STAR-CCM+ Version 4.06 New Features October 2009 Introduction - - PowerPoint PPT Presentation

www.cd-adapco.com

New Features STAR-CCM+ Version 4.06

October 2009

Introduction

• Product development is driven by

– Anticipation of simulation needs in targeted industry segments – Customer-centric approach

• Developing on all strategic fronts

Anticipating Market Needs & Customer Responsive

Closed and Resolved Issues in STAR-CCM+ 4.06

• 170+ Development Tasks
• 70+ Customers Enhancement Requests

Geometry & Meshing

• Surface Preparation (Manual repair)

– Repair Feature mode » Automatic identification of glitches, browsing capability » Automatic fixing (either add missing edges or delete bad ones) » Undo/redo since in manual repair mode – Multi-Region Imprint mode for automatic interfacing » Browse of close regions to accept/reject pairs » Imprint one pair at a time or imprint all at once » Each imprint not only connects the two regions, but automatically creates

interfaces for user!

» Undo/redo since in manual repair mode

Why? Who?

• Increase User Productivity
• Shorten the geometry preparation

time on large number of parts

• ALL

Geometry & Meshing

• Prism layer mesher

– Introduction of new parameter: “layer reduction percentage” » Allows for gradual (automatic) reduction in layers in tight cavities

Mesh retracted region in previous versions STAR-CCM+ 4.06 Layer reduction at 100% (no retraction) STAR-CCM+ 4.06 Layer reduction at 50% (50% retraction)

Why? Who?

• Prism layer quality improvement
• Add more user control
• Reduce cells count
• Improve solution accuracy
• Mainly

External aero

Geometry & Meshing

• Generalized Cylinder Mesher

– Create an extruded mesh on portions of regions when possible – Detect “cylindrical” boundaries (user can accept/reject boundaries) – Mesh controls: layer #, extrusion type, stretching ratio

Why? Who?

• Reduce cells count
• Create meshes aligned with flow

direction

• ALL

Where?

• Pipe flow
• Cyclones
• Intake port

flows

Physics

• Broadband Aeroacoustics

– Lilley source term for steady analysis » Reconstruct shear based sources from steady RANS result

• Reconstructed fluctuations available as output to acoustic codes

» Locates maximum sources quickly – Different models for broadband steady-state synthesis

• Linearized Euler
• Proudman
• Goldstein – Specifically designed to model jet noise
• Curle – Predicts noise from boundary layer sources

» Less computational overhead than Lilley but more specific

Why? Where? Who?

• Aeroacoustics is the study of noise generation

through turbulent fluid motion or aerodynamic forces interacting with surfaces

• Increasingly a concern for manufacturers
• STAR-CCM+ now allows identification and analysis
• f possible noise sources
• For far-field analysis, 3rd party codes (SYSnoise,

Actran) can read CCM file

• Reinforce the multi-disciplinary positioning
• Vehicle design

−Wing mirrors −Sunroofs

• Aircraft design

−Landing gear −Flaps

• HVAC
• Occupant

Comfort

• Automotive

OEMs

• Aircraft

manufacturers

• Tier 1 suppliers

Physics

• Real Gas Modeling

– Helps predict the true behavior of gas including molecular effects » Where gas is near condensation or critical point » At extremely high pressures and low temperatures – New model IAPWS-IF97 » Industry standard model for water and steam » Formulated in 1997 by the international association for the properties of

water and steam in response to demand from steam power industry

– User defined equation of state available for both liquids and gasses

• Compressibility for liquids

Why? Where? Who?

• Complete range of density options

including user specified now available

• Complete flexibility with flow regimes

from low speed through to hypersonic covered for both liquids and gasses

• Hypersonics

(including space craft re-entry)

• Gas turbines, high

pressure combustors

• Steam turbines
• Aero and

astrospace

• Power generation
• Gas turbines

manufacturers

• Oil and Gas

Physics

• Porous media and heat exchanger models

– Anisotropic porous energy » Tensor profile input for Thermal conductivity & Specific heat » Heat transfer in fin & tube heat exchangers display anisotropy » High thermal conductivity (k), low specific heat (Cp) along length of fins

the reverse normal to the fins (due to gaps)

– Heat exchanger first iteration » Specify at what iteration heat exchanger source is activated » Adds additional control to aid stability

Why? Where? Who?

