Nov 29: update in progress ..
“Smart grid” refers to a very large collection of technologies that would vastly increase the complexity and sophistication of the power generation and distribution system. The September 2009 issue of Power Engineering magazine quotes Brian Seal, senior project manager, power delivery and utilization, Electric Power Research Institute (EPRI) as saying: “Smart grid is a marketing term that is devoid of technical definition.”
The scope of the smart grid includes just about everything an engineer could imagine such as grid connected home solar arrays, wind farms, small hydroelectric plants, using electric cars as batteries for the grid, hundreds of thousands of electric car charging stations, complex control systems, new transmission lines and transmission corridors, smart meters in every home and a huge data network to support millions of new computer systems to monitor and control its components. The current scope is undefined and every power company has their own idea of what should be included. If this was an information technology project, monstrous budget overruns would be guaranteed, as well as fantastic profits for those implementing it and selling the hardware.
Power in North America is still quite inexpensive, making alternative energy sources uneconomical without subsidies, mandates and permission for rate increases to pass these costs back to consumers. Given the potential scale of expenditures (in the order of a trillion dollars) and the opportunity for considerable profits, it is not surprising that the power industry is excited. A quick look at opensecrets reveals a great deal of lobbying by the major players. For example, Southern Company:
Most consumers have heard of “smart meters”, and assume that they are needed to support the smart grid, or are indeed the smart grid. Other than enabling new consumer billing schemes (think cell phone bills), they are a very tiny component of the smart grid initiative. Some of the larger components include:
- Monitoring and Control: Millions of new sensors for voltage and current would be installed all over the grid — at transformers, substations, points along power lines and connected to computers that in turn make decisions (disconnecting lines, compensating for reactive loads, changing voltage taps, etc.) and/or send the data over a data network to another control system. The acronym SCADA, supervisory control and data acquisition, is often used to describe these types of control systems. Your laptop running Linux is probably far more powerful, but these systems are likely more reliable and fault tolerant.
Someone, presumably the USA or Israel, has written a worm called stuxnet to infect SCADA systems manufactured by the German company Siemens to attack Iran’s nuclear industry, we should expect the favor to be returned. From the New York Times article linked above; “It’s like a playbook,” said Ralph Langner, an independent computer security expert in Hamburg, Germany, who was among the first to decode Stuxnet. “Anyone who looks at it carefully can build something like it.” Mr. Langner is among the experts who expressed fear that the attack had legitimized a new form of industrial warfare, one to which the United States is also highly vulnerable. “
- Increased Line Capacity: Real-time monitoring of transmission lines allows a more accurate determination of maximum load before the lines heat up and sag into the trees (not a good thing). This would allow more power to be transmitted over existing power lines. Better transmission cables capable of handling larger loads are also available.
- Changing Consumer Behavior: Computerized meters in homes allow new billing schemes, and can reduce demand at peak times by either disconnecting some loads, or by terrorizing customers via power bills. The euphemism for this is “price signal“. It wouldn’t take too many $400 surcharges to teach you not to run that air conditioner during a heat wave. Reducing peak load allows constant power sources like nuclear and hydroelectric generators to provide more of the load and minimizes the cost of supplemental generators (such as gas turbines) that are more expensive to run.
- Power Routing: The bi-directional routing of power across large geographic areas requires computer system integration across many regional power companies. Presumably the direction of power flow would be dictated by the highest bidder and that price premiums will be charged rather than “poor Californians, let me help you out with that heat wave”. Enron would have loved this. You could cause congestion and then charge peak rates for it.
- New Data Networks: Vast quantities of data would be generated, requiring either extensive use of the existing cellphone networks, or entirely new, and very fast network infrastructure with coverage everywhere there is electricity.
- Integration with Electric Cars Vehicle to Grid (V2G) integration would use automobile batteries for grid support (stabilizing voltages) and storage.
This is not an exhaustive list. There are hundreds of ideas that could be “smart grid” material. There are however, a great deal of unknowns that have cannot be simulated with computer models, so utilities around the world are experimenting with various pilot projects to see what happens. For example, do people really reduce demand when hit with a “price signal”, and if so, by how much? Would electric car owners participate in plans that kept their car unused from 6pm to 6am so that it could contribute to storing wind energy at night?
Consider the following quotes from the Electric Power Research Institute, Demonstration Initiative Two Year Update:
“to better understand and demonstrate the technologies, business models, and prices required to further commercialize the concepts” – KCP&L Smart Grid Demonstration Project or
“By deploying and demonstrating integration of these technologies and applications, it will address many of the unknowns” Southern Company Smart Grid Demonstration Project.
