Title

1
• DECEMBER 2019

Minimizing the Preprocessing Time Sink:

A Guided, Fluent Meshing Workflow for
the Aerospace and Defense Industry

Varun Chitta, PhD

Application Engineer, Fluids

Advanced Modeling and Simulation (AMS) Seminar Series

NASA Ames Research Center, August 17, 2021
Your Presenter Today
Varun Chitta
Application Engineer, Fluids at Ansys
About Me:
• Work closely with Aerospace and Automotive industry customers for
deployment and adoption of fluids and multiphysics solutions
• Background: Ph.D. in aerodynamics and turbulence modeling

Let’s stay connected!

[email protected]

3
Agenda
• Fluids meshing tools: Overview
• Fluent meshing workflows
‐ WTM
‐ FTM

• Mosaic technology
• Fluent meshing workflows: Case studies
‐ Live demonstration: WTM workflow

• Automated workflows

5
Best-Of-Breed Simulation Across All Major Physics

6
 Ansys Driving Simulation Themes For Aerospace & Defense Industry
Improve engineering productivity while improving trust in simulation technology
Workflow

Preprocessing
Solver

• Dedicated external aero

workflow: parametric,
scriptable
• End-to-end workflows
• Mosaic technology
• Parallel meshing • Improved robustness: HSN, PMN
• Clean/dirty CAD • Built-in NASA 9-coeff curve fits
• Integrated in workflow for material properties
• Built-in partial slip-wall BC
• Built-in surface ablation model

7
 Ansys Driving Simulation Themes For Aerospace & Defense Industry
Improve engineering productivity while improving trust in simulation technology
Workflow

Preprocessing
Solver

• Dedicated external aero

workflow: parametric,
scriptable
• End-to-end workflows
• Mosaic technology
• Parallel meshing • Improved robustness: HSN, PMN
• Clean/dirty CAD • Built-in NASA 9-coeff curve fits
• Integrated in workflow for material properties
• Built-in partial slip-wall BC
• Built-in surface ablation model

8
Fluids Meshing Tools
Fluids Meshing Tools: Overview

Block-Structured
Unstructured
Hex Multiphysics
Meshing Turbomachinery
ICEM CFD & Users
Fluent Meshing & TurboGrid
SpaceClaim Ansys Meshing
Workflows
Meshing

10
Fluent Meshing Robust mesh generation for complex assembly

• Handles large geometry

• Generates huge meshes
• Parallel meshing
• Advanced methods
• Poly-hexcore / Poly / Tet-hexcore / Tet
• CAD import to solution in one interface
Unstructured
• Linux and Windows support
Meshing • Automation
Fluent Meshing ‐ Workbench
‐ Scripting

11
Fluent Meshing Workflows Robust mesh generation for complex assembly

Step by step guided workflows

✓ Fast learning
✓ Automated
✓ Persistent
✓ Efficient parallel meshing
✓ Customizable workflow
Unstructured ✓ Can be automated with
Meshing python-based journal file
Fluent Meshing
Workflows

12
Fluent Meshing Workflows

Watertight Meshing Workflow (WTM) Fault-Tolerant Meshing Workflow (FTM)

• Prerequisite: clean watertight geometry • Tolerant to bad quality CAD/STL
• Share topology at CAD/mesh level • For dirty, non-watertight geometries
• Conformal CHT mesh • Wrapping technology
• Automatic handling of gaps and overlaps
• Significantly reduces the pre-processing
and CAD-cleanup effort required to
prepare the models for simulation
• Non-conformal CHT mesh

13
When to Use the Watertight Geometry (WTM) Workflow?
• What do we mean by watertight
geometry?
‐ Relatively clean
‐ Watertight solid/fluid regions
• Single body: any imported CAD model
• Multi bodies: share topology at CAD level Aircraft cabin

• Supports surface mesh file as input

• Disconnected bodies are allowed if one
body is entirely inside another, e.g.,
aircraft inside farfield domain
• Can be meshed by surface meshing and
then volume filling, i.e., no wrapping
required Generic combustor

14
What is Wrapping?​
• The wrapping approach can be likened to
‐ Shrink-wrapping luggage to wrap from outside of the geometry, or
‐ Blow molding to wrap from inside of the geometry
• Wrapping has several strengths:​
‐ Automated handling of gaps and overlaps​
• Removes need for manual geometry-based closure of the geometry​
• No need to have geometry in form of a watertight region
‐ Feature suppression​
• Mesh sizes > local features can be used for local defeaturing
• Inside Fluent Meshing FTM workflow
‐ Part management for assembly preparation
‐ Wrap-extract of external and internal flow domain
‐ Creation of porous blocks, e.g., heat exchangers
‐ Automatic offset and wake refinements
‐ CHT applications via non-conformal auto-pairing workflow

15
When to Use the Fault-Tolerant Meshing (FTM) Workflow?

• Complex faulty CAD where it would be impractical

to clean into a good quality watertight geometry
• Faceted geometry
‐ Scanned geometry
‐ STL geometry
• Watertight CAD with complex surface patches
where surface meshing results in poor quality
mesh​
• Cases where mesh-based defeaturing is required​
• Applicable for all industries

