Rachel Maddow, a kindred spirit whose heart beats a little faster at the word “infrastructure,” has been campaigning recently for more infrastructure spending in the stimulus package. Pointing to the mass blackouts caused by Midwest storms, she asked the other day on her MSNBC show, "Can I put in a request for a grid that works, even in the snow?"

Yes, Rachel, you can! What you want is a smart grid rich in distributed energy resources.

First, it is important to be clear that we have two power grids: a transmission grid, which consists of the big lines carrying power from distant generating stations, and a distribution grid, which carries power in the local area to homes, businesses, etc. Failures on the transmission grid, that’s T to us geeks, lead to the really big blackouts like that in the Northeast in August 2003. But most failures — around 90 percent — happen on the distribution, or D grid, and they are usually not well publicized.

Electric Power Research Institute estimates that, overall, blackouts and other power disturbances cost the U.S. economy in the range of $119-188 billion (see p. ES-3 [PDF].) By comparison U.S. power customers paid a total of $343.7 billion for electricity in 2007.

The shocking fact is that the costs of an aging and technologically backward power grid adds something like one-third to one-half to our annual electricity costs. Ghost Town Louisville is a poster child, but most power problems do not receive national publicity.

You would think that this level of costs would be driving the utility sector to invest in advanced technologies that would make the grid more resilient in the face of natural and human-caused disturbances. (Good thing Al Qaeda hasn’t hit a big substation or T line yet. We are highly vulnerable.) But you would be mistaken because the utility sector is probably the world’s worst mesh of capitalism and socialism.

Here’s how it works. The utility comes to the state utilities commission and states its expected costs. After the inevitable arguments, it then receives cost recovery though the rates — a guaranteed rate of return. The problem has been this system favors the status quo. Commissions are hesitant to approve new technologies, and utilities are hesitant to move into them. It’s a kind of comfortable, codependent relationship that promotes more of the same.

Another problem facing the utility sector, as in many industries, is an aging workforce. Working for a power utility has not exactly been the high tech glamor job attracting the best and brightest. So there are simply going to be less people to do the work — fewer linemen who can fix the problems. I can’t say that is what is happening now in the icy regions, but it sure is a proxy for what is going to happen if we don’t amp up emphasis on recruitment and job training. Since we need to rebuild the grid for efficient, clean operations, these are the green jobs.

That was a long set-up to get at your question, Rachel Maddow. Now here are some answers:

1. On today’s dumb grid, power outages must be physically discovered by line crews. This is not the Internet where problems are automatically detected and flows of information are automatically routed around trouble spots. The kind of sensor and automation technology that could do this is only slowly coming to the grid. Literally, this is about putting an automated communications and control backbone on top of the grid, which can detect problems and switch power to another route once a disruption is detected. This exists to some extent on T grids, but very little has been implemented on D grids.

2. Grids are going to get hit by weather. Lines are going to go down. But if you have a grid rich in distributed energy resources, it has much more ride-through capability. What are those resources?

Local power plants — This can consist of solar on roofs, combined heat and power plants at local institutions such as business parks or college campuses, or the existing local plants. Small-scale wind turbines that could mesh into urban and suburban environments are also on the way. Today’s power grid is essentially one-way. With smart communications and control systems, power from local sources can be more easily integrated into the grid, in a more Internet-like fashion. This is where we want to go on a standard basis, and this kind of energy network would be more likely to hold up under extreme conditions.

Microgrids — With those local power plants you might already have what is essentially a micro grid distributing power to a neighborhood, campus, business park, etc. So even if the bulk grid is out, your local grid is still on. Some people see the eventual smart grid as a confederation of microgrids linked by the bulk grid.

Plug-in vehicles running vehicle-to-grid (V2G) — It is generally not known that over 90 percent of potential U.S. energy generating capacity is under the hoods of vehicles. It takes a lot of energy to move around our steel cocoons. But vehicles stay parked generally around 90 percent of the time. As we move to plug-ins, that means we are going to have massive batteries parked in our garages capable of running our homes for several days. Many believe we will use that capacity to store electricity from variable resources like the sun and wind and feed it back into the grid with V2G. That will require smart vehicles connected to a smart grid. In any event, smart charging systems that coordinate plug-in charging with the grid are in development. This is an absolute necessity to make sure plug-ins are not charging during peak hours when it’s not absolutely necessary. The unintended consequence we want to avoid is plug-ins that drive up demand at peak hours, and thus increase emissions from fossil-fuel driven peaker plants.

Smart appliances — Also in development, and successfully tested in the Pacific Northwest, these home appliances have a built-in intelligence that provides them with grid awareness. When they detect a problem they can automatically cycle down, for instance the heating elements in your clothes dryer or water heater shut down for a few minutes, or your refrigerator lets the temperature go up. No food will be spoiled, and you probably will not notice that your dryer has extended its cycle for a few minutes or there is slightly less water to take a shower. But the effect of thousands or millions of small adjustments is to provide “shock absorbers” for the grid. That makes it easier for the grid to ride through problems.

Of course the green energy stimulus package has the laser-beam focus of everyone in the smart grid community. There is $11 billion in President Barack Obama’s package for smart grid. $10 billion is designated for T, and $1 billion will go to D. While I cannot argue with the need for more T to carry resources from our Persian Gulf of wind, running from Montana to Texas to the coasts, there are dangers that it will also transmit more coal-fired electricity. From a climate standpoint this is a perverse result. So we need some “green wires” restrictions on coal transmission. In addition, we need to make sure the D investments provide a migration path to the genuine two-way local D grid.

Druthers, I would have been happier with a more even balance between T and D in the package, but this is only a down-payment on the larger project of full-scale smart grid deployment. From the power disturbance cost figures cited above, it’s easy to argue that this will pay back rapidly and magnificently.

More on this at my new blog, New Energy Nexus.