DeepOPF-AL: Augmented Learning for Solving AC-OPF Problems with Multiple Load-Solution Mappings

TL;DR

DeepOPF-AL introduces an augmented learning method that enables neural networks to effectively learn unique mappings for AC-OPF problems with multiple load-solution mappings, resulting in faster and more optimal solutions.

Contribution

The paper proposes DeepOPF-AL, a novel augmented-learning approach that improves neural network performance in solving non-convex AC-OPF problems with multiple mappings.

Findings

Achieves better optimality than recent DNN schemes.

Maintains similar feasibility and speedup performance.

Requires higher training complexity but uses the same DNN size.

Abstract

The existence of multiple load-solution mappings of non-convex AC-OPF problems poses a fundamental challenge to deep neural network (DNN) schemes. As the training dataset may contain a mixture of data points corresponding to different load-solution mappings, the DNN can fail to learn a legitimate mapping and generate inferior solutions. We propose DeepOPF-AL as an augmented-learning approach to tackle this issue. The idea is to train a DNN to learn a unique mapping from an augmented input, i.e., (load, initial point), to the solution generated by an iterative OPF solver with the load and initial point as intake. We then apply the learned augmented mapping to solve AC-OPF problems much faster than conventional solvers. Simulation results over IEEE test cases show that DeepOPF-AL achieves noticeably better optimality and similar feasibility and speedup performance, as compared to a recent DNN scheme, with the same DNN size yet elevated training complexity.