External aerodynamics
application examples

16
Mosaic Technology
Accuracy and Solution Time Are Highly Dependent on the Mesh

• Various geometries and flow regimes require

different meshes
• Hybrid meshes containing hex in the core are
good alternatives
HEXCORE TET
• But there’s no good solution for mesh
transitions
− Reduced mesh quality
− Excessive cell counts
− Forced to compromise on a common element
type to minimize mesh transitions
PRISMS NATIVE POLY

18
Mosaic Meshing Technology
• Mosaic technology enables polyhedral connections between disparate mesh types
‐ Conformally connects polygonal prisms on the boundary to bulk Cartesian aligned hexahedral cells
• Hex dominant allows for faster meshing and a better-quality volume mesh
‐ Solve time reduction of 30-50%

Isotropic Poly Prisms Hexahedral Cells Mosaic Polyhedral Cells

20
 Parallel Poly-Hexcore Metrics
• Parallel poly-hexcore quality, robustness,
scalability improvements
‐ 31.5M elements Poly-Hexcore mesh includes
‐ 5.2M Cartesian aligned Hexahedral elements
and
‐ 15.4M polyhedral prismatic elements

10X
speed up

21
Fluent Meshing Workflows:
Case Studies
Watertight Meshing of NASA X-43
• Hypersonic flow past a X-43 (mockup) geometry at Mach 9
• No yaw angles are considered -> used symmetry assumption

Geometry cleanup
SpaceClaim geometry performed in Domain with boundary conditions
Ansys SCDM

23
 Watertight Meshing of NASA X-43

Mesh stats:
Total mesh count = 24.6M
3.4M hexahedral cells
21.2M polyhedral cells
Total mesh time ≈ 4.08 mins
with parallel poly-hexcore
elements
Max cells/min seen = 6.04M (64
processors)
Max skewness = 0.88

24
 Fault-Tolerant Meshing of NASA High-Speed CRM
• Transonic flow past a NASA common research model (CRM) representative of a wide-body
commercial transport aircraft

Unstitched surfaces
Intersecting bodies
Overlapping bodies
Original geometry
Internal bodies
Gaps/Holes

25
Fault-Tolerant Meshing of NASA High-Speed CRM

Mesh stats:
No CAD preparation
Total mesh count = 51,210,409
42.6M hexahedral cells
8.6M polyhedral cells
Total mesh time ≈ 57.2 mins
with parallel poly-hexcore
elements
Number of cores = 32
Max skewness = 0.9
Original geometry

Mesh

26
Additional Case Studies: Pre-processing

27
Live Demonstration: WTM
Workflow
 Problem Definition
L = 3.6m
• Hypersonic flow past a NASA X-43 (mockup) geometry at Mach 9

Pressure outlet

BOI’s
Wall, no-slip

Ma∞ = 9

Pressure-far-field
Symmetry plane

SpaceClaim Geometry Domain with Boundary Conditions

29
Watertight Meshing of NASA X-43: Live Demo

30
Watertight Mesh to Simulation Workflow for NASA X-43
• Clean geometry in SCDM
• Setup named selections for boundaries
Geometry

• Transfer clean CAD into Watertight workflow

• Setup mesh and boundary layers controls
• Generate surface and volume mesh
Meshing
• Save workflow for future mesh edits / mesh automation

• Load mesh into Fluent and solve using Density-based solver

• Solution transcript written by default
Setup/Solve • Save the journal file for future simulations (automation)

• Post-process in Fluent
Post- • Can be exported into Ansys Ensight / CFD-Post
Process

31
 End-to-End Automated
Workflow for Pre-Processing
Automation Framework
• Ansys simulation software can be readily automated and incorporated as part of a pre-
existing solution with python scripting

• Use wrapper scripts and journal files to automate workflow execution

‐ CAD manipulation, meshing, simulation execution and more…
‐ Highly customizable

• Allows for batch processing and can further extend integration with 3rd party/in-house
tools Wrapper Scripts
(Perl, Python, Shell, etc.)

Post-Processed
User Inputs
Results

33
 Case Study: Automated Workflow Flowchart
• A fully automated workflow was developed. Using Python scripting to drive the CAD-
preparation, meshing, setup, execution and post-processing of the simulation
Named Selection creation in SpaceClaim Mesh is generated automatically
using the Fault tolerant workflow
FMD
File
Solver execution
Post-processing
Journal
File

Python script produces

meshing journal file Python Input File

Text-based input file

34
Case Study: Automated Meshing Operation

35
 Top Reasons to Use Ansys Fluent Meshing for A&D Applications

Parallel
User friendly
scalability

Automated Robust mesh

meshing offerings

Customizable
Ecosystem
workflows

36
Learning Resources
• Ansys Learning Hub (ALH)

• Fluent Webinar Series

• Ansys Customer Excellence (ACE)

Support

37
Thank You!