Prediction of Hydraulic Blockage at Culverts using Lab Scale Simulated Hydraulic Data

ABSTRACT

Blockage of culverts causes reduction in hydraulic capacity and is one of the main contributors to trigger urban flooding. However, the highly non-linear nature of debris interaction during the flood and lack of blockage-related data from actual flooding events make conventional numerical modelling almost impossible. Literature investigating blockage phenomena reports blockage as a complex hydraulic process, which suggests exploring adaptive solutions using latest technologies. In this context, motivated by the success of data-driven algorithms, in this article, four data driven models (i.e., K-NN, ANN, SVR, 1D-CNN) are implemented to predict the hydraulic blockage at culverts. A new numerical Hydraulics-Lab Blockage Dataset (HBD) is established from a series of lab-scale hydraulic experiments. From the experimental investigations, the ANN model was reported as the best with a R² score of 0.95. A potential use-case of presented research for real-world application is also discussed to demonstrate the practical feasibility.

KEYWORDS:

• Hydraulic blockage
• Learning-Based approaches
• machine learning
• physical modelling
• culverts
• artificial neural network (ANN)
• one-dimensional convolutional neural network (1D-CNN

Acknowledgements

I would like to thank the Wollongong City Council (WCC) for funding this investigation. This research was funded by the Smart Cities and Suburb Program (Round Two) of the Australian Government, grant number SCS69244. Further, I would like to thank the Higher Education Commission (HEC) of Pakistan and the University of Wollongong (UOW) for funding my PhD studies. Thanks to Professor Wanqing Li for providing technical assistance in designing the simulations.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.