• Allows extremely accurate

prediction of heat transfer in fin & tube heat exchangers

• Improves stability especially for high

temperature heat exchangers

• Vehicle thermal

management

• Occupant comfort
• Building heating

and ventilation

• Originally from

automotive sector

• Aerospace
• Electronics cooling

Physics

• Large Eddy Simulation

– Bounded Central Differencing » Addition solver option for LES calculations » LES requires central differencing to ensure accuracy but CD is highly

sensitive to mesh quality resulting in occasional instability

» Bounded CD “blends” lower order schemes to aid stability

• Blending consists of first and second order upwind (default for other solvers)

with CD

• CD is still favored in the blending but where skewed cells are encountered

SOU ensures robust solution while minimizing accuracy compromise

Why? Where? Who?

• Broadens applicability of LES

calculations

• Greatly improved stability on

“industrial” meshes

• LES on complex

geometries

• Vortex shedding
• ff vehicles
• Aeroacoustics
• Requested by nuclear

industry

• Applicable to a wide

range of industries including automotive and aerospace

Physics

• Thermal Nox model

– New model improves on previous version – Now it is not fuel dependent and does not require a pure fuel stream

Why?

Nox prediction is important with strict emission standards New model is more flexible than previous version

Who?

Turbomachinery Oil & Gas

• Coal Combustion

− Model combustion of reacting coal particles and gasification − Includes vaporization, devolatilization, char oxidation

Why?

The STAR-CCM+ model is a more advanced model than STAR-CD 4.10 with greater accuracy and flexibility

Who?

Power-generation

Physics

• Lagrangian Multiphase

– Injector values upgraded to profiles » Increased flexibility in injection specification » Use of field functions and tables for specifying injection values – Cone injection type » Specify inner and outer cone angle with particles injected at random

within specified cone

– Surface injector » Particles injected on part faces as specified by a mass flux

Why? Where? Who?

• Continuing to enhance the

Lagrangian multiphase capabilities and to migrate STAR-CD features

• Add flexibility when specifying

droplet injection

• Vehicle soiling
• Parts subject to

erosion

• Chemical sprays
• Combustion
• Automotive (vehicle &

powertrain)

• Oil and Gas
• CPI & consumer

products

• Biomedical

Physics

• Eulerian Multiphase

– Turbulent Eulerian multiphase flow » K-ε turbulence modeling now applicable for multiphase cases where

turbulence is applied to all phases

» Greatly widens applicability specifically in bubbly flow turbulent mixing

Why? Where? Who?

• The EMP used extensively when

VOF or Lagrangian not suitable

• Drag forces and virtual lift may

be included

• Addition of turbulence increases

applicability and is part of the

• n-going work to shift

multiphase capabilities from STAR-CD

• Granular & immiscible

fluid mixing −Static −Stirred

• Nuclear reactor design
• Separation
• Settling tanks
• Bubble columns
• Chemical

Processing

• Oil & Gas
• Nuclear
• Consumer products
• Manufacturing

Physics

• Harmonic Balance

– Blade flutter capability added » Rapid self-feeding motion, potentially destructive, excited by aerodynamic

forces

» Prescribed flutter deforms mesh to simulate blade motion

• Blade motion modes calculated from structural analysis
• Mesh morpher called at each time level

» Motion may be differently phased across blade rows

Why? Where? Who?

• Blade flutter can cause serious damage in

turbines so mitigation prior to design is important

• Transient phenomena in rotating machinery but
• ften neglected due to timescales involved
• HB More economical than full unsteady

simulation, 2 orders of magnitude saving in computational costs for viscous flows

• More accurate than conventional steady state

methods (e.g. frozen rotor)

• Gas Turbines
• Turbomachines
• Fans
• Blades
• Wind Turbines
• Turbomachinery

−Gas turbines −Aero engines −Steam turbines

• HVAC

−Radial and axial fans

• Powergeneration

−Wind turbines −Pelton turbines

Physics

• Finite Volume Stress

– Large displacements » Account for large displacements when parts are deformed

• New methodology for grid movement

– Implicit FSI » Increased stability and accuracy » Small displacement only

Why? Where? Who?

• Large displacement is necessary when the

deformation of the solid part becomes too important for being modeled with small displacement model

• Implicit treatment for FSI is required for

increasing stability and accuracy of the solution

• Flow, Thermal Stress in one code !
• Riser VIV
• Pipe simulation
• Cooling

applications

• Nuclear reactors
• Oil and Gas
• Nuclear
• Automotive

powertrain

• Turbomachiery

Physics

• MRF and rigid body motion with DFBI

– Ability to have rotating regions with DFBI solver – Primarily used in propulsion calculations – propellers etc

• Wave profiles decoupled from DFBI

– DFBI solver no longer required to automatically generate wave

profiles

• Saturation pressure specified by field function and

table

– Non constant saturation pressure now possible in cavitation model – Note that pressure is now specified absolute

Why? Where? Who?

