Bucketized Active Sampling for Learning ACOPF

TL;DR

This paper introduces Bucketized Active Sampling, an active learning method that efficiently trains high-fidelity OPF proxies by partitioning input space and adaptively selecting samples within a time limit.

Contribution

The paper proposes a novel active learning framework, BAS, that improves OPF proxy training efficiency by partitioning input space and using adaptive sampling strategies.

Findings

BAS reduces the number of OPF solves needed for training.

BAS achieves higher proxy fidelity within time constraints.

Experimental results validate BAS's effectiveness.

Abstract

This paper considers optimization proxies for Optimal Power Flow (OPF), i.e., machine-learning models that approximate the input/output relationship of OPF. Recent work has focused on showing that such proxies can be of high fidelity. However, their training requires significant data, each instance necessitating the (offline) solving of an OPF. To meet the requirements of market-clearing applications, this paper proposes Bucketized Active Sampling (BAS), a novel active learning framework that aims at training the best possible OPF proxy within a time limit. BAS partitions the input domain into buckets and uses an acquisition function to determine where to sample next. By applying the same partitioning to the validation set, BAS leverages labeled validation samples in the selection of unlabeled samples. BAS also relies on an adaptive learning rate that increases and decreases over time. Experimental results demonstrate the benefits of BAS.