Chapter2

ME 437 Computational Fluid Dynamics

CFD Solution Procedure

  • Presented by Dr. Hediye ATIK

Chapter 1 Summary

Definition of CFD

  • CFD stands for Computational Fluid Dynamics, which is the simulation of fluid flow and related phenomena using computers.

  • Combines multiple fields:

    • Engineering (Fluid Mechanics)

    • Mathematics & Related Disciplines

    • Numerical Analysis

    • Computer Science

Applications of CFD

  • Research Tool: Used in various studies for theoretical advancements.

  • Educational Tool: Helps in teaching concepts of fluid dynamics effectively.

  • Design Tool: Applicable in various engineering domains:

    • Aerospace

    • Automotive Engineering

    • Biomedical Science and Engineering

    • Chemical and Mineral Processing

    • Civil and Environmental Engineering

    • Metallurgy

    • Nuclear Safety

    • Power Generation and Renewable Energy

    • Sports

Future of CFD

  • Expected to be extensively utilized in numerous industrial applications.

  • Proper implementation and application of CFD methodologies is crucial for success.

Hardware and Software Requirements

  • Hardware Needs:

    • Workstations

    • Clusters

    • Supercomputers

  • Software Needs:

    • Commercial codes

    • Research codes

    • Third-party mesh generators and post-processors

Steps of a Typical CFD Study

  1. Solver: Governing equations solved on a mesh.

  2. Preprocessor:

    • Creation of geometry, transport equations, physical models, mesh generation, material properties, boundary conditions.

  3. Postprocessor:

    • Includes settings, initialization, monitoring solution, and reporting.

Fluent Software Overview

  • Fluent@Guan: Involves settings for 3D problems, physics selection, user-defined settings, and post-processing steps.

  • A graphical user interface guides the user through various steps involved in the CFD analysis.

Detailed Steps in a Typical CFD Study

Step 1: Problem Setup

  • Step 1a: Creation of Geometry and Define the Problem Domain

    • Classify flow as internal (confined by surfaces) or external (not confined).

    • Identify problem domain and impose artificial boundaries, ensuring sufficient boundary condition information is available.

Step 1b: Mesh Generation

  • Divide domain into smaller subdomains (mesh).

  • Mesh quality impacts the accuracy of results; a balance between accuracy and computational cost is essential.

  • Types of Cells:

    • Structured Cells: Regular shapes (quadrilaterals in 2D, hexahedra in 3D).

    • Unstructured Cells: Irregular shapes, suitable for complex geometries.

Step 1c: Selection of Physics and Fluid Properties

  • Important to understand and select appropriate physics models in CFD software.

  • Choices include:

    • Steady vs. Unsteady

    • Incompressible vs. Compressible

    • Laminar vs. Turbulent

    • Viscous vs. Inviscid

Step 1d: Specification of Boundary Conditions

  • Defined conditions at the boundaries that govern the flow.

  • Examples include inlet and outlet conditions, wall conditions.

Step 2: Solution

Step 2a: Initialization and Solution Control

  • Converts nonlinear governing equations to linear algebraic equations for computational efficiency.

  • Finite Volume Method (FVM) is commonly used for solving CFD problems.

Step 2b: Monitoring Convergence

  • Monitor convergence during the solution process; residual plots are essential for assessing solution accuracy.

Step 3: Result Reporting & Visualization

  • Generates considerable data; visualization aids in interpreting CFD results effectively.

  • Common visualization types include:

    • X-Y plots

    • Vector plots

    • Contour plots

    • Streamline plots

    • Animations

Evaluation and Critique of Results

  • Assessing the correctness of CFD results is often challenging, particularly for unfamiliar problems.

  • Address potential errors:

    • Modeling Errors: Incorrect mathematical representations.

    • Discretization Errors: Low-quality mesh and scheme issues.

    • Iteration Errors: Convergence verification.

    • Programming Errors: Potential bugs in code.

Best Practices

  • Verification and validation procedures are crucial for establishing credibility in results.

  • Be wary of results that may look appealing but lack accuracy.