Do We Need a Smart Grid, A Strong Grid, Or Less Grid?
One of the reasons that I am a strong advocate of smaller nuclear energy plants is that I resist the notion of the inevitability of greater interconnectedness and dependence. I like systems that have firewalls, circuit breakers, and air gaps to ensure that a failure in one location can be isolated. I do not like tightly integrated systems that have little margin for error. I dislike electronic systems that easily communicate with each other without the need for human decision making – it is too easy for them to pass bad information or viruses too rapidly to halt the spread.
Based on my personal experiences during the past couple of decades on the Internet (I sent my first email in 1985), the talk of smart grids with appliances and generators exchanging two way communications with “the grid” scares me. Being on the Internet is a lot of fun and hugely informative – as long as you have taken the proper precautions and as long as downtime is not fatal. The risk of infection or lack of access is low enough to make it worth taking in order to allow connection and communications.
I cannot make the same statement about electricity. I cannot see any real advantage to being tightly connected. I do not have to be connected to anyone in order to use electricity if I have a generator or a battery. My car’s systems work great on electricity without any connectivity to any other object; so did my submarine’s power systems. If I have a generator AND a battery, the backup provided by tight connections to others is worth even less if the battery lasts long enough to get the generator running again.
I am not the only one for whom tightly integrated, wide spread “smart” grids is a matter of concern. Nick Rosen, the author of How To Live Off-Grid: Journeys Outside the System, gave a talk on October 22, 2009 at an event sponsored by Mondo Energy titled: The Truth About Nuclear Energy.
Rosen has been an anti-nuclear activist for his whole life, starting off with participation as a child alongside his parents in protests against the Bomb. Later, he transitioned that active opposition to nuclear weapons to active opposition to the construction of nuclear power plants; people who know him might have been surprised by his participation at the event.
However, Rosen resisted nuclear power plant construction not because he did not understand that producing power was different from producing weapons, but because he thought that the plants being built were simply too large, too centralized, and too scary. (His comment is that any mistake would be a big mistake. Of course, we know from experience at Three Mile Island that the health consequences of even an extensive failure at a nuclear plant can be minimized, but we also know that the financial consequences are just what Rosen predicted – BIG.)
In recent years, however, Rosen has started to think about the possibilities of building smaller nuclear plants for those people who do not want to be completely connected to everyone else. He sees community sized plants as something entirely different from the large, central station plants that he resisted. In comparison to most anti-nuclear activists, Rosen had an advantage in making that kind of leap in his thinking. His father, though a protester against the development of the British bomb program, was a nuclear physicist who helped him understand the concept and the attraction of making reliable power from tiny amounts of material.
I like imagining how the world would be different if there were nuclear fission energy systems available in as wide a size variety as the combustion systems available today. (Note: I actually like humanity and do not immediately believe that they would turn useful tools into deadly weapons. I never worry that my local hardware store has a whole row of axes, wondering if one of my neighbors will buy one and use it while going on a rampage.)
Part One:
Part Two:
By it’s nature, the grid is tightly integrated in the sense that you have synchronous machines electro-mechanically connected. That does not mean it has little margin for error or is not robust (although we have moved in that direction). “N-1” planning has always been paramount in power system engineering to allow for failures. In this regard, expenses associated with spinning reserve and a topology that allows reasonable partitioning in the event of failure is required. Unfortunately, we are moving in the opposite direction.
I am not a fan of the Smart Grid. To me, those are code words to perform demand management in order not to build the plants and infrastructure necessary to maintain a robust electrical system.
My recollection is that the reason there are no “consumer neutronics” is the expense. Even with mass production, you would need higher enrichments (and the associated SWU cost) to have small reactors in every home. There were more practical ideas such as the Canadian Slowpoke reactor for apartment buildings, but they had bad timing (i.e., they were tested when natural gas prices were low).
With the recent achievements in HTGR fuels, I would imagine a 150 -250 MWe reactor that could provide the flexibility of a gas turbine in the medium future (10-15 years).
Arcs_n_Sparks is right, the ‘Smart Grid’ is nothing more than a Trojan horse for demand side management. The other issue is that is also a way of avoiding needed upgrades to the grid, (like new transmission) that will in the end leave the system less stable, not more.
Speaking from experience, the “smart meter” the utility recently installed is very nice. I watched the installation occur in under 5 minutes. I can see the kW draw in easy to read numbers (no more cryptic codes), the total kWh’s, and a graph of usage online – though I’ve not used that feature yet. The meter also sends the usage data wirelessly to the utility thereby saving them labor to read the meters. I know this is only one aspect of the “smart grid” system but I think this is very useful for the end user.
The smart grid is problematic from several perspectives:
1. Introduction of unnecessary complexity into a system that should be straightforward and as simple as possible. Rather than achieving better failure tolerance and “self-healing” capabilities, one adds in additional components that can fail and cause failure cascades, as well as the possibility of byzantine faults, as well as new modes of failure – including unforeseen modes of failure. Software glitches could shut down entire regions, hackers could turn the lights off remotely, and an EMP could insta-fry the computer logic of the grid itself! The intelligence of the grid is the human components of it, who engineer it, build it, operate it, and maintain it.