The complete package of technologies being considered part of a national (USA) “smart grid” has been estimated by industry groups to cost between 300 and 400 billion dollars with an uncertainty of +/- 100 billion. ($338 to $476). This is what is called a SWAG (scientific wild assed guess) and the real cost of most highly complex projects far exceeds the initial estimates. Take a look at this GAO report on NASA budget overruns as an example:
“GAO assessed 18 NASA projects with a combined life-cycle cost of more than $50 billion. Of those, 10 out of 13 projects that had entered the implementation phase experienced significant cost and/or schedule growth. For these 10 projects, development costs increased by an average of 13 percent from baseline cost estimates that were established just 2 or 3 years ago and they had an average launch delay of 11-months. In some cases, cost growth was considerably higher than what is reported because it had occurred prior to the most recent baseline. Many of the projects we reviewed experienced challenges in developing new technologies or retrofitting older technologies as well as in managing their contractors, and more generally, understanding the risks and challenges they were up against when they started their efforts.”
All of these costs will be passed onto consumers, so it is very important that it not be allowed to get out of control and that lobbying be monitored. Have a look the EPRI document “Methodological Approach for Estimating the Benefits and Costs of Smart Grid Demonstration Projects” to get a feel for the complexity of estimating potential benefits, and then ask yourself if it seems credible that all this would save between $1.3 to $2 trillion, meaning that the smart grid would actually pay for itself.
Smart grid reliability is often promoted as a cost saving potential, yet more complicated systems operated with lower spare capacity (or increased utilization) is a recipe for large scale outages. The old saying that the most reliable part of any system is the piece you leave out is apt. Although the existing grid is becoming less reliable as it ages, a lot of the problems can be only fixed the old fashioned way — keeping right of ways clear, replacing aged equipment and towers and insulators, adding new lines and power stations, better training and maintaining adequate staff. Adding a large SCADA layer to the network will alert you to failure quicker but isn’t going to address the antique physical infrastructure.
Distributed power generation and storage have many possibilities. For example, a hospital emergency generator could be fired up remotely by the power company to make up for a local shortage in another community, and power that normally flows one way in a grid, could become bi-directional.
Electric Cars have large batteries that can be used for regulating grid voltage or to store excess wind energy during windy periods. In the article “The economics of using plug-in hybrid electric vehicle battery packs for grid storage” the authors looked at a hypothetical situation where Chevy volts were used by their owners for driving between 8am and 5pm, and then charged and made available for grid storage. If the owners are clairvoyant and know in advance when it is cheapest to recharge, and the most profitable hour to resell, a driver in Boston might make between $12 and $48 a year (life reducing, extra wear factored in). This isn’t enough to be worth the inconvenience, and car manufacturers would have to modify their warranty agreements to include such usage. Each home would also need a 7200 watt grid tie inverter to convert DC back into 220 VAC/60Hz with low harmonics (a nice, clean looking sine wave). If enough people joined in to actually balance the load, there would be no opportunity for arbitrage. Governments could pay car owners willing to participate for the sake of being better able to use variable sources like wind farms but any payment system has the potential to be misused.
The article Using fleets of electric-drive vehicles for grid support looks at using company owned vehicle fleets to provide voltage regulation services for power companies. In some regions, vehicle to grid (V2G) power can be profitable, but is highly dependent on market rates.
All this sounds like it has some potential, but at some point the big question has to be asked. Why are we doing this, and at what cost? If a trillion in added costs are to be added on top of the ever rising price of fossil fuels, we need to be very clear about what the objectives should be.
The standard answer to the first part of this rhetorical question is “to keep up with the ever increasing demand for power”. This unfortunately is not even remotely possible. We have an exponentially growing population that increases about 13% per decade and declining non-renewable resources.
Growth can’t go on indefinitely despite all the political statements talking about the importance getting the economy growing again. The White House has a report called “Living within our means” and has the ludicrous subtitle – “…plan for economic growth”. The report is absolutely littered with the references to economic growth. For a dose of reality, watch professor Albert Barlett’s lecture – it has more than three million hits on YouTube.
Fossil fuels will be available forever, but there will come a time where the energy used to extract them (e.g. expensive deep water drilling) will be equal to the energy in the fuel itself. At this point, it no longer makes sense to extract it except for special cases. One of the more well known examples of dubious yields is the return on corn based ethanol (see theoildrum). Although there is considerable argument on what to include in calculations for energy return on energy invested (EROI), there is general agreement that it is deteriorating. See “The Economic Cost of Energy, EROI, and Surplus Energy” at theoilldrum for some in depth discussion.
Uranium based nuclear isn’t long term either. The NEA states “There is enough uranium known to exist to fuel the world’s fleet of nuclear reactors at current consumption rates for at least a century, according to the latest edition of the world reference on uranium published today.” A hundred years is nothing — many people born today will be around then and if we scale up nuclear power, we will run out a lot faster. As uranium becomes scarce, Thorium reactors are possible, but this technology is still a long way off.
