Publication Type

Journal Article

Version

acceptedVersion

Publication Date

6-2024

Abstract

Advances in machine learning and increased computational power have driven progress in energy-related research. However, limited access to private energy data from buildings hinders traditional regression models relying on historical data. While generative models offer a solution, previous studies have primarily focused on short-term generation periods (e.g., daily profiles) and a limited number of meters. Thus, the study proposes a conditional diffusion model for generating high-quality synthetic energy data using relevant metadata. Using a dataset comprising 1,828 power meters from various buildings and countries, this model is compared with traditional methods like Conditional Generative Adversarial Networks (CGAN) and Conditional Variational Auto-Encoders (CVAE). It explicitly handles long-term annual consumption profiles, harnessing metadata such as location, weather, building, and meter type to produce coherent synthetic data that closely resembles real-world energy consumption patterns. The results demonstrate the proposed diffusion model's superior performance, with a 36% reduction in Fréchet Inception Distance (FID) score and a 13% decrease in Kullback-Leibler divergence (KL divergence) compared to the following best method. The proposed method successfully generates high-quality energy data through metadata, and its code will be open-sourced, establishing a foundation for a broader array of energy data generation models in the future.

Keywords

Generative models, Building energy, Smart meter, Deep learning, Diffusion model, Computer vision

Discipline

Energy Policy | Engineering

Research Areas

Integrative Research Areas

Publication

Energy and Buildings

Volume

312

First Page

1

Last Page

14

ISSN

0378-7788

Identifier

10.1016/j.enbuild.2024.114216

Publisher

Elsevier

Additional URL

https://doi.org/10.1016/j.enbuild.2024.114216

Download

Share

COinS