Neural Network Optimal Power Flow via Energy Gradient Flow and Unified Dynamics

TL;DR

This paper introduces a novel neural network approach for solving optimal power flow problems by transforming them into energy minimization tasks, enabling unsupervised, physics-constrained learning that improves efficiency and physical consistency.

Contribution

It proposes an energy gradient flow-based neural network method that unsupervisedly learns optimal power flow solutions without pre-labeled data, ensuring physical constraints are satisfied.

Findings

Achieves physics-consistent solutions via energy minimization.

Unsupervised training reduces reliance on pre-solved datasets.

Improves computational efficiency over traditional methods.

Abstract

Optimal Power Flow (OPF) is a core optimization problem in power system operation and planning, aiming to minimize generation costs while satisfying physical constraints such as power flow equations, generator limits, and voltage limits. Traditional OPF solving methods typically employ iterative optimization algorithms (such as interior point methods, sequential quadratic programming, etc.), with limitations including low computational efficiency, initial value sensitivity, and low batch computation efficiency. Most existing deep learning-based OPF methods rely on supervised learning, requiring pre-solving large numbers of cases, and have difficulty guaranteeing physical consistency. This paper proposes an Optimal Power Flow solving method based on neural network dynamics and energy gradient flow, transforming OPF problems into energy minimization problems. By constructing an energy function to measure the degree of deviation from the constraint manifold, and guiding networks to learn optimal solutions that simultaneously satisfy power flow constraints and minimize costs through gradient flow. Neural networks are trained unsupervised by directly minimizing physical residuals, requiring no labeled data, achieving true "end-to-end" physics-constrained learning.