Announcements Headlines:
    Assessment of the Performance of the ASBM Turbulence Closure in Biomedical Fluid Flows

    Sponsor

                  

    The project is sponsored by the Research Promotion Foundation (RPF) of Cyprus. The project is part of the "Nikos Simeonides" 2010 prize for Excellence in Research, awarded to Prof. S. C. Kassinos (the coordinator of the project).  

    http://www.research.org.cy/EN/index.html/

    Cyprus. 

     

    ASBM

    The ASBM is an advanced Reynolds Averaged Navier-Stokes (RANS) model of turbulence that holds significant promise for improved turbulence predictions in complex engineering flows. Having originally been formulated at Stanford University, the ASBM has, over the last decade, been further developed at the Computational Sciences Laboratory (UCY-CompSci) of the University of Cyprus. The model allows improved flow predictions for many important benchmark cases, involving two-dimensional mean flow configurations, but has so far not been extensively tested in flows with three-dimensional effects, such as separation and re-attachment. However, a number of important applications, spanning diverse engineering fields, such as body-wing junctions in aerodynamics and airflow and particle deposition in the human respiratory system, involve strong three-dimensional effects. Therefore, an important next step in the continued development of ASBM is the validation, and if necessary further development of the model, for a number of benchmark cases involving internal flow with significant three-dimensional effects. To achieve this goal we compute a number of benchmark cases of increasing complexity leading finally to the computation of the flow in a simplified model of the human respiratory system.

     

    Project Members  

    Prof. Stavros Kassinos (University of Cyprus, Department of Mechanical and Manufacturing Engineering)

    Dr. Dimokratis Grigoriades (University of Cyprus, Department of Mechanical and Manufacturing Engineering)

    Dr. Xavier Albets-Chico (University of Cyprus, Department of Mechanical and Manufacturing Engineering)

    Mr. Fotos Stylianou (University of Cyprus, Department of Mechanical and Manufacturing Engineering)

    Mr. Konstantinos Panagiotou (University of Cyprus, Department of Mechanical and Manufacturing Engineering)

    Collaborative Members 

    An ongoing collaboration exists with the following members:

       Prof. Rene Pecnik (Delft University of Technology, Process and Energy)

       Dr. John O'Sullivan (The University of Auckland, The Department of Engineering Science)

       Prof. Karthik Duraisamy (Stanford University, Department of Aeronautics and Astronautics)

    for more information see the joint website:

    http://www.structurebasedmodel.org/

    Project Purpose

    The project aims at the further development and validation of the Algebraic Structure-Based Model (ASBM) for three-dimensional internal turbulent flows, with the long term objective to use the model in simulations of air flow and aerosol deposition in the human respiratory system. 

     

    High-Level Objectives 

    1) To train the Young Researcher on turbulence modeling, and in particular on the ASBM closure,

    2) To refine the ASBM closure for internal flows with strong three-dimensional effects,  

    3) To promote the adaptation of the ASBM in a number of engineering fields, such as in biomedical engineering.

     

    Scientific and Technological Objectives

    1) To optimize the existing implementation of ASBM in a highly scalable, MPI-based, parallel Navier-Stokes solver (CDP2.4), on structured/unstructured grids, for efficiency and stability.

    2) To carry out a systematic validation of the ASBM in a progression of internal turbulent flows of increasing complexity and with significant three-dimensional effects that have traditionally challenged RANS closures. These cases are:

        a. Fully-developed turbulent flow in a pipe.

        b. Fully-developed turbulent flow in a square-duct.

        c. Fully developed flow in a bifurcating pipe (simplified model of trachea and bronchi).

    3) To refine the model by addressing potential shortfalls identified during the testing phase.

    4) To collaborate with international individual researchers for the thorough validation of the model and for exchange of knowhow on the numerical implementation of the ASBM, thus further promoting the international dimension of the ASBM.

     

    Events 

    UCY-CompSci is organizing a workshop in Cyprus during which the main scientific and technological outcomes of the project will be presented. The workshop is arranged on 23 of May 2014 at 10:00. It will take place at the conference room at the Nireas building 96 Kyrenias street, Aglantzia. 

     

    Turbulent Pipe Flow

    DNS of a fully developed turbulent pipe flow, at Re_bulk= 5300 and Re_tau=180. Contours of instantaneous streamwise velocity U_x. Non- dimensionalization is carried using the bulk velocity (u_bulk=1), and radius of the pipe (R=1). The upper part of the video is an x-y cross-section (at z/R=0), while the lower part of the video is a z-y cross-section (at x/R=7.5).

    Project Objectives

    The project aims at the further development and validation of the Algebraic Structure-Based Model (ASBM) for three-dimensional internal turbulent flows, with the long term objective to use the model in simulations of air flow and aerosol deposition in the human respiratory system. 