2. Introduction of capabilities to monitor and record more than plain electricity use. Already the database state grows in all dimensions. We do not need more intrusive examination of the character of the very loads that we place on the electrical network by the state – or convenient proxies thereof (such as the power company). With the NSA warrantless surveillance controversy, we see how upstanding institutions – telecommunications companies – can be easily induced through whispers of patriotism and civic duty to turn over loads of personal information to the government on a silver platter. How can we expect the power companies to behave any differently under pressure? We cannot. Nor should we. Power corrupts.
3. Predatory behaviors – markets for electricity do not work well as demand is inelastic. Failure to meet demand will result in exorbitant rates and windfall profits for those who do not deserve them while the “Average Joe” and the local power company gets screwed. See Enron’s pump-and-dump scams. Not saying that California didn’t deserve to have the lights go out for obstructing construction of generation resources (this is the rightful, logical, and deserved consequence of blocking construction of necessary grid resources), but the consumers – and PG&E – didn’t deserve to be nearly bankrupted by scammers shutting down power plants for unscheduled “maintenance” to drive up rates. This is the logical consequence of allowing time of use metering and market-based electricity rates. Electricity is a vital commodity, not a privilege; pricing should be based on cost, with an allowance for a reasonable, regulated (and well deserved) profit, not a windfall profit based on what the market can tolerate.
4. Ability to control electricity use. Controlling demand is dangerous. During the winter – or the summer, in certain areas – it can be fatal. Allowing grid users to choose when to use electricity preserves their safety – while taking that choice away from them endagers everyone. The smart grid would allow that control to occur.
This is not to say that meters shouldn’t be upgraded to display instantaneous electrical use, or auxiliary meter interfaces – allowing for an in-house panel showing instantaneous consumption – shouldn’t be provided. Demand charges – for the peak amount of electricity used during a specific time period (e.g. a 1 hour period) during a specific billing period (e.g. a month) – are reasonable to uphold. But beyond that, time of use metering is not a smart thing – and introducing unproven electronic components – capable of being hacked, fried, failing bizarrely, being used to manipulate rates, being used to control people’s electricity use, or being used in a spying scheme – into an already stable system that works well – is not a great idea.
If more transmission lines need to be built, build them. Don’t use them as an excuse, though, to put needless complexity into something that already works quite well.
Adovocates of so-called smart grids often emphasise that it will be a way of allowing distributed intermittent power plants to somehow bring ‘power back to the people’ through being decentralised. This ignores the fact that the current grid is already decentralised in a way that brings true, meaningful power to the people: Reliable electrical power that can be accessed when THEY deem it desireable. A ‘smart grid’ (ie, an intrusive, centralised grid command system) would radically centralise the decision making which is currently distributed through the end-users of the grid, disempowering the masses in very concrete terms.
Who would willingly bet that the concentration of power thus entailed in the hands of a technocratic/bureuacratic/corporate elite would never be abused?
The Internet grid is a wonderful analog for a robust electric grid. DARPA designed the internet to HAVE NO CENTRAL CONTROL so that it was not vulnerable to outages. The “smartness” is localized in each node, which cooperates via the grid, but always exhibits independent, protective intelligence. An internet-like smart grid node would have local power for local demand, with crossconnects to other nodes, supplying or borrowing power as needed, following the rules of the grid, or risking being cut off. By the way, this approach also lessens the looming problem of cyberterrorism shutting down “the grid”.
Well that’s the Party line on this, but as wonderful as it sounds it is not the way it is planned to be implemented. It is very clear that the initial steps will involve demand-side management via ‘dynamic pricing’ at first and by the requirement to surrender control over things like heating, hot water, A/C and other inductive loads to the grid. One will be expected to adjust one’s behavior to suit this management; don’t expect that three people will be able to shower in the morning, or put your washing in the dryer and wait for the grid to decide when to start the machine, and so on.
As for ‘local generation’ (which is a euphemism for wind and solar) we have discussed this before. The bottom line is that any local installation will have to have the capacity to produce power not only for its own loads, but for every other load that may be demanding power from it. Consider the situation in the U.K. a few weeks ago: record lows, the whole island becalmed under overcast sky’s. Just exactly where could power be wheeled from then?
Even if was possible to network generators the way the internet is, consider too that trafficking in and switching really significant amounts of energy cannot be done as cavalierly as is the case with data. Controlling the VAR’s alone, never mind things like switch-arcs and such, is non-trivial, and an error can cause disaster-level physical damage cascading through the system. We are talking weeks without power type failures. This being the case the level of flexibility a ‘power internet’ could have would by sharply limited to not very much more that the current grid architecture allows at the moment.
Smart-grid a buss-word for a set of service reductions/price increases without any real value added from the consumers perspective. And that is about all it will ever be.
