Korea's 24-month refuelling pilot would add about 49 megawatts of steady reactor output

The case rests on a 2022 Pusan National University study of the OPR1000, Korea's standard reactor design. With 60 days of planned maintenance assumed under either cycle, the most economical 24-month fuel loading produced output equal to 333.7 days a year at full capacity. The 18-month cycle produced 324.8 days ^[6]. Those 8.9 days, times 24 hours, times two reactors of 1 GW each, come to about 427 GWh a year ^[13]. Sedaily, which reported the plan, says that is enough for about 120,000 Seoul households ^[7].

Divided by the 8,760 hours in a year, 427 GWh is about 49 MW of continuous output ^[14]. Park Jong-bae, a professor of electrical and electronics engineering at Konkuk University, made the same conversion from a rounder figure. "Converted into a facility producing 400 GWh continuously throughout the year, it is similar to the annual generation of a plant of about 45 megawatts," he said ^[8]. Peak capacity and annual energy measure different things, so any comparison with the demand forecast is rough. Planners added at least 26.6 GW to the 2040 peak ^[4]. The pilot equals about 0.18% of that, a gap of more than 500 to one ^[15].

On the study's numbers, each converted reactor's capacity factor rises from about 89.0% to about 91.4% ^[16], and the total gain scales with the number of units that convert. Sedaily's report does not give a fleet-wide schedule, a count of eligible reactors or the value of the design contract. If the programme stops at two pilots, it adds 49 MW. Rolled across the fleet, long cycles add output from 2030, seven years before the first new reactor in the 11th Basic Plan ^[18]. The third outcome is the one Korea has already reached twice ^[9]. KEPCO Nuclear Fuel hired Westinghouse for 24-month core design consulting in 2017, and Korea Hydro & Nuclear Power's 2019 work on Hanul 3 and 4 never reached a domestic plant ^[9].

I think the pilot's most useful output is a measured outage length. The study's gain holds only if maintenance stays near 60 days on the longer cycle ^[6]. A 24-month cycle averages half an outage a year ^[5], so every extra outage day costs half a day of output. The 8.9-day gain is gone once the outage reaches about 78 days ^[17]. The counter-case is that reactors abroad already run the cycle ^[10]. On that view the outage data already exists, and the slow part in Korea is the civil and piping design now under contract ^[2]. If the Korean pilot reports outages near 78 days, the supply plan falls back on new build and on the continued-operation reviews the Nuclear Safety and Security Commission plans to streamline ^[12].