Effect of continuous refueling on decay heat

wasija

Estimate the decay heat rate in a 3000 MWth reactor in which 3.2% mU-enriched
U02 assemblies are being fed into the core. The burned-up fuel stays in the core for 3 years before being replaced. Consider two cases:

1. The core is replaced in two batches every 18 months.

2. The fuel replacement is so frequent that refueling can be considered a continuous process. (Note: The PHWR reactors and some of the water-cooled graphite-moderated reactors in the Soviet Union are effectively continuously refueled.)

Compare the two situations at 1 minute , 1hour, 1day, 1month, and 1year.

I would really appreciate some help.

I am supposed to use the following equation

P/Po =0.066((t-ts)^-0.2 - t^-0.2)

But i don't understand what t is in case 1 and case 2.
Thanks

Astronuc

As far as I know, only pebble bed or liquid fueled reactors are refueled continuously.

The PHWRs, e.g., CANDUs and derivatives, and perhaps RBMKs (some of the water-cooled graphite-moderated reactors in the former Soviet Union ), can be refueled 'on-line', i.e., while the unit is operating. On-line refueling is not the same as continuous refueling.

When looking at decay heat of the fuel, one would have to consider time of operation. If one is considering decay in the core, one would have to consider each batch.

In a two batch core, one batch is operating for some time in the current cycle, while the older batch has been operating the same time as well as the time in the previous cycle.

Let's say a core has a cycle burnup of 18 GWd/tU, and is half-way through a cycle. Then, assuming the batches are sharing comparable power distribution, one batch has a burnup of 9 GWd/tU, and the older batch (half of the core) would have a burnup of 27 GWd/tU, and thus two different rates of decay power.

For a continuous refueling, one would have to look at the irradiation time for each set of fuel assemblies in the core. On-line refueling can approach continuous refueling if small sets of assemblies - e.g., 4 or 8 as might be the case in a CANDU, are added on the basis of several days, or weekly.

So rather than 1/2 core reload every 18 months (547 calendar days), one could reload ~1% of the core every day or two. I haven't run the numbers, but one would have to determine the batch fraction and reload schedule assuming some uniform period between reloads.