Hypersonic Aerothermodynamics
Introduction:The main challenge of the research project “Aero-elastic couplings and dynamic behavior of rotational periodic bodies” is to develop validated mathematical model of the flutter phenomena in the framework of fluid-structure systems characterized by the rotational periodicity of a flexible structure and to investigate the approaches for flutter suppression. Such problems bring the issue of possible phase shifts between the particular parts of a vibrating structure (e.g. blades in a cascade) which can lead to unstable oscillation. I am developing reduced order aerodynamics tool estimate the aerodynamics loads on the blade cascade, these load will be fed used in structural model as external excitation or aerodynamics damping. In the first phase the separate CFD and FEM numerical analysis will be perform to identify the dynamics parameters of the blade cascade. Once validated with experiment the flow solver will be loosely coupled with structural solver to perform complete Aeroelastic coupled simulation.
FEM Model:The structural model is represented by the non-linear beam model. The blade and the tower is criticized into several beam elements of each elements have 6 degree of freedom. The blade and entire turbine Beam element model is presented below.
Aero-servo-elastic co-simulation model: For the aero-elastic coupling partitioned based coupling stagey is considered in which both flow solver and structural solver MECANO is coupled loosely. Both solver are coupled in Matlab/Simulink environment using s-function tool. The load transfer function between both both the model is achieved by special elements called 'DIGI' elements which acts as 1-D sensors. The model is first tested on single bladed wind turbine model and then three bladed model. The complete co-simulation model is presented in the figure below.
Fig: wake shape
Fig: FEM beam model
Fig: Aero-servo-elastic co-simulation model
Results: The Blade tip deflection vs time and the blade 1st bending mode is presented in the figure right.
Also two different wind velocity is used one with constant wind speed and other one step wind speed.
Fig: Example of steam turbine
Aero-elastic simulation of Blade cascad: In the above figure the aero-elastic of blade cascade is presented. There simulation is performed in the strongly coupled environment. However, the in the future the loosely coupled strategy will be used with ROM for both flow and structural solver.
FEM model of Bladed wheel: To identify the dynamic behavior of the bladed wheels. The simplified lab model is created using 30 prismatic blades. The main aim of the experiment is to identify the vibration characteristics of this kind of model under external excitation. The FEM result of modal analysis for the wheel is present in the right figure which shows the first mode shape and bending frequency has good agreement with experimental results.
Fig: 1st Mode shape of Bladed wheel
Aerodynamics and aeroelasticity of HAWT
FSI for gas turbine and Steam turbine
Introduction: This research work was carried out in joint collaboration with University of Liege and Von Karmmen institute, Belgium. One of the prime objective of the work is to accuratly estimate the aerothermal loading on the re-entry bodies.
During this work I have successfully implemented new type of numerical scheme based on approximate Riamenn solver to simulate the hypersonic flows e.g. HLLC, AUSM, AUSM+, AUSMDV. These scheme are implemented in the in-house CFD code. The code is density based highly parallel solver.
One of the key issue was to solve the Carbuncle problem in the numerical scheme. In the figure (left) the Carbuncle problem in Roe's scheme is removed by use of AUSM+ scheme. OpenMPI, PETSc and ParMETIS is used for the parallelization of the code. The code has been tested on >1000 core.
Research activities
Fig: Blade 1st Bending mode fz=0.85hz
Fig: Blade deflection at 2 different wind vel
Introduction: Horizontal axis wind turbines (HAWT) are one of the most popular machines for producing renewable energy in the world. Over the last two decades, a significant shift in government policy towards renewable energy has led to bigger and more efficient wind turbines, a development that has stretched the capabilities of traditional wind turbine design methods. High-fidelity and integrated multi-disciplinary models of wind turbine systems are important for the correct evaluation of the performance and the load analysis, and thus might reduce the failure rate during the design stage.
Therefore the main aim of the research work to develop more advanced design tools, that can model the complete wind turbine, including nonlinear structural and control effects as well as unsteady aerodynamics flows around the rotor. Higher fidelity aroservoelastic modeling of complete wind turbine model is the main focus of the present work. The project work is funded by Belgium Govt under DynaWind project with joint collaboration of Siemens and University of Liege, Belgium.