Generative Modeling and Data Augmentation for Power System Production Simulation

Linna Xu, Yongli Zhu
12/10/2024
eess.SYcs.AIcs.LG

No official code found. Best implementation: Becklishious/NeurIPS2024 (1 star, last commit 2024-12-13).

Abstract

As a key component of power system production simulation, load forecasting is critical for the stable operation of power systems. Machine learning methods prevail in this field. However, the limited training data can be a challenge. This paper proposes a generative model-assisted approach for load forecasting under small sample scenarios, consisting of two steps: expanding the dataset using a diffusion-based generative model and then training various machine learning regressors on the augmented dataset to identify the best performer. The expanded dataset significantly reduces forecasting errors compared to the original dataset, and the diffusion model outperforms the generative adversarial model by achieving about 200 times smaller errors and better alignment in latent data distributions.

View on arXivView PDF

Code Implementations(1)

Becklishious/NeurIPS2024Community100%
11Python, Jupyter NotebookOct 14, 20241 years ago

Similar Papers

From Shared Demand Patterns to Local Uncertainty: Probabilistic Load Forecasting by Mixing Compact Adaptations

83% match2026

Deep Generative Methods for Producing Forecast Trajectories in Power Systems

83% match2023

TimeGPT in Load Forecasting: A Large Time Series Model Perspective

83% match2024

Few-Shot Load Forecasting Under Data Scarcity in Smart Grids: A Meta-Learning Approach

83% match2024

Load Forecasting in the Era of Smart Grids: Opportunities and Advanced Machine Learning Models

83% match2025

Analysis of Weather and Time Features in Machine Learning-aided ERCOT Load Forecasting

82% match2023

Cite this paper

@article{xu2024generative,
  title  = {Generative Modeling and Data Augmentation for Power System Production Simulation},
  author = {Linna Xu and Yongli Zhu},
  year   = {2024},
  eprint = {2412.12146},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2412.12146v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2412.12146.svg)](https://codeofpaper.com/paper/2412.12146)

Discussion