One of the distinctive characteristics of the Smart Grid is the notion of communication between the utilities and consumers, creating (ideally) a beneficial feedback loop of virtuous behavior, in which the utilities get consumers to help balance loads by choosing not to overtax the system, and the consumers get price breaks for lessening their energy use at peak times. The challenge is to figure out what kind of information will get consumers to play along and what kind of information flows will let this dynamic system function effectively without becoming unstable.
Researchers from MIT's Laboratory for Information and Decision Systems looked at this problem and concluded that if too many people set energy-hungry appliances/tasks to occur based on the price of electricity, then this could result in a huge (and unwelcome) spike in demand. But, as they outlined in their paper presented to the IEEE Conference on Decision and Control, utilities could avert this particular type of chaos by using fairly simple price controls. (To read the entire paper in all its mathy glory, read the MIT News article, "The too-smart-for-its-own-good grid" and click on the link.)
Meanwhile, EPRI published a preliminary report on the Commonwealth Edison smart meter pilot project (the full report is available at the Smart Grid Information Clearinghouse) where they basically tested out how consumers electricity consumption was affected by combinations of dynamic pricing rates, incentives, and enabling technologies. A report on Phase 2 is due out in fall. People being the contrary beasts that we are, the pilot is part of an effort to figure out how to get demand response to work. What kind of information and technologies will get consumers to alter their electricity consumption habits as desired? The report makes interesting if dry reading: the number of participants who actually change their behavior in response to the various tested approaches is low, but the handful of motivated participants who were exposed to the dynamic pricing projects did reduce their load by significant amounts. And, of the approaches tested, the highest number of respondents (8.7%) was among those using real-time pricing with day-ahead notifications, next highest at 6.7% were those using critical-peak pricing rates, and third highest at 4.9% was respondents exposed to peak-time rebate rates.
I am easily amused, so I find a kind of wry humor in the fact that, in dealing with such a highly complex system as the Smart Grid, it's humanity that's throwing a wrench in the works. It's only fitting, I suppose, since humanity is the reason that we need the Smart Grid in the first place!