Approximate-and-Optimize for SCOPF

Repository for An Approximate-and-Optimize Method for Security-Constrained AC Optimal Power Flow

Components

There are two parts in this repository:

• approx-optimize-scopf: Implemented in Julia, mainly containing the codes for decompose and optimize, as well as for collecting base case solutions for training the surrogate model.
• scopf-approx: Implemented in Python, mainly containing the codes for approximate, that is to train the surrogate neural network. The part for generating generating scenario datasets is also included in this module.

For more details, please refer to the README.md for each module respectively.

Acknowledgment

This research is supported by the National Key R&D Program of China (Grant No. 2022YFA1003900), Shenzhen Science and Technology Program (Grant No. RCBS20221008093309021), and National Natural Science Foundation of China (Grant No. 12301416).

Citing

• Jinxin Xiong, Shunbo Lei, Akang Wang, Xiaodong Luo. "An Approximate-and-Optimize Method for Security-Constrained AC Optimal Power Flow".