ANSYS Mechanical APDL for Finite Element Analysis

Mary Kathryn Thompson, John Martin Thompson

• Publisher: Butterworth-Heinemann
• Year: 2017
• Print ISBN: 9780128129814
• eBook ISBN: 9780128131107
• Topic: Technology & Engineering
• Subtopic: Engineering General
• Index: Technology & Engineering

ANSYS Mechanical APDL for Finite Element Analysis provides a hands-on introduction to engineering analysis using one of the most powerful commercial general purposes finite element programs on the market. Students will find a practical and integrated approach that combines finite element theory with best practices for developing, verifying, validating and interpreting the results of finite element models, while engineering professionals will appreciate the deep insight presented on the program's structure and behavior. Additional topics covered include an introduction to commands, input files, batch processing, and other advanced features in ANSYS.The book is written in a lecture/lab style, and each topic is supported by examples, exercises and suggestions for additional readings in the program documentation. Exercises gradually increase in difficulty and complexity, helping readers quickly gain confidence to independently use the program. This provides a solid foundation on which to build, preparing readers to become power users who can take advantage of everything the program has to offer.- Includes the latest information on ANSYS Mechanical APDL for Finite Element Analysis- Aims to prepare readers to create industry standard models with ANSYS in five days or less- Provides self-study exercises that gradually build in complexity, helping the reader transition from novice to mastery of ANSYS- References the ANSYS documentation throughout, focusing on developing overall competence with the software before tackling any specific application- Prepares the reader to work with commands, input files and other advanced techniques

Chapter 1

Introduction to ANSYS and Finite Element Modeling

Abstract

This chapter provides an introduction to finite element analysis and the ANSYS Mechanical APDL family of software. It begins with an overview of the finite element method, its benefits, and its limitations. Next, it introduces a basic 10-step procedure for finite element analysis. This is followed by a brief history of ANSYS and finite element software programs. Next, it summarizes the current ANSYS Mechanical APDL products and program capabilities. Finally, it describes the program’s evolution and how that influences the use of ANSYS, Inc. products.

Keywords

Analysis procedure; engineering analysis; finite element analysis; ffinite element method; ANSYS Inc.; ANSYS Mechanical APDL; history of ANSYS; software license; forward compatibility; backward compatibility

Chapter Outline

1.1 What Is the Finite Element Method?

1.2 Why Use the Finite Element Method?

1.3 Basic Procedure for Finite Element Analysis

1.4 Engineering Software—Not an Engineer

1.5 A Brief History of ANSYS and Finite Element Analysis

1.6 ANSYS Today*

1.7 ANSYS Licensing

1.8 Functionality and Features of the ANSYS Mechanical APDL Family

1.9 ANSYS: Backward Compatibility and Legacy Code

Suggested Reading Assignments:

None

This chapter provides an introduction to finite element analysis and the ANSYS Mechanical APDL family of software. It begins with an overview of the finite element method, its benefits, and its limitations. It summarizes the current ANSYS Mechanical APDL products and program capabilities. Finally, it describes the program’s evolution and how that influences the use of ANSYS, Inc. products.

1.1 What Is the Finite Element Method?

The finite element method (FEM) is a mathematical technique for setting up and solving systems of partial differential (or integral) equations. In engineering, the finite element method is used to divide a system whose behavior cannot be predicted using closed form equations into small pieces, or elements, whose solution is known or can be approximated. The finite element method requires the system geometry to be defined by a number of points in space called nodes. Each node has a set of degrees of freedom (temperature, displacements, etc.) that can vary based on the inputs to the system. These nodes are connected by elements that define the mathematical interactions of the degrees of freedom (DOFs). For some elements, such as beams, the closed form solution is known. For other elements, such as continuum elements, the interaction among the degrees of freedom is estimated by a numerical integration over the element. All individual elements in the model are combined to create a set of equations that represent the system to be analyzed. Finally, these equations are solved to reveal useful information about the behavior of the system.

Just as a regular polygon approaches a perfect circle as the number of sides approaches infinity, a finite element model approaches a perfect representation of the system as the number of elements becomes infinite. Since it is impossible to divide the system into an infinite number of elements, the finite element method produces the exact solution to an approximation of the problem that you want to solve. When the number of elements becomes sufficiently large, the approximation becomes good enough to use for engineering analysis. However, this may increase the number of equations to be solved beyond the point where it is practical or desirable to solve them by hand. For this reason, the finite element method is associated with computer programs that set up, solve, and visualize the solutions of these large sets of equations for you.

