Question 8 of 8: Nuclear and Wind Power

Question 8: Nuclear and Wind Power (10 marks)

Question text not reproduced: the examination questions are © Engineers and Geoscientists BC. Open the official past paper (linked at the top of this page) to read the question, then follow the worked solution below.

A Canadian utility adding roughly 2000 MW of non-emitting capacity by 2029 is choosing between two mature technologies that behave very differently on the system. The comparison below works through the aspects listed in the question and closes with a recommendation.

Number of generating units and land requirement. Two thousand megawatts of nuclear capacity is two to three units. A CANDU 6 rates about 700 MWe, so three units serve; a pair of AP1000-class or a fleet of small modular reactors would be the alternative arrangement. The whole station occupies a few square kilometres of a single controlled site with a cooling water source, and the exclusion zone around it is the dominant land take. Two thousand megawatts of wind, by contrast, is roughly six hundred modern 3–4 MW machines, or well over a thousand of the 1.8 MW class considered in Question 5. Spaced at the five-to-seven rotor diameters needed to limit wake losses, that fleet is spread over several hundred square kilometres, necessarily across multiple sites and often multiple municipalities. The land is not consumed — farming and grazing continue between the towers — but it is committed, and the number of landowner agreements, road crossings and municipal approvals scales with the machine count rather than with the megawatts.

Environmental impact, effluents and emissions. Neither option emits carbon dioxide, sulphur oxides or particulates in operation, which is the premise of the question. Their residual impacts are of quite different kinds. A nuclear station discharges warm condenser cooling water and small, regulated quantities of tritium and noble gases; it produces spent fuel that must be stored in bays, transferred to dry storage, and ultimately placed in a deep geological repository — the mandate of the Nuclear Waste Management Organization in Canada — and it produces decommissioning waste at end of life. It also carries a low-probability, high-consequence accident risk that is managed through the CNSC licensing regime, defence in depth and emergency planning zones. Wind produces no effluent at all, but has a diffuse footprint: avian and bat mortality, low-frequency noise and shadow flicker for nearby residences, visual intrusion on a landscape scale, and blade and foundation waste at decommissioning. Public acceptance is the practical constraint in both cases, but it attaches to a single site for nuclear and to hundreds of neighbours for wind.

Capacity factor, reliability and the ability to follow load. This is the decisive difference. A well-run CANDU or PWR station achieves a capacity factor of 85–92  %; its output is dispatchable, available on demand, and interrupted only by planned outages of known duration and by occasional unplanned trips. Two thousand megawatts of nuclear nameplate therefore delivers about 1800 MW of firm, schedulable capacity. Onshore wind in most of Canada achieves 30–40  %, so the same nameplate yields an average of 600–800 MW and a firm contribution at peak — the capacity credit the system operator will actually count — of only perhaps 10–20  % of nameplate. Worse, the resource is uncorrelated with demand: Canadian winter peaks occur on cold, still, high-pressure mornings, exactly the conditions under which wind output collapses. Wind can be curtailed downwards on command but cannot be called upwards, so it cannot follow load. Nuclear can follow load in principle — French PWRs and CANDU units do manoeuvre — but it is usually operated baseload because its costs are almost entirely fixed and because xenon transients complicate rapid manoeuvring. Neither technology, in short, is a natural load-follower; but only one of them is firm.

Grid stability and connection. A nuclear station connects through a handful of transformers at one switchyard onto the extra-high-voltage network. Its large synchronous generators contribute rotating inertia, short-circuit strength and reactive power support — all of which help hold frequency and voltage during disturbances. The counterpart risk is that the loss of one 700–1100 MW unit is a large single contingency that the system must be able to absorb. A wind fleet connects through hundreds of converter-interfaced machines dispersed across the province, often at points where the transmission system is weak because the wind is best where the load is not. It supplies little natural inertia and, unless grid-forming converters and synchronous condensers are added, it degrades system strength as it displaces synchronous plant. Substantial new transmission, and probably storage or firm backup, has to be built alongside it.

Construction and maintenance. Nuclear construction is a single decade-scale project with high engineering intensity, a long licensing path through the CNSC, and a well-documented history of schedule and cost overrun on first-of-a-kind units; against a 2029 deadline set from today that is a serious constraint, and only a refurbishment of existing units or a committed SMR programme would plausibly meet it. Wind is modular and fast: individual farms are built in twelve to twenty-four months and generate revenue as each phase is commissioned, so 2000 MW can be staged over the period with far less schedule risk. In operation nuclear needs a large, highly trained, licensed staff, planned outages every one to three years, and mid-life refurbishment; wind needs a small staff but visits several hundred nacelles a year, with gearbox and blade repairs at height, and its maintenance burden scales with unit count.

Recommendation. Neither option alone is a sound answer to the requirement as stated, and the recommendation should say so. If the utility genuinely needs 2000 MW of dependable capacity to serve forecast demand in 2029, nuclear is the appropriate choice, because it is the only one of the two that delivers firm, dispatchable, inertia-supplying capacity at that scale on a single site — provided the utility already has a licensed site, an established CANDU or SMR supply chain and a licensing path capable of meeting the date. If that lead time cannot be met, the practical recommendation is a staged portfolio: commit to the nuclear units for the firm requirement while building wind in phases to cover energy growth in the interim, and pair the wind with storage, demand response or existing hydro so that its energy is usable when the system needs it. In a jurisdiction with large reservoir hydro — British Columbia, Manitoba, Quebec, Newfoundland and Labrador — that hydro is the natural partner for wind and shifts the balance decisively towards a wind-plus-storage build; in a hydro-poor system such as Ontario, Saskatchewan or New Brunswick, the nuclear case is much the stronger. The engineering conclusion is that capacity and energy are different products, and 2000 MW of wind nameplate is simply not the same thing as 2000 MW of nuclear nameplate.