Tuesday, August 24, 2010

Smart grid is not low hanging fruit.

$100 Million to install 24,000 smart meters: http://www.denverpost.com/business/ci_15847180

That is just over $4000 per electricity meter to enable communication.

For comparison, a 1000 sq ft of R60 attic insulation costs about $1500 before subsidies.

Replacing every light in your house with CFLs costs about (25 lights x $5) $125. Or prettier LEDs for $1250.

Adding a timer to your water heater so it only heats the water during the hours when you might use it is another $25.

The difference being that all of the things on this list save energy by wasting less energy, while the smartest grid in the world can only tell you how you are wasting energy.

On the other hand, the project will likely result in everyone else’s energy prices going up and raising energy prices does encourage conservation.

Wednesday, August 11, 2010

maintenance and smart grid

I read this and was blinded by rage:

http://www.cnn.com/2010/TECH/innovation/08/09/smart.grid/index.html?hpt=C1

Apparently, the smart grid would have prevented the 2003 blackout that took out the Northeast and caused $6B in damage.

I don’t know how the magic of smart grid would have prevented the utility from slashing its tree trimming budget to boost profits or how it would have prevented the SCADA system from failing to communicate that the lines were overheating or how it would have convinced the operators to listen to the operators from the neighboring utilities or how it would have created an effective regional decision-making structure to force utilities to shed load to prevent the problem from spreading, but somehow, magically, it would have. Maybe by letting the utility know that 20,000 toasters were being used in Tulsa, smart grid would have saved the day.

Outages doubled because of one reason: lack of maintenance. Anyone who says otherwise is trying to sell an expensive and largely useless smart grid.

Seriously, this is a simple problem that has been badly mishandled. And the fundamental cause is weird financial arrangements. Utilities use loans to build their system. The interest rate they pay on these loans is based on their credit rating. Their credit rating is based largely on how much they owe on outstanding loans as a fraction of what their system is worth. The value of their system is based on money spent to upgrade the system. Additionally, for profit utilities have the rate they are allowed to charge their customers based on the value of their system, so money put into the system means more profits.

So, if they get loans and use the money to improve their system it improves their credit rating, making it cheaper to get loans, which makes delivering power less expensive and/or more profitable.

Money spent on maintenance isn’t treated the same way. It is a current expense – money down the drain.

From an MBA CEO perspective, this is the entire story. Money spent on system improvements makes profits while money spent on maintenance is just gone. Only an idiot would throw money away on maintenance. So nobody does. Instead, you run equipment until it fails, then replace it using a capital improvement budget. And when you can convince your regulators to let you do it, you throw more layers of expensive useless crap like the smart grid on top of your teetering system because it is another way to make more profits. Even if it doesn’t serve any useful purpose, the utility still gets to charge their customers for installing it and extract a profit.

The solution is to let utilities treat maintenance costs the same way capital improvements costs are treated and to increase the cost of blackouts through a fine of some variety. Maybe something like a free week of service for every hour of an outage not caused by unusual weather would do the trick.

Sunday, February 28, 2010

thesis idea: geographic diversity of renewable energy

One of the common claims about renewable energy is that geographic diversity is the saving grace. The wind will always be blowing and the sun shining somewhere. And this is true - to a point.

Earlier this winter, the UK had an uncommonly strong snow storm that was accompanied by several days of cold with little wind. Their nationwide wind energy output went to virtually zero. I don't know of any statistics about their solar power, but it seems reasonable to assume that a few inches of snow would pretty effectively reduce solar energy output to near zero as well. Meanwhile, the cold weather lead to peak load conditions. If the UK relied on wind or solar power to meet any portion of the nationwide peak load, they would have had to take extraordinary measures (like brownouts or rolling blackouts) to keep the lights on. And do this at a time when doing so would cause the greatest harm to the population.

So, geographic diversity of renewable energy production has to be greater than can be found in the UK if we are going to make the power system rely on it. Barring the invention of very large scale energy storage, no amount of wind and solar energy (or smart grid or conservation) will ever be sufficient to meet basic power system reliability criteria in the UK. They will always have to have a non-intermittent power system in place that can serve 100% of the peak load independent of the weather.

The thesis paper's goal would be to determine roughly how much geographic diversity is necessary and possibly to use this as a way to question the wisdom of renewable energy development zones (like the wind energy development zones in West Texas).

It is a very important question for the future of wind and solar power. If the UK wanted to meet the goal of reducing their CO2 emissions by 80% by way of building wind and solar energy, a significant fraction of the 20% of emissions left would have to be consumed by the back-up power system that needs to be kept in place and running on standby ready for severe weather events that only happen once every 50 years.

The solution, of course, is to use excess energy production to fix CO2 into methane (or some other easily stored fuel) and to use it in natural gas peaking plants, but that is the subject for an entirely different type of thesis paper.

Saturday, February 13, 2010

sociology of climate science

A quick thought experiment: If climate scientists were infinitely intelligent, would their results be any more useful for policy?

My claim is "no."

And the reason is because of a selection bias for people who enter the field and sociological influences.

If scientists where infinity intelligent, they’d still be herd animals like the rest of us. If the herd is going in one direction, only a small portion of the population will choose to go the other way. This theoretical herd of infinitely intelligent scientists will apply their intelligence to mock them.

To my mind, the best scientists are borderline autistic (a la big bang’s Sheldon). These people can abandon their own ideas with no thought to social consequences.

And herein lies the problem with the "save the world" sciences (social work, environmental science, climate science, etc). They inadvertently select for people who want to save the world: people who are both highly socially aware and who have chosen their field because they see it as a way to bring positive change.

Being socially aware is a hindrance in science because it makes it harder to bear the scorn of the herd when you disagree with them.

Choosing a science as one's path in life based on the hope to be able to help humanity is a hindrance because it means that if it turns out that there actually isn't anything wrong (that there is no danger to save humanity from) is a form of failure to achieve one's hopes. Additionally, it'd likely mean the eventual loss of one's funding, since climate science without anthropogenic global warming is pretty boring.

So, no I don't think greater intelligence or capability would be remotely useful, but a little bit of impartiality would go a long ways.