     

    ASBM

    The ASBM is an advanced Reynolds Averaged Navier-Stokes (RANS) model of turbulence that holds significant promise for improved turbulence predictions in complex engineering flows. Having originally been formulated at Stanford University, the ASBM has, over the last decade, been further developed at the Computational Sciences Laboratory (UCY-CompSci) of the University of Cyprus. The model allows improved flow predictions for many important benchmark cases, involving two-dimensional mean flow configurations, but has so far not been extensively tested in flows with three-dimensional effects, such as separation and re-attachment. However, a number of important applications, spanning diverse engineering fields, such as body-wing junctions in aerodynamics and airflow and particle deposition in the human respiratory system, involve strong three-dimensional effects. Therefore, an important next step in the continued development of ASBM is the validation, and if necessary further development of the model, for a number of benchmark cases involving internal flow with significant three-dimensional effects. To achieve this goal we compute a number of benchmark cases of increasing complexity leading finally to the computation of the flow in a simplified model of the human respiratory system.

     
     

    ASBM

    The ASBM is an advanced Reynolds Averaged Navier-Stokes (RANS) model of turbulence that holds significant promise for improved turbulence predictions in complex engineering flows. Having originally been formulated at Stanford University, the ASBM has, over the last decade, been further developed at the Computational Sciences Laboratory (UCY-CompSci) of the University of Cyprus. The model allows improved flow predictions for many important benchmark cases, involving two-dimensional mean flow configurations, but has so far not been extensively tested in flows with three-dimensional effects, such as separation and re-attachment. However, a number of important applications, spanning diverse engineering fields, such as body-wing junctions in aerodynamics and airflow and particle deposition in the human respiratory system, involve strong three-dimensional effects. Therefore, an important next step in the continued development of ASBM is the validation, and if necessary further development of the model, for a number of benchmark cases involving internal flow with significant three-dimensional effects. To achieve this goal we compute a number of benchmark cases of increasing complexity leading finally to the computation of the flow in a simplified model of the human respiratory system.

     
     

    High-Level Objectives 

    1) To train the Young Researcher on turbulence modeling, and in particular on the ASBM closure,

    2) To refine the ASBM closure for internal flows with strong three-dimensional effects,  

    3) To promote the adaptation of the ASBM in a number of engineering fields, such as in biomedical engineering.

     

    High-Level Objectives 

    1) To train the Young Researcher on turbulence modeling, and in particular on the ASBM closure,

    2) To refine the ASBM closure for internal flows with strong three-dimensional effects,  

    3) To promote the adaptation of the ASBM in a number of engineering fields, such as in biomedical engineering.

     

    Scientific and Technological Objectives

    1) To optimize the existing implementation of ASBM in a highly scalable, MPI-based, parallel Navier-Stokes solver (CDP2.4), on structured/unstructured grids, for efficiency and stability.

    2) To carry out a systematic validation of the ASBM in a progression of internal turbulent flows of increasing complexity and with significant three-dimensional effects that have traditionally challenged RANS closures. These cases are:

        a. Fully-developed turbulent flow in a pipe.

        b. Fully-developed turbulent flow in a square-duct.

        c. Fully developed flow in a bifurcating pipe (simplified model of trachea and bronchi).

    3) To refine the model by addressing potential shortfalls identified during the testing phase.

    4) To collaborate with international individual researchers for the thorough validation of the model and for exchange of knowhow on the numerical implementation of the ASBM, thus further promoting the international dimension of the ASBM.

     

    Events 

    UCY-CompSci will organize a seminar in Cyprus during which the main scientific and technological outcomes of the project will be presen

    Events 

    UCY-Co

    Events 

    UCY-CompSci will organize a seminar in Cyprus during which the main scientific and technological outcomes of the project will be presented. The targeted audience will include experts in Computational Fluid Mechanics from both research institutions and SMEs. The Seminar will be organized during the last two months of the project, but the exact date will be specified at a later time in order to maximize the impact of the event (and to avoid potential conflicts with other events).

    Mean streamwise velocity profile

    Normalized structure componentality tensor

    Normalized structure dimensionality tensor

    Normalized structure circulicity tensor

    Turbulent Pipe Flow

    DNS of a turbulent bifurcation pipe. For the construction of the geometry we have used the physiologically realistic bifurcation model of T. Heistracher and W. Hofmann (J. Aerosol Sci., Vol. 26, No. 3, pp. 497-509, 1995). The mathematical description of the carina region has been modified to achieve a smoother join of the two outlet branches. To achieve physical turbulent inlet profile we use the recycling boundary condition. The video shows contours of the instantaneous velocity magnitude U_mag.