Can Renewable Energy Save the day?
In Sustainable Energy – without the hot air by David JC MacKay, a professor at the University of Cambridge, looks at the renewable potential of the USA using truly extreme examples, and then adds them all up to get a feeling for the upper limits of what is possible. Here are some excerpts:
“The windiest spots are in North Dakota, Wyoming, and Montana. They reckoned that, over the whole country, 435 000 km2 of windy land could be exploited without raising too many hackles, and that the electricity generated would be 4600 TWh per year, which is 42 kWh per day per person if shared between 300 million people. ”
“If we assume that shallow offshore waters with an area equal to the sum of Delaware and Connecticut (20 000 km2, a substantial chunk of all shallow waters on the east coast of the USA) are filled with offshore wind farms having a power density of 3 W/m2, we obtain an average power of 60 GW. That’s 4.8 kWh/d per person if shared between 300 million people.”
“The hydroelectric facilities of Canada, the USA, and Mexico generate about 660 TWh per year. Shared between 500 million people, that amounts to 3.6 kWh/d per person. Could the hydroelectric output of North America be doubled? If so, hydro would provide 7.2 kWh/d per person.”
He adds these kind of things up to 62 kWh/d per person which is a far cry from our current average of 250 kWh/d per day. Even worse, if outsourced jobs and manufacturing ever return, we will be using a lot more energy in North America. How could energy demand be reduced by 75%? How about giving up cars, air conditioning, air travel and winter heat? Too extreme? Perhaps eliminating of agricultural fertilizers (and see food prices soar), street lighting, rail transport and all those products from Walmart while keeping the car and winter heat? Of course, it isn’t all or nothing, the most likely income is that all these areas would be reduced and that the average person will fly about as often as they used to fly the Concorde. Our great grandparents would be laughing at us.
The production of renewable energy in a scale vast enough to replace our current demands runs into a number of problems:
- The mega projects required would be astronomically expensive. Diverting massive military budgets to rebuilding infrastructure is one of those things that ought to happen, but wont, without regime change.
- Much of the energy available from wind and solar is not near major population centers. Unlike previous centuries when people lived near rivers for transportation and water, or in great farming areas — we live all over the place because we can. No water? Drill the aquifer or divert a river. Cold as hell? Use electric heat. No electricity? — build thousands of miles of transmission line. No food? Build highways and truck it in. This is enormously wasteful and great deal of energy is lost in long transmission lines. Fixing this would require population migrations.
- It takes up front energy to manufacture replacement technology such as wind turbines (steel, copper, Aluminum must be mined and manufactured) and to build enough capacity to keep up with the decline in oil production, we would have to divert almost all our spare capacity into alternate energy (e.g. suffer now for capacity later). This will not happen.
- The best hydro resources with large reseviors are already developed, with a few exceptions like the lower Churchill in Canada.
If there is going to be a war like effort to solve the upcoming crisis, it needs to have a larger scope. Will we ever see propaganda this again? We could treat our energy dependency as a common enemy.
There is more to energy than just electricity
The trillion dollar smart grid initiative, even if it could be flawlessly implemented along with equally large investments in solar capacity would still come nowhere to enabling the continuation of the current North American lifestyle. Most heating is via natural gas, the vast highway system consumes asphalt, building materials like cement and aluminum require large energy inputs, liquid fuels are indispensable for air travel because of the high energy density, agriculture consumers natural gas as a fertilizer feed stock and much more.
A far better use of resources is to not use the energy in the first place. This is not really that difficult, however disruptive technologies will be attacked. There are trillions in existing infrastructure, and “preserving the investments in the existing infrastructure” is going to be a major concern of corporate lobbyists.
I see the following as being essential.
- Low energy homes. Each climatic region needs custom designs (this is a good task for public universities) for solar passive homes. Regulations need to be implemented to set minimum thermal standards. It is possible to build homes that do not need air conditioning, and only minor axillary heat in the winter using inexpensive local building materials.
- Long range city planning. Residential subdivisions are really dumb. These often have no place to work as a wage slave, forcing excessive commuting, nor enough land to be self sufficient even on a subsistence level. Every lot should have septic, water and room for a small garden. It is ridiculous that 15% of Americans are on food stamps and that they have no ability to feed themselves without government assistance. Subsistence living is not a goal — but it should be an option.
The existing power infrastructure is fine if we slash our power needs, keep up the grid maintenance and stop economic and population growth. We don’t need a trillion dollar “smart grid” upgrade. We do need self sufficiency at the homestead level. 15% of Americans rely on food stamps and nearly 20% are unemployed. We need to reevaluate our priorities.
Also note the view count on the video — more than a quarter of a million views in one day and the ratio of likes to dislikes is 92%.
