Synthetic Power Flow Data Generation Using Physics-Informed Denoising Diffusion Probabilistic Models

2025-04-24

AI Summary

To address the severe scarcity of high-quality power flow data in smart grids—stemming from privacy constraints and operational limitations—this paper proposes a physics-informed denoising diffusion probabilistic model (DDPM) framework. Our method innovatively formulates power flow equations as a differentiable physics-informed loss term and incorporates a multi-task auxiliary training strategy to jointly ensure statistical fidelity and physical feasibility (e.g., strict satisfaction of power flow equations). Experiments on IEEE 14- and 30-bus systems demonstrate that the generated samples achieve 99.7% feasibility, reduce key statistical feature errors by over 42%, and exhibit significantly stronger out-of-distribution generalization than three baseline models. To our knowledge, this is the first work to jointly optimize both physical consistency and statistical quality in DDPM-based synthetic power system data generation.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Probabilistic Circuits and Graphical Models

Application Category

• User Modeling, Personalization and Recommendation: User privacy protection in personalized systems
• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems

Abstract

Many data-driven modules in smart grid rely on access to high-quality power flow data; however, real-world data are often limited due to privacy and operational constraints. This paper presents a physics-informed generative framework based on Denoising Diffusion Probabilistic Models (DDPMs) for synthesizing feasible power flow data. By incorporating auxiliary training and physics-informed loss functions, the proposed method ensures that the generated data exhibit both statistical fidelity and adherence to power system feasibility. We evaluate the approach on the IEEE 14-bus and 30-bus benchmark systems, demonstrating its ability to capture key distributional properties and generalize to out-of-distribution scenarios. Comparative results show that the proposed model outperforms three baseline models in terms of feasibility, diversity, and accuracy of statistical features. This work highlights the potential of integrating generative modelling into data-driven power system applications.

Problem

• Generating synthetic power flow data with privacy constraints
• Ensuring data feasibility using physics-informed diffusion models
• Improving statistical fidelity in power system applications

Innovation

• Physics-informed DDPMs for power flow data
• Auxiliary training ensures statistical fidelity
• Outperforms baselines in feasibility and diversity