Estimation of air concentration in chute spillway using metamodel methods

Document Type : Research Paper

Authors

1 Professor, Department of Civil Engineering, Faculty of Civil Engineering, University of Tabriz, Tabriz, Iran

2 Department of Civil Engineering, Faculty of Civil Engineering, University of Tabriz, Tabriz, Iran

Abstract

One of the ways to prevent creating negative pressure and cavitation in spillways is to introduce air into the flow over the spillways. Understanding the distribution of air concentration variations along the spillway is of significant importance for estimating the aeration level. This study explores the application of GPR and SVM molels in predicting air concentration. To achieve this, a dataset of 2268 laboratory experiments obtained from hydraulic models of chute spillways was utilized in the modeling process. Various input models were defined based on different combinations of measured parameters. The results demonstrate the high capability of both methods in estimating the required air concentration over the spillway. In predicting air concentration in the chute spillway under artificial aeration conditions, flow discharge (QW), longitudinal distance ratio from the end of the deflector to the channel width (L/W), and depth ratio (perpendicular to the spillway) to channel width (Y/W) significantly influenced the outcomes. Statistical indices, including R, DC, and RMSE for this case were 0.9214, 0.8451, and 1.008, respectively, in the GPR, and 0.9333, 0.8662, and 0.937 in the SVM. For scenarios without artificial aeration, the model with input parameters QW, L/W, Y/W, and ΔP (pressure difference between atmospheric pressure and the pressure under the jet) achieved the best performance in the GPR method with values of R=0.9222, DC=0.8644, and RMSE=0.914. In the SVM, the same model with values of 0.87, 0.7543, and 0.123 for R, DC, and RMSE, respectively, was selected as the superior model.

Keywords

Main Subjects


Estimation of air concentration in chute spillway using metamodel methods

EXTENDED ABSTRACT

Introduction

Water supply for various agricultural, industrial, and particularly drinking purposes holds significant importance, especially in arid and semi-arid regions. In recent decades, with technological advancements, the construction of reservoirs and hydraulic structures has expanded considerably. Given the increasing trend in building tall dams and the need to enhance dam safety, hydraulic engineers are increasingly focusing on the economic and reliable design of these structures. One hydraulic structure playing a crucial role in ensuring sufficient dam safety during flood inflow into the reservoir and discharge downstream is spillways. Due to the long-standing use of hydraulic models in spillway design and the considerable accuracy of their results, engineers are compelled to construct these models and conduct experiments to achieve more economical and safer spillway designs. Due to the high flow velocity and pressure drop in tall spillways, the potential for cavitation phenomena has increased, leading to severe damage to the spillway. Therefore, a specialized structure called an aerator, also known as an air vent, must be incorporated in locations where natural aeration from the free surface of the flow is insufficient for proper protection. Understanding the distribution of air concentration variations along the spillway is crucial for estimating aeration levels. Introducing innovative and creative methods that can effectively address this issue holds immense importance in this context.

Materials and Methods

Here, Support Vector Machines and Gaussian Process Regression methods were employed as modeling approaches for parameter estimation, and their results were subjected to evaluation. Various models were defined with different input parameters, aiming to estimate air concentration. The concentration of air pollutants was predicted, and the influential parameters for each case were identified. This research endeavors to contribute to previous studies by demonstrating the application of Support Vector Machines and Gaussian Process Regression in estimating air concentration distribution and assessing the practicality of these methods in comparison to laboratory experimental data in this domain. For assessing the performance of the Support Vector Machine and Gaussian Process Regression methods in predicting air concentration distribution in spillways, experimental data from Chanson (1988) comprising 2268 data points have been utilized.

Results and Discussions

To investigate the impact of various parameters on the estimation of air concentration distribution, sensitivity analysis has been employed. This involved systematically excluding superior model parameters and re-running the model using Support Vector Machines and Gaussian Process Regression. Evaluation metrics (R, DC, RMSE and KGE) were calculated, and the influence of the removed parameters was examined. According to the sensitivity analysis results, parameters Y/W and L/W play a crucial role in estimating air concentration distribution along the spillway. Moreover, the exclusion of the Qw parameter has a minor impact on the results. It is observed that the L/W parameter is significant in predicting air concentration along the spillway. Similar to scenario 1, the removal of the Qw parameter has a negligible effect on the final results.

Conclusion

Here, a wide range of experimental data was utilized to examine the modeling capabilities in predicting the required air concentration in the spillway of the Clyde Dam. For this purpose, 2268 experimental data obtained from the hydraulic model of the Clyde Dam were employed in the modeling process. Multiple models were developed based on different combinations of measured parameters, and by analyzing the obtained results, the impact of each parameter on estimating air concentration was determined. The results indicated that both Support Vector Machine (SVM) and Gaussian Process Regression (GPR) methods exhibit high accuracy in solving the specified problem. In the application of Support Vector Regression, among various kernel functions, the Radial Basis Function (RBF) kernel yielded the best results in estimating the required air concentration in spillways. The estimation of air concentration in the spillway, particularly for scenarios where artificial aeration is not performed (1113 data), resulted in a model with four input parameters, including Qw, L/W, Y/W, and ΔP. This model, with values of R=0.9229, DC=0.8644, and RMSE=0.0914 in Gaussian Process Regression, and R=0.87, DC=0.7543, and RMSE=0.123 in Support Vector Machine, was selected as the superior model. For the estimation of air concentration in scenarios where artificial aeration is conducted (1155 data points), a model with three input parameters, Qw, L/W, and Y/W, demonstrated superior performance. This model, with values of R=0.9214, DC=0.8451, and RMSE=0.1008 in Gaussian Process Regression, and R=0.9333, DC=0.8662, and RMSE=0.0937 in Support Vector Machine, was chosen as the top-performing model. According to the obtained results, it is evident that SVM and GPR methods can serve as reliable approaches with acceptable and desirable results for estimating the required air concentration.