1.2 Why Use the Finite Element Method?

The cost, in terms of the manpower and computer resources, required to set up and solve a finite element model for a simple problem like a cantilever beam is very high compared to the benefit. Simple problems can—and should—be solved with simple methods (or obtained from engineering handbooks). But not all problems are simple. For example, if a bridge is built using a simple truss supported by two piers, the deflections and stresses in the bridge can be found using information taught in an introductory statics and strength of materials class. But as the complexity of the truss increases, solving this problem using the engineering fundamentals becomes more difficult, leaving the analyst with long hours of error-prone calculations. As system complexity continues to increase, closed-form analysis rapidly becomes impossible. The real benefit of finite element analysis lies in the ability to solve arbitrarily complex problems for which analytical solutions are not available or which would be prohibitively time consuming and expensive to solve by hand.

1.3 Basic Procedure for Finite Element Analysis

There are 10 basic steps in any finite element analysis. First, the solid model geometry is created, the element type(s) and material properties are defined, and the solid model geometry is meshed to create the finite element model. In ANSYS, these steps are performed in the Preprocessor (PREP7). Next, loads and constraints are applied, solution options are defined, and the problem is solved. These steps are performed in the Solution processor (SOL). After the solution is ready, the results are plotted, viewed, and exported in one of the postprocessors (POST1 or POST26). Finally, the results are compared to first-order estimates, closed-form solutions, mathematical models, or experimental results to ensure that the output of the program is reasonable and as expected. (Processors will be addressed in more detail in chapter 2.)

/PREP7

1. Define the Solid Model Geometry

2. Select the Element Types

3. Define the Material Properties

4. Mesh

/SOLUTION

5. Define the Boundary Conditions

6. Define the Loads

7. Set the Solution Options

8. Solve

/POST1 or /POST26

9. Plot, View, and Export the Results

10. Compare and Verify the Results

It is sometimes possible to omit one or more steps. For example, the default solution options are often sufficient for a simple analysis. It is possible to perform some steps out of order. For example, the element types and material properties can be defined in either order. Similarly, the loads and boundary conditions can be defined in either order. It is occasionally necessary to perform these steps out of order. For example, solid model geometry is not required for a finite element analysis. When the nodes and elements are generated directly, the element type(s) must be specified before the geometry can be created. Finally, complicated analyses may involve multiple trips through one or more processors.

For simplicity, this 10-step procedure will be used in this book whenever possible.

1.4 Engineering Software—Not an Engineer

As with all computer programs, the quality of your results will depend on the quality of your model. This includes the accuracy of the material properties, the appropriateness of the material models, how closely the simulated geometry and loads match the actual geometry and loads, and the validity of the simplifications and assumptions made. Simply put, Garbage In=Garbage Out. Finite element software programs can be thought of as very sophisticated calculators that help you to analyze engineering systems that could not otherwise be evaluated. They integrate the section properties of the system with the material properties to generate the equations to be solved. They convert the applied loads to the appropriate forms and apply them to the specified DOFs. They solve the generated system of equations. And, they help you to visualize and understand the results. But a finite element program will not comment on the validity of any assumptions made in setting up the model as long as the laws of physics are not violated. It also will not ensure that you are using the correct laws of physics for a given problem. Any errors that the program reports will be associated with the use of the program, and not with the physical or analytical system. In addition, it will not provide any commentary on the quality or implications of the results. Finite element software is only a tool. In the end, you, and you alone, are responsible for determining whether or not the results of your finite element model can be used to make or justify engineering decisions.

1.5 A Brief History of ANSYS and Finite Element Analysis

The finite element method was first proposed in the early 1940s as a numerical technique for solving partial differential equations. At that time, a mesh of elements could be defined and the interaction of the elements could be used to create the system of equations to be solved. However, the system of equations still had to be solved by hand. This limitation rendered the finite element method an academic curiosity until the early 1960s when computers that could solve large systems of simultaneous equations started to become available. This made it possible to apply the finite element method to general problems. As a result, interest in using the finite element method in engineering practice began to grow.

Early finite element programs were specialty codes that were developed to solve a specific type of problem. They generally contained a single element type (e.g., beams, axisymmetric shells, or plane stress solids) and included a single type of physics (structural, thermal, etc.). This limited the type of problem that each program could solve. It also meant that there were no standard analysis tools. It was common for different groups in the same organization to use different computer programs. In many cases, each group of engineers developed and used its own finite element code. This led to concerns about the compatibility of results from different programs, the overall quality of those results, and whether the engineers’ time was being used efficiently.

