Deep learning architectures for inference of AC-OPF solutions
Machine Learning
2020-12-02 v2 Systems and Control
Signal Processing
Systems and Control
Data Analysis, Statistics and Probability
Abstract
We present a systematic comparison between neural network (NN) architectures for inference of AC-OPF solutions. Using fully connected NNs as a baseline we demonstrate the efficacy of leveraging network topology in the models by constructing abstract representations of electrical grids in the graph domain, for both convolutional and graph NNs. The performance of the NN architectures is compared for regression (predicting optimal generator set-points) and classification (predicting the active set of constraints) settings. Computational gains for obtaining optimal solutions are also presented.
Cite
@article{arxiv.2011.03352,
title = {Deep learning architectures for inference of AC-OPF solutions},
author = {Thomas Falconer and Letif Mones},
journal= {arXiv preprint arXiv:2011.03352},
year = {2020}
}
Comments
5 pages, 4 tables, 3 figures. Climate Change AI Workshop - Tackling Climate Change with Machine Learning (NeurIPS 2020)