Abbaszadeh, H., Norouzi, R., Sume, V., Kuriqi, A., Daneshfaraz, R., & Abraham, J. (2023a). Sill role effect on the flow characteristics (experimental and regression model analytical). Fluids, 8(8), 235.
Abbaszadeh, H., Daneshfaraz, R., & Norouzi, R. (2023b). Experimental investigation of hydraulic jump parameters in sill application mode with various synthesis. Journal of Hydraulic Structures, 9(1), 18-42.
Abbaszadeh, H., Daneshfaraz, R., Sume, V., & Abraham, J. (2024). Experimental investigation and application of soft computing models for predicting flow energy loss in arc-shaped constrictions. AQUA—Water Infrastructure, Ecosystems and Society, 73(3), 637-661.
Chakib, B. (2013). Numerical computation of inception point location for flat-sloped stepped spillway. International Journal of Hydraulic Engineering, 2(3), 4752.
Chanson, H. (1988). Study of air entrainment and aeration devices on spillway model. A thesis submitted in partial fulfilment of the requirements for the Degree of Doctor of Philosophy in Civil Engineering at the University of Canterbury, New Zealand.
Daneshfaraz, R., Norouzi, R., Abbaszadeh, H., Kuriqi, A., & Di Francesco, S. (2022a). Influence of sill on the hydraulic regime in sluice gates: an experimental and numerical analysis. Fluids, 7(7), 244.
Daneshfaraz, R., Norouzi, R., Abbaszadeh, H., & Azamathulla, H. M. (2022b). Theoretical and experimental analysis of applicability of sill with different widths on the gate discharge coefficients. Water Supply, 22(10), 7767-7781.
Daneshfaraz, R., Norouzi, R., Ebadzadeh, P., Di Francesco, S., & Abraham, J. P. (2023a). Experimental study of geometric shape and size of sill effects on the hydraulic performance of sluice gates. Water, 15(2), 314.
Daneshfaraz, R., Norouzi, R., Ebadzadeh, P., & Kuriqi, A. (2023b). Influence of sill integration in labyrinth sluice gate hydraulic performance. Innovative Infrastructure Solutions, 8(4), 118‏
Franc, J. P., & Michel, J. M. (2006). Fundamentals of cavitation (Vol. 76). Springer science & Business media.
Hassanzadeh, Y., & Abbaszadeh, H. (2023). Investigating discharge coefficient of slide gate-sill combination using expert soft computing models. Journal of Hydraulic Structures, 9(1), 63-80.
Kermani, E. F., Barani, G. A., & Hessaroeyeh, M. G. (2013). Investigation of cavitation damage levels on spillways. World Applied Sciences Journal, 21(1), 73-78.
Norouzi, R., Sihag, P., Daneshfaraz, R., Abraham, J., & Hasannia, V. (2021). Predicting relative energy dissipation for vertical drops equipped with a horizontal screen using soft computing techniques. Water Supply, 21(8), 4493–4513.
Pfister, M. (2011). Chute aerators: Steep deflectors and cavity subpressure. Journal of Hydraulic Engineering, 137(10), 1208–1215.
Pfister, M., & Hager, W. H. (2010). Chute aerators: Air transport characteristics. Journal of Hydraulic Engineering, 136(6), 352–359.
Rahmeyer, W. J. (1981). Cavitation damage to hydraulic structures. American Water Works Association, 73(5), 270-274.
Rasmussen, C. E., & Williams, C. K. (2006). Gaussian processes for machine learning (Vol. 1, p. 159). Cambridge, MA: MIT press.
Raza, A., Wan, W., & Mehmood, K. (2021). Stepped spillway slope effect on air entrainment and inception point location. Water, 13(10), 1428.
Roushangar, K., & Homayounfar, F. (2019). Prediction characteristics of free and submerged hydraulic jumps on horizontal and sloping beds using SVM method. KSCE Journal of Civil Engineering, 23(11), 4696-4709.
Roushangar, K., & Shahnazi, S. (2020). Prediction of sediment transport rates in gravel-bed rivers using Gaussian process regression. Journal of Hydroinformatics, 22(2), 249-262.
Salmasi, F., Abraham, J., & Salmasi, A. (2021). Effect of stepped spillways on increasing dissolved oxygen in water, an experimental study. Journal of Environmental Management, 299, 113600.
Süme, V., Daneshfaraz, R., Kerim, A., Abbaszadeh, H., & Abraham, J. (2024). Investigation of clean energy production in drinking water networks. Water Resources Management, 38(6), 2189-2208.
Vapnik, V. N. (1995). The Nature of Statistical Learning Theory. Springer-Verlag, New York.
Wu, J., Ma, F., & Dai, H. C. (2011). Influence of filling water on air concentration. Journal of Hydrodynamics, 23(5), 601-606.