1.5.1 The Development of NASTRAN

In 1965, the United States National Aeronautics and Space Administration (NASA) issued a request for proposals to create a computer program that could be used by all of its engineering organizations to solve a variety of structural problems related to the development of lunar exploration technology. The resulting program was known as NASTRAN. In 1969, NASA began to develop coupled thermal-structural capabilities in order to predict the optical performance of a large space telescope system that was exposed to changing orbital thermal conditions. By 1971, NASTRAN® was available for commercial use. It is still the default finite element program in the aerospace industry today.

1.5.2 The Development of ANSYS

While NASA was focused on lunar exploration, Westinghouse Electric Corporation was developing nuclear reactors for space propulsion and nonconventional energy production. Like their aerospace counterparts, the Westinghouse mechanical and nuclear engineers needed to predict transient stresses and displacements in reactor systems due to thermal and pressure loads. Dr. John Swanson, then an employee at the Westinghouse Astronuclear Labs in Pittsburgh, believed that an integrated, general-purpose finite element program would save both time and money when doing these types of calculations. He began developing such a program, called STASYS, for Westinghouse in 1969.

In 1970, John Swanson left Westinghouse and founded Swanson Analysis Systems, Inc. (SASI) where he continued to develop a commercial general-purpose finite element program that he called ANSYS®. The original version of ANSYS contained 40 elements of various types (springs, dampers, beams, bricks, etc.) including several elements with thermal degrees of freedom. Westinghouse became ANSYS’s first customer by the end of the year. The program was rapidly adopted by other companies and became the default finite element program for much of the power industry. Today, ANSYS products are used in all major engineering fields including the aerospace, automotive, chemical processing, construction, consumer goods, electronics, energy, health care, offshore, marine, and materials industries.

1.5.3 The Evolution of ANSYS

With every new release since 1970, new features and functionality have been added to ANSYS. Many additions were specifically developed for the program. For example, the first elements with thermoelectric (1975) and electromagnetic (1983) DOFs were developed by ANSYS engineers. Some capabilities have been added by interfacing ANSYS with other programs. For example, computational fluid dynamics (CFD) capabilities were first added in 1989 by building an interface between SASI’s ANSYS and Compuflo’s FLOTRAN. Similarly, explicit dynamics capabilities were added in 1996 by developing an interface between ANSYS and Livermore Software Technology Corporation’s LS-DYNA™. Finally, some capabilities have been added by incorporating other programs into ANSYS. For exam...

Table of contents

• Cover image
• Title page
• Table of Contents
• Copyright
• Dedication
• Preface
• Acknowledgments
• Chapter 1. Introduction to ANSYS and Finite Element Modeling
• Chapter 2. Interacting with ANSYS
• Exercise 2-1. Static Axial Loading of a Notched Plate in Tension
• Chapter 3. Creating and Importing Geometry
• Exercise 3-1. Bottom-Up Solid Modeling of a Plate With a Central Hole Using Quarter Symmetry
• Exercise 3-2. Top-Down Solid Modeling of a Pipe Flange Using Symmetry
• Exercise 3-3. Structural Analysis of a Simple Warren Truss Using Direct Generation
• Chapter 4. Elements and Element Input
• Exercise 4-1. Modeling a Simple 1D Cantilever Beam Using Beam Elements
• Exercise 4-2. Modeling a Simple 2D Cantilever Beam Using PLANE Elements
• Exercise 4-3. Modeling a Simple 3D Cantilever Beam Using SOLID Elements
• Chapter 5. Defining Material Properties
• Exercise 5-1. Temperature-Dependent Plasticity Analysis of a Plate with a Central Hole
• Chapter 6. Meshing
• Exercise 6-1. Determining the Mesh Convergence of a Heated Plate With a Central Hole
• Chapter 7. Selecting Entities
• Chapter 8. Solution
• Exercise 8-1. Time Varying Heat Conduction Through a Composite Wall
• Chapter 9. Postprocessing
• Exercise 9-1. Postprocessing an Axisymmetric Cylindrical Pressure Vessel Using Element Tables
• Exercise 9-2. Postprocessing a 3D Thermal Model With Geometric Discontinuities Using Power Graphics
• Exercise 9-3. Postprocessing a Cylindrical Structural Shell Using PowerGraphics, Results Coordinate Systems, and Load Case Combinations
• Chapter 10. Input Files
• Exercise 10-1. Using the Sequential Method to Create an Input File for 1D Steady-State Conduction Through a Steel Clad Copper Pan
• Exercise 10-2. Using the Concurrent Method to Modify an Input File for Steady-State Conduction Through a Cladded Plate
• Exercise 10-3. Using the Direct Method to Create a Batch File for Steady-State Conduction Through a Cladded Plate With Varying Surface Temperatures
• Appendix. Chapter and Section Numbering for Selected ANSYS Mechanical APDL 17.2 Documentation
• Index