Whenever I launch into one of my frequent rants about the need for utility-scale energy storage to support the roll-out of more Solar and Wind generation there are always people that point to pumped storage as a credible solution. And every time that happens I argue very strenuously (and I hope convincingly) that pumped storage is NOT a viable solution because there are not enough locations where it works.
The Bath County facility in Virginia (the world’s largest pumped storage) is an engineering marvel – I encourage anyone interested in storage to check out their site and the video which is quite inspiring. But to create a second large reservoir close to the main reservoir in other locations is difficult. There may be many sites being proposed but are they large enough to make a real impact?
To get an idea of the scale required it would take 3 Bath County sized facilities to be able to store the output from the existing wind generation capacity in Texas. It took 8 years from the time the Bath County facility was licensed until it went into production. And these days the development of large reservoirs is subject to significant resistance from some segments of the environmental community.
I support the development of as much pumped storage as we can reasonably do as quickly as possible. Ironically, even good sites for pumped storage projects can encounter environmental and cost barriers – examples being the Iowa Hill project in California and the Pioneer-Burdekin project in Queensland, Australia. I just don’t think these kinds of projects even if approved will be large enough or fast enough in North America to help us deal with wind variability in the next 5-10 years. On the other hand, pumped storage may be a significant part of the solution in Europe if they can overcome the cost/environmental challenges of building the undersea interconnects required.
In terms of North America it seems to me that there may be a way to achieve the same result using a different approach.
Every hydro facility has penstocks and generators sized to make use of the average stream flow of the river that feeds the reservoir behind the dam. It would not make sense to design in a lot more generating capacity because using more water would quickly draw down the reservoir to a level where power generation would no longer be possible. In dry years some of the penstocks are closed and in wet years water is spilled but on average the facility is designed to use all of the water supply available and no more.
Now if we decided to use existing hydro facilities in a different way it would be possible to provide backup generation for wind and solar. What I am proposing is that additional penstocks and turbines be added to large-scale hydro facilities in North America – perhaps as much as double the current capacity. In most cases this could be done using tunnels from the existing reservoir as shown below.
In order to make this work the average capacity of wind generation (typically 25-30% of nameplate) would have to be roughly equal to the excess capacity built into the hydro facilities.
When wind generation dips below the average the hydro facilities could make use of the excess capacity installed to make up the difference. This of course would lower the level of reservoirs but typically reduced wind conditions only last for a few hours or at most a few days.
When wind generation exceeds the average the hydro generation could be cut back and reservoirs could be refilled or water spilled. By balancing the two generation sources the regional grid would always have access to the same amount of power despite the variability of wind.
This exact situation is already occurring in Scandinavia where excess wind energy from Denmark is often available at night and excess hydro is available during the day. The difference with “Unpumped Storage” is that the entire system will be designed to balance excess wind capacity whenever and wherever it might occur with excess hydro built into existing facilities.
There are some significant challenges with this approach (as there are with any concept that embraces renewable sources in a major way).
- Grid Capacity: Balancing wind generation with hydro that might be a thousand miles away will require new transmission lines – very large new transmission lines. In the worst case with a large high pressure zone sitting over the Mid-West wind generation could drop to essentially zero from a high of something like 10 GW or more (this exact situation happened in December, 2012 in Texas where a new wind generation record of 8.6 GW was followed the very next day with 6 hours of no appreciable wind at all). So Unpumped Storage would require a very significant investment in new inter-connections between regional grids
- Reservoir Levels: Unpumped Storage would cause reservoirs levels to drop more quickly and to a lower level than with normal hydro operations. This would have ecological impacts that need study and could have a significant impact on recreational activities.
- Energy Prices: This approach would not work in a deregulated environment where wind and hydro producers were effectively competing for market share in a “spot market”. Hydro, being the reliable source, could demand almost any price when wind generation dropped dramatically. By way of example, Texas raised the ceiling price of electricity to $9,000/MW-Hour (the average annual price in Texas is $55/MW-Hour) to try and entice utilities to build more base-load generation capacity. The concept seems to be that if you let base-load plants charge 170x the average price for the few hours that they can get access to the grid then utilities will spend the $billions required to build new facilities. Call me a skeptic but I would say “that dog don’t hunt.” In the other extreme when the wind is blowing hard the spot price can drop to zero (actually to less than zero in Texas about 10% of the time because of Production Tax Credits earned by wind producers).
- Jurisdictional Issues: To make Unpumped Storage work the jurisdictions in which the hydro and wind were located would have to cooperate in every way; regulations, import/export policies, transmission facility planning and control, pricing mechanisms, and financial incentives would all have to be aligned across the entire region regardless of how many state and provincial borders the electrons crossed. Would this be easy? No. Would it be expensive? Very. But would it work? In many regions of North America the answer is a definite yes.