    Simulation Parameters:

    1) daughter branch radius: R_d=1.0

    2) parent branch radius: R_p=1.244444

    3) bifurcation angel: phi=70 degrees

    4) viscosity: nu=3.77358x10^-4

    5) time step: dt=0.004

    6) daughter bulk flow velocity: u_b^d=1.0

    7) parent bulk flow velocity: u_b^p=1.29145

    8) daughter Reynolds number Re_b^d= 5300

    9) daughter Reynolds number Re_b^p= 8518

    Note that the Reynolds numbers are based on the local pipe diameter and local bulk velocity.

     

    DNS of a turbulent Y-junction pipe. For the construction of the geometry we have used the physiologically realistic bifurcation model of T. Heistracher and W. Hofmann (J. Aerosol Sci., Vol. 26, No. 3, pp. 497-509, 1995). The mathematical description of the carina region has been modified to achieve a smoother join of the two inlet branches. To achieve physical turbulent inlet profiles we use the recycling boundary condition (the arrow in the upper branch in pointing in the wrong direction). The video shows contours of the instantaneous velocity magnitude U_mag.

    Simulation Parameters:

    1) daughter branch radius: R_d=1.0

    2) parent branch radius: R_p=1.244444

    3) bifurcation angel: phi=70 degrees

    4) viscosity: nu=3.77358x10^-4

    5) time step: dt=0.004

    6) daughter bulk flow velocity: u_b^d=1.0

    7) parent bulk flow velocity: u_b^p=1.29145

    8) daughter Reynolds number Re_b^d= 5300

    9) daughter Reynolds number Re_b^p= 8518

    Note that the Reynolds numbers are based on the local pipe diameter and local bulk velocity.

    Related Projects

    Biomedical flows was one of the focus areas of the TOK-DEV project and an area where UCY-CompSci has developed significant know-how. We have developed the capability to carry out high-fidelity simulations of aerosol deposition in models of the human respiratory system. We have shown that turbulence generated in the upper airways persists at least down to the first few generations of the bronchi and at least for a part of the inhalation cycle. Hence, Large Eddy Simulations (LES) are ideally suited for probing the fundamental aspects of these flows and for gaining a better understanding of the factors (such anatomy, chronic disease etc.) that affect aerosol deposition. On the other hand, for routine patient evaluation of customized treatment strategies LES is still too expensive. We are thus also engaged at developing the capability to carry RANS based simulations using the Algebraic Structure-Based Model (ASBM).

    UCY-CompSci is also active in hemodynamics, for example we are using LES to study the flow factors that contribute to the rapture of aneurysms in the carotid. Aneurysm rapture remains notoriously difficult to predict, an indication that we still do not understand all the mechanism that determine the evolution of aneurysms to the point of rapture. Using real carotid geometries and velocity data we have shown that even for nominal Reynolds numbers that would correspond to laminar flow, in the vicinity of the aneurysm the flow becomes unstable and turbulent-like at least for part of the heartbeat cycle. This inherent flow unsteadiness is thought to contribute to a modified distribution of shear stresses on the walls of the aneurysm and could prove to play role in raptures.

    Related Publications

    H. Radhakrishnan and S. C. Kassinos, Modeling Pulsatile Flow In An Aneurystic Carotid ArteryProceedings of the 8th Euromech Fluid Mechanics Conference, Bad Reichenshall, Germany (2010)

     
    H. Radhakrishnan and S. C. Kassinos, CFD Analysis of Turbulent Flow And Particle Deposition In A Symmetrically Bifurcating Human Airway Model6th World Congress on Biomechanics, Singapore (2010)
     
    H. Radhakrishnan, and S. C. Kassinos, Numerical modeling of turbulent airflow and particle deposition in a bifurcating airway modelProceedings of the 11th World Congress on Medical Physics and Biomedical Engineering, Munich, Germany (2009)
     
    H. Radhakrishnan, and S. C. Kassinos, CFD Modeling of Turbulent Flow and Particle Deposition in Human LungsProceedings of the 31st Annual International IEEE EMBS Conference, Minneapolis, Minnesota, USA (2009)
     
    H. Radhakrishnan, and S. C. Kassinos, Using LES To Model Turbulent Particle Transport In Human LungsProceedings of the 6th International Symposium on Turbulence and Shear Flow Phenomena (TSFP6), Seoul, Korea (2009)
     
    H. Radhakrishnan and S. C. Kassinos, Using LES to model turbulent particle transport in the airways of the human respiratory system7th International ERCOFTAC Symposium on Engineering Turbulence Modelling and Measurements, Limassol, Cyprus (2008)
     
    H. Radhakrishnan and S. C. Kassinos, Modeling particle deposiotion in human lungs6th GRACM International Congress on Computational Mechanics, Thessaloniki Greece (2008)