Robert – I have to agree with a few other commenters that the specifics of the Internet model are completely different from electricity. It may sound attractive to imagine a system with “No Central Control” but we all have far different standards for electrical continuity than we do for data delivery. As many have pointed out, the Internet delivers PACKETS not a continuous stream of electrons that, if interrupted for even a fraction of a second, can cause serious equipment damage. All of us have experienced the lengthy page loads, the service interruptions, the intermittent page not found errors, and the occasional delivery delays for email. Most of those occurrences are so commonplace that people do not even notice them.
The grid has always had intelligence in it – circuit breakers, voltage regulators, and frequency control devices operate in automatic. In addition, the grid harnessed the intelligent behavior of hundreds of millions of people who can turn lights on and off, start and stop appliances, and generally go about their lives with a steady stream of electric servants ready to do what they are told to do on demand.
A long lingering last salute to the Balkanized DUMB GRID –
There are relatively few defenders of the “Balkanized DUMB Grid” but for all the many decades of our lives when you plugged in a toaster to DUMB GRID things just worked.
The complex, incrementally piecemeal implemented, SMART GRID with its complex communication and control protocols roughly patterned after features of the Intranet remains incompletely designed and its ultimate efficiency and reliability is unknown. Hopefully, government engineered massive additional complexity will result this time in massively greater reliability and security.
I personally do not think the internet message switching/redundancy model is a good one for Smart Grid. The problem of stable continuous power delivery versus intermittant, time flexible, but robust delivery of messages (or packets) are quite different problems. Switching heavy power delivery infrastructire is not that similar to switching packets on a fiber optic small signal message delivery infrastructure.
Preserving parts of old fashioned and centralized DUMB grid while you piecemeal invent and implement SMART GRID is likely to be less risky than radically completely dismantling DUMB Grid in an effort to more quickly shift all service to SMART Grid. (Engineering best practice suggests that a system that is first fully engineered and then built tends to be least costly and more reliable than a system that is incrementally designed and implemented).
While SMART metering seems innocent enough and letting people know not only how much but when they used energy appears enlightening, there are problems I anticipate that may come from follow on developments when SMART appliances for the home, including SMART Heating and Air Conditioners, become mandatory. At some point SMART Grid communication and control features are likely to be used to permit utility managers to remotely turn off SMART GRID intelligent furnaces and air conditioners in your home at their, instead of your, discretion. Instead of you deciding when you plug in the toaster to make toast, the utility manager decides when your utility region experiences a rolling blackout to match the supply of intermittent climate sensitive energy sources available.
Deciding when you use energy is a practical personal freedom that you have enjoyed in the era of the Balkanized DUMD GRID. You may wistfully miss the loss of this freedom someday when you want to plug in the toaster and just make toast. Charging more for power at times of the day when climate sensitive energy is not available is one thing. Actually shutting off power for discretionary use at certain times of the day reduces my freedom and quality of life.
I would just like to emphasize that I mean no offense to anyone from the Balkans when discussing the “Balkanized DUMB GRID”.
If I had the courage I would say that what produces the most value for the least money is a GRID design that features nuclear plants located just on the outskirts of the urban areas for which they supply power. I personally like the underground mounting concept for new nuclear (as proposed by Edward Teller and others). Underground mounting at approximately 10 meter depth could permit the extremely small physical footprint and energy dense nuclear power plants to be put right under the city they serve while preserving total freedom regarding the surface use of the real estate. Changes to the GRID would be minimized by this approach and use of materials for power distribution (copper) would also be minimized.
The more conscious and politically correct might combine local underground mounted nukes with distantly generated sustainable wind and solar power. This would be a “belt and suspenders” GRID approach that would have the value of keeping the stronger local nukes in the picture ready to carry the load and provide stability and high quality power anytime the climate dependant wind and solar disappears because of a cloud or a momentary dearth of wind.
I am only modestly courageous and only infrequently politically correct. I just send letters to my Congressman about the relative cost effectiveness of locally sited new nuclear versus the poor cost effectiveness of distantly generated renewable energy requiring expensive and still incompletely engineered and unverified GRID upgrades. I regularly send in my letters to my mostly liberal and environmentally sophisticated and connected Californian decision makers who ignore my pleas and chart me a course down the river to a lower quality, less robust, and a less plentiful energy future.
While I do not deny that some of the technology that will be deployed for the Smart Grid can be abused, I do not accept that fact that you remain powerless to do anything about it.
If you do not like the way this is all going, rally your neighbors and pressure your elected representatives and local PUC! They did it in the Netherlands, where “the Dutch government had to back down after consumer groups raised privacy concerns. Instead of a mandatory role out smart meters will be voluntary”.
There is a lot of attention currently on smart metering, time of use pricing, and demand response to clip the demand peaks. But this is far from the complete picture of Smart Grid. Even if one assumes a zero CO2 100% nuclear future, there are still many advantages to be had for a smarter grid.
Substation automation, faster outage detection and repair, reduced costs for meter reading, synchronization of the major regions of the grid (NASPINet), replacement of aging grid infrastructure, enabling distributed generation (including small modular nukes), etc. Reliability and availability are a huge piece of this. It ain’t just about smart meters.