Julio Isaac Montenegro Gambini, Julio Martín Kuroiwa Zevallos, Luis Fernando Castro Inga
Wednesday 1 july 2015
9:45 - 10:00h at Africa (level 0)
Themes: (T) Water engineering, (ST) Experimental facilities and instrumentation
Parallel session: 8F. Engineering - Instrumentation
The design of a hydraulic engineering project is a challenge because involves many aspects, generally an iterative process. As part of the design of these structures an indispensable tool is its modeling, usually to check the hydraulic performance of its configuration and if necessary, improve the design. The barrages are the most used hydraulic structures in intakes and dams ensuring a more regular feeding and retain a constant level, being tested on scale models to study the flow behavior and interactions. Three-dimensional numerical modeling of these structures don’t have much background due to the recent development of computational software and high-capacity processors. This paper presents the numerical and physical modeling of Mazán Hydroelectric Project barrage in Iquitos with 24 radial gates which 3 are reproduced in 1/30 scale physical model to study their operation in normal conditions and extraordinary floods obtaining a discharge curve. FLOW3D is used in this research employing the finite volume method solving the Navier-Stokes equations numericaly and generating a nested mesh, allowing considerable savings on the number of items and less computational effort. For validation, we made some comparison observing flow patterns, free surface levels and velocity fields in cross sections. Comparisons satisfactory results. The free surface in the mathematical model turns out be in acceptable accuracy, including the display of the standing waves. The validation of the mathematical numerical model will provide recommendations for future modifications to the structure and its optimization without opting for a new physical model saving material, human and time resources. Today although there are trends suggest that numerical models can replace entirely to the physical modeling, but it’s best to be complementary since efficient participation of both models is a useful way to improve designs.