+ All Categories
Home > Documents > Evaluating collections of Simulation Results with Tecplot ... · PDF fileEvaluating...

Evaluating collections of Simulation Results with Tecplot ... · PDF fileEvaluating...

Date post: 28-Mar-2018
Category:
Upload: phamkhuong
View: 228 times
Download: 0 times
Share this document with a friend
14
National Aeronautics and Space Administration www.nasa.gov Evaluating collections of Simulation Results with Tecplot Chorus and Tecplot 360 Dr. Scott Imlay CTO Dr. Durrell Rittenberg VP Product Management 1
Transcript

National Aeronautics and Space Administration

www.nasa.gov

Evaluating collections of Simulation Results with Tecplot Chorus and Tecplot 360!

Dr. Scott Imlay CTO!Dr. Durrell Rittenberg VP Product Management!

1

Agenda For Tutorial!•  Creating a project from the raw results!-  Using File Crawling to pull results!-  Creating an index file!

•  Extracting standard results from the collection!-  Creating Templates (In Tecplot 360)!-  Leveraging Macros (In Tecplot 360)!-  Custom Actions (In Tecplot Chorus)!

• System modeling with surrogate models!-  Multidimensional models!-  Exporting model results!

•  Report Generation!-  Image and data exports!-  Session files!

Driving Force is Moore’s Law!

Impact of Low Cost Computational Resources!

NAS: NASA Ames (CA)

Growing with Moore’s Law

CFD Solutions Are Getting Complex!

5

1991 2011

Com

plex

ity

Complexity growing with Moore’s Law

System Engineering “Vee”!

Feasibility Study

Concept of Operations

System Requirements

High-Level Design

Detailed Design

Software / Hardware Development

Field Installation

Unit/Device Testing

Subsystem Verification

System Verification & Deployment

System Validation

Operation & Maintenance

Changes Upgrade

Regional Architectures

Lifecycle Process

Timeline Implementation

Development Process

System Validation Plan

System Verification Plan (System Acceptance)

Subsystem Verification Plan

(subsystem Acceptance)

Unit/Device Test Plan

Document / Approval

Multi- Disciplinary Trade Studies and

Optimization

Traditional CAE Methods

Typical Expenditure Profile!

Field-Data Operations for Parametric Data!

• Common Field-Data Operations!- Comparative analysis (what has changed)!-  Integrations (forces and moments)!- Extractions (snapshots)!- Aggregations (statistics, box-plots)!

• Operations over enormous amounts of data!- Example: Aero database development !

•  Thousands of 100M Cell CFD solutions!- Some aggregations require data from all solutions to be

loaded simultaneously !•  If not clever, work through equivalent of 100B Cells!

Tecplot Chorus For Collections of CFD Results!

Evaluating overall system performance and allowing engineers to compare results quickly

Meta-data Analytics and Surrogate-Models!

•  Surrogate Models!-  A simple mathematical model that approximates the

results of the simulation as a function of the input parameters!

-  Generally statistical in nature!•  Internal Application of Surrogate Models!-  Visualization: surfaces and line plots!

•  External/Future Applications of Surrogate Models!-  Substitute for CFD code when output meta-data is

needed as a function of input parameters!-  Sensitivity Analysis!

•  Quantify leverage of each input variable on results!•  ANOVA (Analysis of Variance)!

-  Probabilistic Analysis!•  In reality all input variables have some randomness

(stochastic)!•  Example: weather forecasting ensemble models!

-  Robust Design and Reliability-Based Design!•  Improve quality by minimizing the impact of input

randomness!

Current Surrogate Models!

•  Response Surface!-  Linear least-square fit of an N-

dimensional quadratic equation!-  Note: in spite of name, THIS IS NOT

A SURFACE!•  Kriging!-  Minimizes error over field, not just

points!•  defined correlation function

(variogram)!-  Interpolation (passes through the data

points)!-  Expensive for large sets of meta-data!

Tutorial 1 Hot Water Mixing!

12

Tutorial Overview!

•  Low Speed T-junction mixing problem (CFD tutorial) !-  Input variables: !

•  Main pipe velocity!•  Cooling water temperature!•  Secondary pipe velocity!•  Hot water temperature !

- Output variables: !•  Outflow temperature profile!•  Average temperature at outlet!•  Mass-flow !

13

Work on Large Data Performance!


Recommended