Unsupervised Hybrid Deep Generative Models for Photovoltaic Synthetic Data Generation

This paper contributes to the field of deep generative learning applied to solar photovoltaic (PV) synthetic data generation problems by exploring Deep Generative Model (DGM) that combines Variational Autoencoders (VAE) and Generative Adversarial Networks (GAN), i.e., VAEGAN. We build upon knowledge...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:IEEE Power & Energy Society General Meeting S. 1 - 5
Hauptverfasser: Rosa de Jesus, Dan A., Mandal, Paras, Senjyu, Tomonobu, Kamalasadan, Sukumar
Format: Tagungsbericht
Sprache:Englisch
Veröffentlicht: IEEE 26.07.2021
Schlagworte:
Data models
Deep learning
Feature extraction
Generative adversarial networks
Photovoltaic systems
Privacy
Solar Photovoltaic
Synthetic Data
Variational Autoen-coders
ISSN:1944-9933
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:This paper contributes to the field of deep generative learning applied to solar photovoltaic (PV) synthetic data generation problems by exploring Deep Generative Model (DGM) that combines Variational Autoencoders (VAE) and Generative Adversarial Networks (GAN), i.e., VAEGAN. We build upon knowledge in the area of deep learning to incorporate our Hybrid Deep Neural Network (HDNN), combining convolutional and Long Short-Term Memory (LSTM) layers at the encoding level for producing robust latent representations and subsequently high-quality synthetic PV data samples. The major advantage of these approaches is that it allows the DGMs to perform better feature extraction as well as to capture the historical trends in data effectively. The simulations on actual data acquired from a real PV system demonstrate the effectiveness of the DGMs to produce high-quality samples for multiple seasons of the year.
ISSN:1944-9933
DOI:10.1109/PESGM46819.2021.9637844