Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks
Journal of Imaging · 14.03.2023 · 9 (3) · 68
DOI: 10.3390/jimaging9030068 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 14.03.2023. Verfasst von André Antunes und 3 weiteren Autor:innen. Beteiligte Einrichtung: Sustain.RD, Escola Superior de Tecnologia de Setúbal, Instituto Politécnico de Setúbal, 2914-508 Setúbal, Portugal. Die Arbeit wurde bisher 20-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Fundação para a Ciência e Tecnologia, Fundação para a Ciência e a Tecnologia.
Abstract (Originalsprache)
The current paper presents a hyper parameterization optimization process for a convolutional neural network (CNN) applied to pipe burst locations in water distribution networks (WDN). The hyper parameterization process of the CNN includes the early stopping termination criteria, dataset size, dataset normalization, training set batch size, optimizer learning rate regularization, and model structure. The study was applied using a case study of a real WDN. Obtained results indicate that the ideal model parameters consist of a CNN with a convolutional 1D layer (using 32 filters, a kernel size of 3 and strides equal to 1) for a maximum of 5000 epochs using a total of 250 datasets (using data normalization between 0 and 1 and tolerance equal to max noise) and a batch size of 500 samples per epoch step, optimized with Adam using learning rate regularization. This model was evaluated for distinct measurement noise levels and pipe burst locations. Results indicate that the parameterized model can provide a pipe burst search area with more or less dispersion depending on both the proximity of pressure sensors to the burst or the noise measurement level.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Sustain.RD, Escola Superior de Tecnologia de Setúbal, Instituto Politécnico de Setúbal, 2914-508 Setúbal, Portugal.
- NOVA LINCS, Department of Computer Science, Faculdade de Ciências e Tecnologia, Universidade NOVA de Lisboa, 2829-516 Caparica, Portugal.
- INCITE, Escola Superior de Tecnologia do Barreiro, Instituto Politécnico de Setúbal, 2839-001 Lavradio, Portugal.
ORCID-Profile: André Antunes, Bruno Ferreira, Nelson Carriço
Förderung
- Fundação para a Ciência e Tecnologia
- Fundação para a Ciência e a Tecnologia
Zitieren
APA
Antunes, A., Ferreira, B., Marques, N., & Carriço, N. (2023). Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks. Journal of Imaging, 9(3), 68. https://doi.org/10.3390/jimaging9030068
Vancouver
Antunes A, Ferreira B, Marques N, Carriço N. Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks. Journal of Imaging. 2023;9(3):68. doi:10.3390/jimaging9030068
BibTeX
@article{Antunes2023,
author = {Antunes, André and Ferreira, Bruno and Marques, Nuno and Carriço, Nelson},
title = {Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks},
journal = {Journal of Imaging},
year = {2023},
volume = {9},
number = {3},
pages = {68},
doi = {10.3390/jimaging9030068}
}
EUVIMED