In a WSJ article on the Waxman-Markey bill (discussing the incentives for bringing more renewable power and energy efficiency online), there was a throw-away line at the end that spoke volumes:
This new product research by Whirlpool is most likely referring to that corner of the smart grid that involves demand response—the ability of utilities to turn off (and on) energy loads at the appliance level in an effort to manage their own aggregated demand.
Peak demand, as it's known, is a real problem. The daily demand for power fluctuates considerably, with peak demand usually coming in the late afternoon and early evening when the lights, air-conditioning of commercial buildings are still operating, the motors of manufacturing facilities, and yet people are returning home to turn on their own lights and AC. The combined demand forces utilities and power providers to fire up older, more inefficient power sources and at times pay exorbitant prices for power that they can't pass along to their consumers.
Peak demand has been a problem from the beginning of the electric age. In fact, one of Edison's contributions was to create a light bulb with such high resistance that the user could turn it on or off without blowing out the other bulbs on the circuit. And as soon as Edison Electric brought the first utility-scale networks on line, they found that aggregated demand forced them to build the capacity to provide enough power to meet peak demand, and then not use it the rest of the time. Interestingly, peak demand came in the first few weeks of January, when lighting was needed. Not until the advent of air-conditioning did peak move to the summer.
But I digress. The interesting point of this throw-away line about Whirlpool's visionary product research is that it captures a real, if seemingly small, challenge to our best laid plans.
The demand response technologies of a smart grid promise utilities the ability to manage demand at the consumer level, turning off dryers in the afternoon and turning them on again at 3am, when the utilities are begging for demand. They can already turn smart thermostats up a few degrees to offset, and the entire electric vehicle effort is predicated on the notion that these drivers will come home and not plug in their cars right before turning on their lights and lowering the AC back down a few degrees to its original level. Or, more accurately, plug in their cars but let the utilities decide when to charge them (and maybe, at this peak time of the day, to draw the batteries down further to offset others' demand (something known as vehicle-to-grid energy).
For want of a nail
At an aggregate level, this demand response enable utilities to be both more carbon-efficient, by taking offline their inefficient power generators, and more capital-efficient, by optimizing the use of what's left. But many things that would appear to work at the aggregate level somehow get snagged on the small details of everyday life.
Whirlpool recognizes that if you put wet clothes in the dryer in the afternoon and then don't dry them for 12 hours, they start to smell and get wrinkles. I know this from personal, if unintended, experimentation. In other words, people will be forced to choose between global warming and wrinkles.
I am also not sure how many people will willingly give utilities control over the charge-status of their electric cars. Imagine deciding to go out at night and finding your battery lower than when you came home. Actually, just imagine imagining that possibility, and ask how many people in that situation would give up control of recharging their car.
One response to this nagging question of market acceptance is that smart meters (another piece of the smart grid vision) will enable the utilities to charge consumers based on their time of use. Consumers would pay more for drying their clothes in the afternoon, and less at night: the invisible hand of the market will change consumer behavior.
This is a nice thought if only it was borne out in consumer behavior, energy costs in the US are still so cheap that fluctuations in pricing have little effect on demand. Only when gas nudges against $5/gallon did we see true behavioral change. What additional cost per kilowatt-hour would it take to cede control of your car's charge schedule? Assuming, of course, that people know what price per kWh they're paying now.
Whose smart now, b%$#%?
Proponents of the smart grid point to the myriad ways in which control over today's electrical production, distribution, and consumption could bring huge efficiency gains. There are two problems with this approach. The first, which I'll save for another post, is about trying to improve complex systems by making them more tightly-interconnected. The second, and related, is the notion that top-down control can actually be achieved in soci0-technical systems. And believe me, the electric grid is as much social as technical.
In other words, many aspects of the smart grid promise efficiencies by concentrating control in one set of actors. For example, the utilities get to decide when you run your dryer or recharge your car. As a people, the United States is not particularly good at handing over that control--indeed, a very similar debate rages around the internet.
Right now, that internet is a very dumb grid--packets of information go in one corner and come out the other, whether they are video, email, websites, credit-card transactions, what have you. The TelCos, among others, would like to prioritize the packets--for example, more important messages would get through while others would get side-tracked. Side-tracked, of course, referring to those low-priority trains that are pushed to the side to let the high-priority ones keep to schedule. In California, the passenger trains are side-tracked to the freight trains, with the result that commuting by train requires planning for up to an hour delay on a 2 hour trip. But now I'm side-tracking my point.
The point being, people are not keen on the idea that their internet usage is somehow less important then their neighbor's. The Telcos envision a world in which they are more than happy to oblige by offering you premium (higher priority) service for a premium price. Since the Telcos are only operating the last mile of the internet connection to your home, where this priority is not as critical to the smooth operation of the entire network, this is a bit of a challenge to accept. Net neutrality is the term used to argue that we should avoid allowing some corporate actors to gain control over the internet network
Right now the electric grid is dumb. Making it smart refers, in many ways, to giving control of this dumb grid to somebody who is supposedly smarter than others and will determine the best ways in which to route energy, to recharge cars, and to dry clothes. When it comes to re-routing around a faulty transformer, this is good (and the transformer won't complain). When it comes to re-routing your daily routines, this may not fly so well.
The smart grid vision borders dangerously on techno-utopianism. To work, it requires people to dramatically change their behaviors, which is something people do not do well. In fact, it brings to mind one of my advisor, Bob Sutton's, favorite quotes:
- Finnerty (Kurt Vonnegut, Piano Player, p332)
This isn't to say we should not try to change consumer behavior. On the contrary, we need to understand it so that our investments—and investments in the smart grid are massive—are effective. Right now, the policies around energy and climate change are driven by engineers and scientists on one side and economists on the other. Neither side has variable in their models accounting for wrinkled clothes--but these are just the things that get tangled up in the machinery.
Giacomo 'Peldi' Guilizzoni is the Founder of 
...analyzed 54 prominent studies that used functional magnetic resonance imaging, or fMRI, to identify the brain areas associated with specific emotions — and found that 28 of them used the statistical methods and filters that were most likely to yield data that would confirm the researchers’ hypotheses. According to Vul, this cherrypicking may have been inadvertent; nevertheless, it produced correlations that “exceed what is statistically possible” and are almost certainly false. (from 
