arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.01207eess.SYcs.AIcs.SY
收藏

具有物理约束的神经动力学:一种统一的流形框架用于大规模功率流计算

Physics-Constrained Neural Dynamics: A Unified Manifold Framework for Large-Scale Power Flow Computation

Shanghai Jiao Tong University(上海交通大学)

Xuezhi Liu

AI总结:

本文提出一种基于流形几何和梯度流的神经物理功率流求解方法,通过无监督训练实现端到端的物理约束学习。

摘要:

功率流分析是电力系统分析、规划和运行控制中的基本工具。传统牛顿-拉夫森方法存在初始值敏感性和批量计算效率低的局限,而现有的基于深度学习的功率流求解器大多依赖于监督学习,需要预解决大量案例,难以保证物理一致性。本文提出了一种基于流形几何和梯度流的神经物理功率流求解方法,通过将功率流方程描述为一个约束流形,并构建能量函数$V(\mathbf{x}) = \frac{1}{2}\\\\|\mathbf{F}(\mathbf{x})\\\\|^2$和梯度流$\frac{d\mathbf{x}}{dt} = -\nabla V(\mathbf{x})$,将功率流求解转化为动态系统平衡点寻找问题。神经网络通过直接最小化物理残差以无监督的方式进行训练,无需标注数据,实现真正的\"端到端\"物理约束学习。

英文摘要:

Power flow analysis is a fundamental tool for power system analysis, planning, and operational control. Traditional Newton-Raphson methods suffer from limitations such as initial value sensitivity and low efficiency in batch computation, while existing deep learning-based power flow solvers mostly rely on supervised learning, requiring pre-solving of numerous cases and struggling to guarantee physical consistency. This paper proposes a neural physics power flow solving method based on manifold geometry and gradient flow, by describing the power flow equations as a constraint manifold, and constructing an energy function \(V(\mathbf{x}) = \frac{1}{2}\|\mathbf{F}(\mathbf{x})\|^2\) and gradient flow \(\frac{d\mathbf{x}}{dt} = -\nabla V(\mathbf{x})\), transforming power flow solving into an equilibrium point finding problem for dynamical systems. Neural networks are trained in an unsupervised manner by directly minimizing physical residuals, requiring no labeled data, achieving true "end-to-end" physics-constrained learning.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