growthregion deployment of reactors with finite lifetimes

[Daniel Wojtaszek]

Has anyone tried running a simulation where the deployment of reactors with finite lifetimes is governed by growthregion?  I tried doing this but I'm finding that the replacement of reactors at the end of their lifetime results in the total generated power dropping below the demand.

Cheers.

Dan

[Daniel Wojtaszek]

Here are the input and database output files for a case that demonstrates the drop in power due to reactor decommissioning in the growth region.

I've also noticed that the timeseriesdemandfresh_htgr_comm table in cyclus.sqlite has 35 rows for each reactor requesting a whole core load of fresh fuel (5914 kg), which is not correct.  Each of the 35 rows per reactor should be ~169 kg of fresh fuel demanded.  There are 35 assemblies in each core, where each assembly is ~169 kg.  The transactions table shows that the correct mass of fresh fuel is actually being sent to each reactor.

Cheers.

Dan

[Jin whan Bae]

That's a good point. I think that's an error on the code (https://github.com/cyclus/cycamore/blob/master/src/reactor.cc#L246). The ``assem_size * n_assem_order`` should be ``assem_size``, or the function call should be outside of the for loop. You can either make an issue or a PR to fix this. I don't think growth region takes into account the decommissioned facilities. May I perhaps direct you to check out d3ploy (https://github.com/arfc/d3ploy)? It is an institution archetype created to do deployment, so maybe this can be of help. Also, there is `cyclus_input_gen` (https://github.com/arfc/cyclus_input_gen), which is an external script that generates a `deployInst` input block that can satisfy your demand equation.

Let me know if you have any further questions!

Thanks,

[Daniel Wojtaszek]

Thanks! I've made an issue for the timeseriesdemand table issue.  I'll check out d3ploy and cyclus_input_gen.

Cheers.

Dan