Nuclear maneuvering in Ontario: planning future agility

In recent posts on Ontario’s long-term energy plan, I have pointed out that wind power is actually gas power. That’s because wind is unreliable and needs to be balanced with gas-fired electricity. The  following chart shows Ontario grid-connected wind farm output between December 22 and yesterday (January 4). Each column represents one wind farm’s hourly output. Wind output was unusually high in that period, thanks to the storms over the U.S. Northeast and Atlantic Canada. Nevertheless, wind output was very erratic.

Click on image to enlarge it

More than two-thirds of the output shown in the chart was between 3 and 87 megawatts. The average capability of the wind fleet during that period was roughly 104 MW; the average capability factor—defined by the IAEA as “the ratio of the available energy generation over a given time period to the reference energy generation over the same time period, expressed as a percentage”—was around 45 percent. That’s high for wind.

But electricity demand over that same period was much more constant and predictable. This meant that the Ontario grid operator (the IESO) needed to balance supply and demand using other generation sources. To keep supply balanced with demand, the grid operator would need a fast-dispatchable source of generation in this circumstance. In Ontario, that means hydro or gas. And since by political fiat the coal-fired generators are being phased out, it really means that gas will play a critical role in meeting demand without blowing up the grid. Gas, as I and others point out, is a dangerous, expensive, CO₂-emitting fossil fuel.

Maintaining a balance between supply and demand also means that some other generators must power down—or go offline—if wind spikes up at a time when demand is low. That is because, again by political fiat, wind generators get pride of place on the grid.

This points up a glaring inconsistency in Ontario’s long-term energy plan: wind is being force-fed onto the provincial grid—at very high prices—because it is ostensibly CO2-free. As I illustrated in my post “Wind power is gas power, and comes with pollution,” wind is patently not CO₂-free.

Because of its erratic and unpredictable nature, wind is dangerously interfering with Ontario’s biggest source of real CO₂-free electricity.

How? Because wind gets pride of place on the grid. This means if the wind spikes up at a time when all supply matches demand, other generators have to power down or come offline so that the grid stays in balance. This means that nuclear generators—yes, the ones that provide the biggest amount of electricity—have to maneuver around wind.

In other words, because of a politically correct policy, nuclear, which produces electricity at roughly 5.8 cents per kilowatt-hour, has to make way for wind, for which Ontario rate-payers shell out more than 12 cents per kWh. It is provincial policy to pay more than twice as much for wind as for nuclear, and wind gets pride of place on the grid. And why does it get pride of place? Because the deciders of our electricity policy have bought the line that it emits no CO₂. But it does emit CO₂.

Don Jones, a former reactor designer at Atomic Energy Canada Limited, has written a number of articles that brilliantly illustrate wind’s impact on the grid. In one of them, published today at Wind Farm Realities, Jones points out the problems that the rush to wind will create on our grid, especially if the provincial government actually implements the current long-term energy plan:

With nuclear limited [in the LTEP] to meeting 50 percent of demand and hydro, depending on water supply conditions, optimistically meeting 25 percent, dispatchable gas generation and wind/solar must pick up the rest. Long and short term dispatchable support of wind will normally be by gas although any available hydro could be used if required in the short term if some gas generation is temporarily out of its dispatchable range. The real concern with a wind and gas combination on the grid is recovery after a “sudden loss of wind event” (SLOWE) when combined cycle units are operating well below their dispatchable power range.—Don Jones, P. Eng. (source: Wind Farm Realities, Jan. 04 2011)

And what happens when a SLOWE—a sudden loss of wind event—does occur? Well, without coal the only instant-dispatchable sources are pump-storage hydro and simple-cycle gas. Hydro is held in strategic reserve to handle daytime peak usage. As you can see in the chart above, wind doesn’t magically kick in during peak periods; it comes and goes on its own schedule. That means simple-cycle gas will have to handle SLOWEs. If Ontario follows through with the amount of wind called for in the current long-term energy plan, we will need more simple cycle gas.

But, as Jones points out,

It would make little environmental or economic sense to have enough simple cycle gas turbine generator capacity available to cater for a SLOWE if that is more than the amount required to meet normal peaking and operating reserve requirements.

Jones argues, and I agree with him, that the dispatch order policy in Ontario is bass-ackwards. Instead of maneuvering billion-dollar nuclear units around expensive, erratic wind, the rule should be that the nuclear units are the last to come offline in periods of surplus demand.

The problem with that is what happens when there is still too much baseload demand. This has occurred a number of times recently. The current nuclear fleet in Ontario—consisting of five sets of CANDUs of different vintage—can respond to maneuvering requests from the IESO but operators have to resort to steam bypass or reactor power changes that must respect chemical and physical limits.

To address this, Jones suggests a combination of refurbishments that either upgrade steam bypass systems and/or enable deeper reactor power maneuvering and new reactor technology that includes load-cycling and -following capability in the design.

The latter bears on an issue that is of critical importance to Canada’s nuclear future: the sale of the CANDU division of AECL. That sale is utterly dependent on the new construction project at Darlington. In 2009 the Ontario government held a reactor competition, from which AECL’s new ACR-1000 emerged as the front-runner. But the ACR is still in development, and that costs money. In fact, the development costs are why AECL’s current owner, the Canadian federal government, wants to sell the CANDU division.

And uncertainty over what the ACR would ultimately cost to build is what led to the shelving of the Darlington reactor project.

This has led some in the industry to suggest forgetting about the ACR, which AECL says can load-follow and -cycle, and going with the tried-and-true CANDU 6, 11 of which have been built in Canada and around the world, the most recent under schedule and under budget.

If Ontario went with the CANDU 6 at Darlington, there would be no issue about when the plant would come online and how much it would cost.

But would it have the maneuverability that AECL claims the ACR has?

If Ontario went with the CANDU 6 that would kill the ACR. Would that make it a more attractive takeover target for other reactor vendors, like Areva or Westinghouse? Those companies’ reactor portfolios are rather limited to big pressurized water reactors. Neither would be interested in acquiring the rights to a machine that competes directly with their own showcase offerings. Areva has gone to lengths recently to prove to critics that it isn’t a one-trick pony based entirely on its 1,650 MW EPR. But the moves it has made in that direction consist of touting two only slightly smaller light water designs—both of which are in the capacity range of the ACR. Could it benefit from having the natural uranium 700-MW CANDU 6 in its portfolio? It certainly could sell these machines—it has truly global marketing clout.

Of course, that might not solve the nuclear flexibility problem in Ontario. What would solve it is if Ontario abandoned the ill-conceived coal phase out. Or AECL, or whoever buys the CANDU division, could apply the modifications Don Jones suggests to the CANDU 6.