Impact of Normalization Techniques on Synthetic Load Profile Generation Using Deep Generative Models
Synthetic load profiles are increasingly employed in power system studies as a cost-effective and privacy-preserving alternative to extensive smart meter deployments, with deep generative models (DGMs) showing promising results in capturing complex demand patterns. However, the impact of data normal...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Access |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/11121160/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849340408979521536 |
|---|---|
| author | Luis H. T. Bandoria Walquiria N. Silva Madson C. De Almeida |
| author_facet | Luis H. T. Bandoria Walquiria N. Silva Madson C. De Almeida |
| author_sort | Luis H. T. Bandoria |
| collection | DOAJ |
| description | Synthetic load profiles are increasingly employed in power system studies as a cost-effective and privacy-preserving alternative to extensive smart meter deployments, with deep generative models (DGMs) showing promising results in capturing complex demand patterns. However, the impact of data normalization on their performance remains insufficiently explored. Using datasets from a university smart grid and industrial consumers in Germany, this work systematically evaluates five normalizations techniques-Min-Max, Standard, Robust, Max-Abs, and Quantile—on four representative DGMs: Wasserstein GAN with gradient penalty (WGAN-GP), variational autoencoder (VAE), nonlinear independent component estimation (NICE), and denoising diffusion implicit models (DDIM). Additionally, this study introduces DDIM for synthetic load profile generation, providing a deterministic and faster sampling approach compared to traditional probabilistic denoising models. Results based on statistical and temporal metrics indicate that Max-Abs normalization consistently yields more accurate and stable synthetic profiles across all models and datasets, while Robust and Quantile methods often degrade essential distributional features. These findings highlight the critical role of normalization in developing effective synthetic data generation pipelines for power system applications. |
| format | Article |
| id | doaj-art-b3eede6d0c1a4514b7d53ac2ea15d1ad |
| institution | Kabale University |
| issn | 2169-3536 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Access |
| spelling | doaj-art-b3eede6d0c1a4514b7d53ac2ea15d1ad2025-08-20T03:43:55ZengIEEEIEEE Access2169-35362025-01-011314090014091310.1109/ACCESS.2025.359716011121160Impact of Normalization Techniques on Synthetic Load Profile Generation Using Deep Generative ModelsLuis H. T. Bandoria0https://orcid.org/0000-0002-1423-2529Walquiria N. Silva1https://orcid.org/0000-0002-5178-6344Madson C. De Almeida2https://orcid.org/0000-0001-6538-8569Department of Systems and Energy, Universidade Estadual de Campinas (UNICAMP), Campinas, São Paulo, BrazilInstitute of Energy and Environment, Universidade de São Paulo (USP), São Paulo, BrazilDepartment of Systems and Energy, Universidade Estadual de Campinas (UNICAMP), Campinas, São Paulo, BrazilSynthetic load profiles are increasingly employed in power system studies as a cost-effective and privacy-preserving alternative to extensive smart meter deployments, with deep generative models (DGMs) showing promising results in capturing complex demand patterns. However, the impact of data normalization on their performance remains insufficiently explored. Using datasets from a university smart grid and industrial consumers in Germany, this work systematically evaluates five normalizations techniques-Min-Max, Standard, Robust, Max-Abs, and Quantile—on four representative DGMs: Wasserstein GAN with gradient penalty (WGAN-GP), variational autoencoder (VAE), nonlinear independent component estimation (NICE), and denoising diffusion implicit models (DDIM). Additionally, this study introduces DDIM for synthetic load profile generation, providing a deterministic and faster sampling approach compared to traditional probabilistic denoising models. Results based on statistical and temporal metrics indicate that Max-Abs normalization consistently yields more accurate and stable synthetic profiles across all models and datasets, while Robust and Quantile methods often degrade essential distributional features. These findings highlight the critical role of normalization in developing effective synthetic data generation pipelines for power system applications.https://ieeexplore.ieee.org/document/11121160/Data normalizationsynthetic load profilesgenerative adversarial networksvariational autoencodersnormalizing flowsdenoising diffusion models |
| spellingShingle | Luis H. T. Bandoria Walquiria N. Silva Madson C. De Almeida Impact of Normalization Techniques on Synthetic Load Profile Generation Using Deep Generative Models IEEE Access Data normalization synthetic load profiles generative adversarial networks variational autoencoders normalizing flows denoising diffusion models |
| title | Impact of Normalization Techniques on Synthetic Load Profile Generation Using Deep Generative Models |
| title_full | Impact of Normalization Techniques on Synthetic Load Profile Generation Using Deep Generative Models |
| title_fullStr | Impact of Normalization Techniques on Synthetic Load Profile Generation Using Deep Generative Models |
| title_full_unstemmed | Impact of Normalization Techniques on Synthetic Load Profile Generation Using Deep Generative Models |
| title_short | Impact of Normalization Techniques on Synthetic Load Profile Generation Using Deep Generative Models |
| title_sort | impact of normalization techniques on synthetic load profile generation using deep generative models |
| topic | Data normalization synthetic load profiles generative adversarial networks variational autoencoders normalizing flows denoising diffusion models |
| url | https://ieeexplore.ieee.org/document/11121160/ |
| work_keys_str_mv | AT luishtbandoria impactofnormalizationtechniquesonsyntheticloadprofilegenerationusingdeepgenerativemodels AT walquiriansilva impactofnormalizationtechniquesonsyntheticloadprofilegenerationusingdeepgenerativemodels AT madsoncdealmeida impactofnormalizationtechniquesonsyntheticloadprofilegenerationusingdeepgenerativemodels |