Fans are applications that are increasingly used in many areas. This increase in use leads to a growing. To meet this demand, it becomes necessary to improve fan efficiency, using versatile and stable fans that consume less energy. To meet this challenge, it is important to design machines that operate outside the optimum point. We also need to improve fan stability that is linked to leakage flow.
In this work, we are interested in the flow of backlash in an axial fan. The clearance is the space between the moving parts (the rotating blades) and the fixed parts (the duct or machine casing). Flow through this gap has unavoidable and significant effects. Reduced inflow leads to complex unsteady phenomena in the gap, causing the machine to stall. It therefore seems necessary to investigate and understand the mechanisms behind performance degradations in these fans. The general objectives of this thesis are, to improve our knowledge of the instationnarities that are very frequent in gap flow to develop reliable modeling of loss sources outside the design point.
Therefore, RANS and Hybrid methods are widely used to better predict the physical phenomena that generate the game flow. The RANS method is less expensive in terms of computation time and allows to access more easily to the stationary fields. But it seems limited, because it does not allow access to the unsteady characters of the vortices (TLV, TCRV, TSV). The SGE simulation is then a complement to the RANS simulation, because it helps to detect and understand the vortex system, their dynamics and their interaction with the stationary or moving surfaces, but it remains very costly in terms of computation time, so the Hybrid approach remains a sophisticated alternative to study instationarities in this thesis. The CFD simulation approaches Reynolds Average Navier-Stokes (RANS) and Shear Blended Eddy simulation (SBES) are used in this work to carry out calculations on the cascade configuration and on the fan.
In this thesis, Deveaux’s (2020) model was evaluated and extended to non-optimal operating points on the VT configuration, then verified on the USI7 configuration, in order to partially address the non-optimal design point. Application of the DMD identifies global modes at 2 frequencies, 90 Hz and 180 Hz, at the operating point close to stall. These frequencies are associated with the unsteadiness that cause machines instability.
| Date | 18 Nov 2024 |
|---|
| Original language | French |
|---|
| Awarding Institution | - École de technologie supérieure
|
|---|
| Supervisor | Marlène Sanjosé (Supervisor) |
|---|
Drame, A. B. (Author),
Sanjosé (Supervisor),
18 Nov 2024Student thesis: Doctoral thesis › Doctorate in Engineering: Engineering