gwenchee / ddca_numerical_exp

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

ddca_numerical_exp

Numerical Experiment for CYCLUS Demand-Driven Deployment

The repository is part of an effort to add demand-driven deployment capabilities into the CYCLUS framework. PI: Kathryn Huff

Assume the algorithm is an INSTITUTION.

GIVEN:

all prototype definitions and parameters
    - reprocessing capacity of one facility
    - reprocessing efficiency
    - reactor specifications
    - enrichment capacity / tails assay
    - spent / fresh fuel compositions

Non-optimizing

  • If it `predicts' the fuel demand correctly ( given the past, )
  • Does all the reactors run? (timeseriespower =! 0 if not in refueling)
  • Does it deploy facilities when it sees we need more capacity?
  • Does it track demand?

Deterministic-Optimizing

  • Are the constraints met?
  • The objective function is maximized?
  • Same result every time?

Commodity Flow

Power / Advanced Reactor Deployment -> mox_fuel (fuel for advanced reactors) -> spent fuel (Pu / Fissile) -> Reprocessing Capacity -> uox -> uox fabrication capacity -> uox enrichment capacity -> natural u

Example Objective Function

2% nuclear electricity generation annually, starting from x P = x*(1.02)^(t/12)

About


Languages

Language:Python 63.5%Language:C++ 36.5%