julia-jump

Translate AMPL to Julia with JuMP

Note: the codes assume you use Gurobi

train-*.jl: proto-codes for implementing a policy

• train-perfectforesight.jl: implements perfect foresight policy for a deterministic problem
• train-cempc.jl: implements certainty-equivalent MPC
• train-sbrmpc.jl: implements scenario-based robust MPC

parallel-*.jl: main-codes for implementing a policy in parallel

prarallel/: dir which contains the MPC functions or data loading functions

• parallel-run-perfectforesight.jl: implements perfect foresight policy for a deterministic problem in parallel
• parallel-run-cempc.jl: implements certainty-equivalent MPC in parallel
• parallel-run-sbrmpc.jl: implements scenario-based robust MPC in parallel

src/: source codes

• source.jl: functions used in the policies such as reading data files

data/: two-layer or multi-layer problem of my master thesis

• *.json: contains the corresponding stochastic parameters
• *.csv: contains (deterministic) problem parameters

The stochastic dual dynamic programming (SDDP)

One can implement the SDDP with running RunSDDP.jl.

• RunSDDP.jl: runs the SDDP
• SDDP.jl: functions used for SDDP implementation
• LoadDataSDDP.jl: loads data for SDDP implementation