Closed HenriDeh closed 8 months ago
it is stated that SDDP performs well with integer variables
Note the caveat: we might find a suboptimal policy.
do you happen to have one
Here are some:
@article{laraEJOR2018,
title = {Deterministic electric power infrastructure planning: Mixed-integer programming model and nested decomposition algorithm},
journal = {European Journal of Operational Research},
volume = {271},
number = {3},
pages = {1037-1054},
year = {2018},
issn = {0377-2217},
author = {Cristiana L. Lara and Dharik S. Mallapragada and Dimitri J. Papageorgiou and Aranya Venkatesh and Ignacio E. Grossmann}
}
@article{laraSDDiP2020,
title={Electric power infrastructure planning under uncertainty: stochastic dual dynamic integer programming (SDDiP) and parallelization scheme},
author={Lara, C.L. and Siirola, J.D. and Grossmann, I.E.},
journal={Optim Eng},
volume={21},
pages={1243–1281},
year={2020}
}
@article{colonetti2023,
author={Colonetti, Bruno and Finardi, Erlon and Brito, Samuel and Zavala, Victor},
journal={IEEE Transactions on Power Systems},
title={Parallel Dual Dynamic Integer Programming for Large-Scale Hydrothermal Unit-Commitment},
year={2023},
volume={38},
number={3},
pages={2926-2938}
}
Amazing thanks a lot.
No problem. I'm going to close this issue. I don't think we need to suggest papers to cite in our documentation. (Happy to keep answering questions though.)
Hello,
In the documentation page on integrality, it is stated that SDDP performs well with integer variables, unlike SDDiP. Since I might use this approach in an article, I will have to provide a source for this claim, do you happen to have one ?
This could be included in the documentation as well.