Session
Session 3 2022
Start Date
10-26-2022 12:00 AM
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Noman S.A., Hussain, A., Danish, M., and Ali, R. (2022). "Prediction of Discharge Coefficient of Circular Side Orifice Through Machine Learning Technique" in "9th IAHR International Symposium on Hydraulic Structures (9th ISHS)". Proceedings of the 9th IAHR International Symposium on Hydraulic Structures – 9th ISHS, 24-27 October 2022, IIT Roorkee, Roorkee, India. Palermo, Ahmad, Crookston, and Erpicum Editors. Utah State University, Logan, Utah, USA, 10 pages (DOI: 10.26077/1d18-ff7d) (ISBN 978-1-958416-07-5).
Abstract
A sharp-crested circular side orifice is a crucial element when it comes to diverting flow from primary source to its subordinate source. Such a flow measurement instrument technique is of immense value in conservation and evaluation of drainage and irrigation networks. Usually, it is placed towards the side of a channel in order to regulate the flow of the fluid. Traditionally, coefficient of discharge was predicted through regression methods which are time-consuming and lack accuracy. Artificial Intelligence (AI) and its applications in this domain have bridged this gap by providing novel alternative methods which prove much more efficient. Repeated studies have pointed out that AI techniques generally give better results when it comes to a myriad of water variables such as rainfall-runoff, evaporation and evapotranspiration, streamflow, and dam water level changes. Total 261 dataset has been collected from the literature review comprising of the fully submerged orifice and for partially-submerged orifice with varying orifice diameter (D) of 5 cm, 10 cm and 15 cm. This study aims to provide a better estimate of prediction of discharge through circular sharp-crested orifice using Artificial Neural Network (ANN). The ANN model has been deployed to randomly select 80% of the data for training, 15% for validation and remaining 5% for testing. In the ANN model, Lavenberg-Marquardt algorithm was used as back-propagation step to assign weights in order to predict the output. The correlation coefficient (R), mean absolute error (MAE) and root mean squared error (RMSE) for complete data of fully and partially submerged circular side orifice are observed to be 0.9765, 0.0228 and 0.0172 respectively.
Prediction of Discharge Coefficient of Circular Side Orifice Through Machine Learning Technique
A sharp-crested circular side orifice is a crucial element when it comes to diverting flow from primary source to its subordinate source. Such a flow measurement instrument technique is of immense value in conservation and evaluation of drainage and irrigation networks. Usually, it is placed towards the side of a channel in order to regulate the flow of the fluid. Traditionally, coefficient of discharge was predicted through regression methods which are time-consuming and lack accuracy. Artificial Intelligence (AI) and its applications in this domain have bridged this gap by providing novel alternative methods which prove much more efficient. Repeated studies have pointed out that AI techniques generally give better results when it comes to a myriad of water variables such as rainfall-runoff, evaporation and evapotranspiration, streamflow, and dam water level changes. Total 261 dataset has been collected from the literature review comprising of the fully submerged orifice and for partially-submerged orifice with varying orifice diameter (D) of 5 cm, 10 cm and 15 cm. This study aims to provide a better estimate of prediction of discharge through circular sharp-crested orifice using Artificial Neural Network (ANN). The ANN model has been deployed to randomly select 80% of the data for training, 15% for validation and remaining 5% for testing. In the ANN model, Lavenberg-Marquardt algorithm was used as back-propagation step to assign weights in order to predict the output. The correlation coefficient (R), mean absolute error (MAE) and root mean squared error (RMSE) for complete data of fully and partially submerged circular side orifice are observed to be 0.9765, 0.0228 and 0.0172 respectively.