Predicting membrane strains with a deep learning encoder-decoder convolutional neural network architecture trained on synthetic finite element data
Benjamin Steven Vien, Thomas Kuen, Louis Raymond Francis Rose, Wing Kong Chiu
Abstract. Melbourne Water Corporation’s Western Treatment Plant, located in Werribee, Victoria, Australia, is pioneering the use of advanced technologies, including artificial intelligence, to enhance asset management practices. This is particularly important for the structural health monitoring (SHM) of anaerobic lagoon floating covers, which are crucial for biogas collection but can be compromised by scum accumulation underneath. This study introduces a novel encoder-decoder network within a convolutional neural network (CNN) framework, designed to predict strain distributions on deformed membranes. Due to the limited availability of real-world data, finite element analysis was utilised to generate synthetic data consisting of displacements and strain fields for model training. The study investigates the optimal quantity of synthetic samples needed for accurate predictions and discusses the proposed CNN architecture and data preparation techniques. The findings indicate that a dataset of at least 10,000 synthetic training samples is required to accurately predict strain distributions, which represents a significant improvement by orders of magnitude compared to using only 100 and 1000 samples. Furthermore, refinement learning methods were demonstrated, where a pretrained CNN model is further trained on a new dataset with lower strain variability. The results indicate that refining the pretrained model with frozen (fixed) weights in the encoder network yields better accuracy in predicting strain, at least 2.3 times better than those without frozen weights. However, the refined model without frozen weights retains more information from the original dataset and is more consistent in predicting strain profiles. The results suggest a high-quality, representative training dataset relating to the application of interest is essential for effective machine learning. These findings lay a fundamental basis for implementing practical deep learning approaches and further utilising unmanned aerial vehicle-based imagery for effective SHM of highly valuable assets.
Keywords
Deep Learning, Membrane, Strain, Convolution Neural Network, Synthetic Data, Finite Element
Published online 3/25/2025, 12 pages
Copyright © 2025 by the author(s)
Published under license by Materials Research Forum LLC., Millersville PA, USA
Citation: Benjamin Steven Vien, Thomas Kuen, Louis Raymond Francis Rose, Wing Kong Chiu, Predicting membrane strains with a deep learning encoder-decoder convolutional neural network architecture trained on synthetic finite element data, Materials Research Proceedings, Vol. 50, pp 61-72, 2025
DOI: https://doi.org/10.21741/9781644903513-7
The article was published as article 7 of the book Structural Health Monitoring
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 license. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.
References
[1] Rowe, R.K. and H.P. Sangam, Durability of HDPE geomembranes. Geotextiles and Geomembranes, 2002. 20(2): p. 77-95. https://doi.org/10.1016/S0266-1144(02)00005-5
[2] Wong, L., B.S. Vien, Y. Ma, T. Kuen, F. Courtney, J. Kodikara, F. Rose, and W.K. Chiu. Development of Scum Geometrical Monitoring Beneath Floating Covers Aided by UAV Photogrammetry. in Mater. Res. Proc. 2021. https://doi.org/10.3390/rs12071118
[3] Wong, L., B.S. Vien, Y. Ma, T. Kuen, F. Courtney, J. Kodikara, and W.K. Chiu, Remote Monitoring of Floating Covers Using UAV Photogrammetry. Remote Sensing, 2020. 12(7): p. 1118. https://doi.org/10.3390/rs12071118
[4] Vien, B.S., L. Wong, T. Kuen, and W.K. Chiu, A Machine Learning Approach for Anaerobic Reactor Performance Prediction Using Long Short-Term Memory Recurrent Neural Network. Structural Health Monitoring: 8APWSHM, 2021. 18: p. 61. https://doi.org/10.21741/9781644901311-8
[5] Vien, B.S., L. Wong, T. Kuen, R.L.R. Francis, and W.K. Chiu, Probabilistic prediction of anaerobic reactor performance using Bayesian long short-term memory artificial recurrent Neural Network Model, in International Workshop on Structural Health Monitoring (IWSHM) 2021: Enabling Next-Generation SHM for Cyber-Physical Systems. 2021, DEStech Publications, Inc. p. 813-820. https://doi.org/10.12783/shm2021/36331
[6] Vien, B.S., T. Kuen, L.R.F. Rose, and W.K. Chiu, Optimisation and Calibration of Bayesian Neural Network for Probabilistic Prediction of Biogas Performance in an Anaerobic Lagoon. Sensors, 2024. 24(8): p. 2537. https://doi.org/10.3390/s24082537
[7] Vien, B.S., T. Kuen, L.R.F. Rose, and W.K. Chiu, Image Segmentation and Filtering of Anaerobic Lagoon Floating Cover in Digital Elevation Model and Orthomosaics Using Unsupervised k-Means Clustering for Scum Association Analysis. Remote Sensing, 2023. 15(22): p. 5357. https://doi.org/10.3390/rs15225357
[8] Nie, Z., H. Jiang, and L.B. Kara. Stress field prediction in cantilevered structures using convolutional neural networks. in International Design Engineering Technical Conferences and Computers and Information in Engineering Conference. 2019. American Society of Mechanical Engineers. https://doi.org/10.1115/DETC2019-98472
[9] Yang, R., Y. Li, D. Zeng, and P. Guo, Deep DIC: Deep learning-based digital image correlation for end-to-end displacement and strain measurement. Journal of Materials Processing Technology, 2022. 302: p. 117474. https://doi.org/10.1016/j.jmatprotec.2021.117474
[10] Ronneberger, O., P. Fischer, and T. Brox. U-net: Convolutional networks for biomedical image segmentation. in Medical Image Computing and Computer-Assisted Intervention–MICCAI 2015: 18th International Conference, Munich, Germany, October 5-9, 2015, Proceedings, Part III 18. 2015. Springer.
[11] Vien, B.S., (2024). bensvien/ARC-Linkage-2024: ARC Linkage Project APWSHM 2024 Strain CNN Version 1.2 (Version APWSHM2024). Zenodo. https://doi.org/10.5281/zenodo.14188874
