day as more power is demanded; there is a large difference between the lowest-demand night hours and the highest-demand afternoon hours.
For better or for worse, then, much of load management comes down to economics. As a consumer, behaviors like turning the furnace down at night and turning off the air conditioner for an hour during a hot afternoon are economically encouraged. Power companies also sometimes have programs in place in which equipment like water heaters are put under the direct control of the utility, to turn off during particularly high peak times. 
|Fig. 2: Energy use follows a regular 24-hour cycle that can be well predicted. Energy cost is highest during peak afternoon hours; the predicted and actual energy use curves show that usage is relatively steady throughout the afternoon. (Made from data from  and .)|
In the discussion above, I mention is that it is in the power company's interest for demand to be as steady as possible. The reason for this is because the steadiest power supplies are also the cheapest--for example, natural gas is a relatively cheap source of power but flows out of the ground at a steady rate, making it a good source of baseline power but not appropriate for peaking demand.  Demand above the base levels is met by using less constant (and more expensive, on a per-joule basis) supplies, like higher-cost fossil fuels, and peaking thermal and hydro power. The peaking thermal power is especially interesting because that often refers to solar power, which is high on hot sunny days, when air conditioners swell the demand. That means that thermal solar power is especially nice for certain types of load peaking, because supply and demand move somewhat together. 
Knowing this, Fig. 2 becomes more interesting to analyze. There is a large swing in total demand between the night and the afternoon hours, and the afternoon plateaus at a high rate. Since power companies are generally quite good at predicting demand on a day-to-day basis, they can plan for this swing and meet the afternoon peak with relatively inexpensive power sources because of the plateau structure. Other demand curves sometimes show shorter peaks on top of the plateau, which would presumably be difficult to meet because they are short-term and the most flexible peak power sources are also generally the most expensive. What Figure 2 does not show is a constant power demand; although a constant demand makes for the cheapest power, it remains true that most activity (and power demand) is during the day.
Electric load management is a complex subject for several reasons: the technology is not trivial, economic complexities strongly influence the overall picture, and much of the data comes from sources (power companies) that are out to make money, not necessarily to elucidate their business strategies. At the same time, one cannot understand the power generation and distribution system without considering fluctuations in demand, and the interplay between supply and demand.
© Katie Malone. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 C. W. Gellings and K. E. Parmenter, "Demand-Side Management," in Energy Management and Conservation Handbook, ed. by F. Kreith and D. Y Goswami (CRC Press, 2008).
 A. J. Pansini and K. D. Smalling, Guide to Electric Load Management, (Pennwell Books, 1998).
 I. K. Yu and Y. H. Song, "Load and Price Forecasting via Wavelet Transform and Neural Networks" in Operation of Market-oriented Power Systems, ed. by Y. H. Song and X. F. Wang (Springer, 2003).
S. Doty and W. C. Turner, Energy Management Handbook (Fairmont Press, 2009).
 O. I. Elgerd and P. D. Van der Puije, Electric Power Engineering, 2nd Ed. (Chapman and Hall, 1998).
 S. A. Kalogirou, Solar Energy Engineering: Processes and Systems, (Academic Press, 2009).