• Completeness of DFBI and VOF capabilities
• Impose a motion in addition to 6dof
• UAVs
• Manoeuvring
• Subsea devices
• Rudders
• Aerospace &

Defense

• Marine

Physics / CAE Integration

• Multiple passive scalars

– Extending current scalar capability – Ability to track fluid paths without multi-species computational

• verhead

» Manifold simulations, residence/mean age of air calculations

• STAR-CCM+ GT-Power Coupling (co-simulation)

– First instance of STAR-CCM+/1D coupling – Equivalent capability to STAR-CD

• STAR-CCM+ Nastran Mapping

– Import/Map/Export data mesh and field surface results

Why? Where? Who?

• CAE Integration
• Intake port flows
• Automotive

Powertrain

Visualization and Analyses

• Surface Integral and Area Average reports for line parts

– Primarily focused on aerodynamic analysis e.g. Cp on airfoil profiles

• γ (ratio of specific heats) added as a standard field function

– Important parameter to visualize for rocket and propulsive flows made up of

combinations of gases

• Export streamline/ribbons derived parts to re-use in CAD tools
• User-specified uniform grid for plotting vectors

– Improves visualization when wide variations in cell sizes are present

Why? Who?

• Improve general usage
• Reinforcing post-processing

capabilities

• ALL

Post-Processing

• “External Heat Transfer Coefficient” field function for convection

boundaries

– Allows the user to “preview” what HTC is set for boundaries – Useful when using mapping from 3rd party software

• New “Bands” option for reports

– Works on force reports and force coefficient reports – Specifically aimed at aerodynamics (cumulative force graphs) – Delivers values of report locations specified by the number of bands

Why? Who?

• Improve general usage
• Saving processing time
• 30Mcells: from 3hours to 3min
• Reinforcing post-processing

capabilities

• External

aero

Usability Improvements

• Table inputs to support crank angle

– Key for reciprocating engine analysis

• Automatic creation of a java macro during interactive sessions

– “Log File” functionality so actions can be replayed

• Detailed summary report

– Provides full information on hardware and software configuration including

elapsed and CPU time

• Automatic saving of last computed iteration of a diverging case
• Creating a field function from tabular data

– Users can now access table data from a field function – Allows the visualization of tabular data

• JT open import ported on Windows platform

Why? Who?

• Improve usability
• Ease the input settings
• Saving processing time
• ALL

Solvers Performance

• Significant improvement in parallel viewfactor calculations

– Reduced memory requirements and increase scalability

• Speed-up on post-processing operations

– Multi-section plots (close to 3 times faster) – Work still on-going to have a competitive offering – Decrease of memory requirements needed for STAR-View+

• AMG solver improvements

– solver optimizations resulting in up to 40% speed-up on

some coupled flow solutions

– improved robustness on large numbers of processors.

• Coupled solver Relaxation factor

– Stabilization of convergence oscillations during steady state runs

• General performance increase on Windows

– Since first releases, STAR-CCM+ running on windows was ~20% slower than Linux on

the same hardware. Windows port is now equivalent if not faster than Linux

Problem size:

• 22M cells, ~500k Patches
• Beams per patch: 512
• Processes used: 64
• Memory per core/proc: 2GB

Viewfactor computation time:

• ray tracing time: 7 minutes
• reciprocity time: 5 minutes
• total time: 12 minutes

Documentation

• Lagrangian Multiphase documentation restructure

– Added a section that provides a step-by-step guide to

setting up Lagrangian Multiphase simulations

• New Tutorials

– Parts – Turbomachinery » Harmonic Balance » Multi-row

• Validation cases

– Natural convection in an eccentric annulus – Radiative heat transfer and scattering with DOM – Evaporation of Hexane/Decane droplets in dry air – Dispersion of material particles in grid generated turbulence – Surface to surface radiation in a cylindrical hole

Summary

• Product development is driven by

– Anticipation of simulation needs in targeted industry segments – Customer-centric approach

• Developing on all strategic fronts

www.cd-adapco.com

New Features STAR-CCM+ Version 4.06

October 2009