Diversifying training data does not improve generalizability of neural network models for PV identification
Journal of Renewable and Sustainable Energy · 01.11.2024 · 16 (6)
DOI: 10.1063/5.0220983 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Renewable and Sustainable Energy, online erschienen am 03.12.2024. Verfasst von Joseph Ranalli und 2 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 25 Literaturangaben.
Abstract (Originalsprache)
Data about behind-the-meter photovoltaics (PV) installations may be difficult to obtain for researchers. A number of investigators have considered deep learning as an attractive solution to this challenge, capable of directly identifying PV installations from aerial or satellite images. Deep learning models are well known to experience challenges when working with data from sources that they have never been exposed to. This study investigated whether generalizability can be improved by diversifying training data across available labeled data sources. We assessed the performance of models trained on all possible combinations of six different labeled datasets of aerial PV imagery, with a fixed number of total training images. Unfortunately, our results indicate that no combination of model training data achieved generalized performance that approaches models trained on data from a target data source. This implies that generalized ResNet models cannot be developed simply by modifying the configuration of the training data. Consequently, researchers should expect that some degree of data labeling is likely to be necessary when adapting these models to new applications, but our results do indicate that significant performance improvements are possible with only small (∼20%) introductions of target data. Future work may investigate alternative architectures, expanded training datasets, or ways to reduce the amount of labeled data necessary to adapt a model for a given application.
Zitieren
APA
Ranalli, J., Zech, M., & Tetens, H. (2024). Diversifying training data does not improve generalizability of neural network models for PV identification. Journal of Renewable and Sustainable Energy, 16(6). https://doi.org/10.1063/5.0220983
Vancouver
Ranalli J, Zech M, Tetens H. Diversifying training data does not improve generalizability of neural network models for PV identification. Journal of Renewable and Sustainable Energy. 2024;16(6). doi:10.1063/5.0220983
BibTeX
@article{Ranalli2024,
author = {Ranalli, Joseph and Zech, Matthias and Tetens, Hendrik-Pieter},
title = {Diversifying training data does not improve generalizability of neural network models for PV identification},
journal = {Journal of Renewable and Sustainable Energy},
year = {2024},
volume = {16},
number = {6},
doi = {10.1063/5.0220983}
}
EUVIMED