Comparing experimental energy dissipation in multi-horizontal submerged jets to Machine learning models

Pedram Ramezani¹ Ehsan Shamaei² Hamed Kharazmi³

1) Graduated in Master of science-Shahid Chamran University of Ahwaz, Ahwaz, Iran
2) Water and Waste Water Company, Isfahan, Iran
3) Department of Computer Engineering, University of Kashan, Kashan, Isfahan, Iran

IOI Code : XKFA-AZZBDC
Publication : Second International Conference on applied researches in civil engineering, architecture and urban planning (icarcau.com/2nd)
Abstract :
    One of the most difficult issues in designing hydraulic structures is addressing energy dissipation in spillways, outlets and stilling basins. There are various methods to raise energy dissipation; they are devoid of foundation scour, streambed erosion and other unfavorable processes. One of these approaches is employing the stilling basin with multi horizontal submerged jets (MHSJ) by two layers, which are suggested for use as an energy dissipation structure in hydropower stations. This study focused on the prediction of energy dissipation in the stilling basin of multi-horizontal submerged jets with two layers and 3 angles including 0º, 5º and 8º. These structure installed in outlets, performed by machine learning and compared with the energy dissipation ratio, as computed by the energy formula. The results indicated that when the angles of lower layer jets (θ2) were larger than those of the upper ones (θ1) (Ɵ2> Ɵ1), the percent of wasted energy would soar. Also, comparison of the numerical results estimated by the Linear Regression and Support Vector Regression and decision tree methods with the experimental results showed a significant correlation between the numerical results and the experimental ones.

Keywords : energy dissipation multi horizontal submerged jet machine learning