Auro Ashish Saha discusses necessary ingredients in the form of modules enabling engineering graduate to become conversant with paradigms of computer aided engineering
Mech

 

Author: Auro Ashish Saha, Department of Mechanical Engineering, Pondicherry Engineering College, Pondicherry, India. Email: asaha@pec.edu

 

Abstract


In this article we discuss all the necessary ingredients described in the form of modules that will enable an engineering graduate to familiarise him/herself and become conversant with the paradigms of computer aided engineering (CAED) and design philosophy practised by the engineering industry.

For seamless integration into the work flow culture of a typical CAED, specific competencies are desirable or need to be cultivated, such as analytical, geometric and computational modelling, design and analysis using equation-based analytical or numerical simulation and scripts; understanding of generating drawing using different methodologies such as geometric models, from CAED script or conventional drafting using a computer aided drawing tool.

The unique feature of this presentation is that the whole training guide is developed using six basic modules and extended on an incremental basis for various applications in CAED.

Introduction to modelling, design and analysis


Product engineering in these competitive times is accomplished through virtual design [1] with the help of simulation. A prior step to simulation is modelling. Modelling has different meanings associated with it: when expressed graphically, it represents geometric features or, alternatively, represents a mathematical formulation.

Repeated analysis of the model for various parametric changes enables one to obtain an optimised design. The following sections introduce an integrated virtual product engineering concept with the help of six basic modules. An incremental learning methodology from a lower level module to a higher level module will help to develop the specific competencies for seamless integration into the work flow culture of CAED.

Insight into geometry and meshing


The first stage in any product engineering starts with creating a geometry of the model. For subsequent computational analysis, this geometry has to be meshed by generating grids on the model. Salome [2] is an open source standalone application used for generation of CAD model, pre and post-processing of numerical simulations.

The featured image shows the use of geometry and mesh workbenches available in Salome for preprocessing that uses a simple primitive 3D box geometry and a uniform structured mesh algorithm.

Complex 3D geometry can be created and meshed using various meshing algorithms available for generating both structured and unstructured meshes and also enabling import and export of different geometry and mesh files.

Geometric modelling and drafting


Solid modelling is used to create virtual models of real objects with surfaces and volumes that are indistinguishable from the actual objects [3].

Features of solid modelling are:

  • Provides greater visual aid for products;
  • Experimenting the assembly before their manufacture to examine any unforeseen problems;
  • Incorporating changes in shape and size before finalising the design required for manufacture;
  • To present concepts (architects use to show the exterior/interior of proposed building; design engineers use to show the concepts of shapes of cars);
  • Doing engineering analysis such as stress calculations, temperature distribution calculations for products subjected to loads or heat transfer.

Approaches to solid modelling are:

aaeng-2

Figure 2: Drawing and Dimensioning with FreeCAD

Bottom-up modelling – using points, lines, areas and volumes. By extrusion or rotating of areas model is created.
Top-down modelling – using area and volume primitives. Using Boolean operations model is created.

FreeCAD is an open source modular parametric 3D modeller [4]. Figure 2 shows drawing of orthographic and isometric views of a cube with circular through hole on its faces. The 3D model is created in the part workbench with standard cube and cylinder primitives, followed by Boolean operations. The drawing is generated using the Drawing and external module Drawing Dimensions workbench.

Figure 3 shows 3D model of Assembled Parts created from Part and external module Assembly 2 workbench of FreeCAD following the top-down modelling approach. Circular edge constraint is applied on parts for assembly. The same 3D model of assembled parts created using Sketcher, Part Design and external module Assembly 2 workbench is shown in Figure 4 following bottom up modelling approach.

aaeng-3

Figure 3: 3D Model of Assembled Parts with Part and Assembly 2 Workbench of FreeCAD

aaeng-4

Figure 4: 3D Model of Assembled Parts with Sketcher, Part Design and Assembly 2 Workbench of FreeCAD

Simulation based CAED


Simulation is used at different stages of the product life cycle, particularly when products become complex. Before product designs are finalised, simulations are carried for alternative design options within the framework of a product life cycle management (PLM) system [5].

Simulation of multi-physical phenomena involves computationally solving the governing equations described by PDEs. Elmer is a finite element method based computational tool available for numerical analysis [6].

aaeng-5

Figure 5: Geometry and Mesh with Elmer

A geometry and meshing capability within Elmer is demonstrated in Figure 5 for structural and thermal analysis. Figures 6 and 7 show the displacement and stress field developed in a cantilever beam of 1.0 m x 0.05 m x 0.1 m size, when subjected to an end load of 2000 N. The material properties used in the linear elasticity solver of Elmer simulation are, Poisson’s ratio = 0.37, Young’s Modulus = 10 x 10 9 /m2, Density = 550 kg/m3.

aaeng-6

Figure 6: Cantilever Beam Deflection with Elmer

aaeng-7

Figure 7: Cantilever Beam Stress with Elmer

Figure 7 shows the temperature field developed in a rectangular longitudinal fin 1.0 m x 0.05 m, using the heat transfer solver of Elmer simulation when subjected to base temperature of 100 0C, heat transfer coefficient 0.4 W/m2 K at external temperature of 25 0C with insulated tip and unit material properties.

 

aaeng-8


Figure 8: Fin Temperature with Elmer

CAED analysis using equation solvers


EES is a general equation-solving program that can numerically solve thousands of coupled non-linear algebraic and differential equations [7]. Equations can be expressed in any order following the syntax rules for a typical mechanical design as shown in Table 1.

aaeng-10

Table 1: CAED analysis using Engineering equation solver and AutoLISP script

Script based CAED and drafting


AutoCAD is a commercial application available for 2D and 3D computer-aided design (CAD) and drafting. AutoLISP programming language enables customising the functionality of AutoCAD at an advanced level [8]. Advanced calculations can be performed through AutoLISP expressions. Some of the AutoLISP function and command syntax like defunc – define function, list – variable with more than one element, setq – primary assignment command, print – print command, +, -, *, /, expt – mathematical expressions has been used for demonstrating the mechanical design as shown in Table 1.

AutoLISP can also execute AutoCAD commands to create entities in the existing drawings or even make complete drawings from the scratch [9]. Table 2 shows the AutoCAD command script and the corresponding AutoLISP script for obtaining orthographic drawing as shown in Figure 9.

aaeng-11

Table 2: Drafting using AutoCAD and AutoLISP script

aaeng-9

Figure 9: Script based Drafting (Typical)

Parallel computing


The general availability of massively parallel CPU/GPU hybrid computing hardware enables to simulate inherently difficult and complex problems that require extensive resources so far unviable with present day only CPU computing systems.

CPUs typically have only four or six or eight processing cores, whereas modern GPUs available from Nvidia and AMD have many hundreds of cores, thus enabling desktop workstations with supercomputing capabilities [10].

OpenCL is an open-source, GPU programming language for general-purpose computations on heterogeneous systems supported by Intel, AMD and Nvidia. CUDA is an open-source GPU programming language developed for NVidia GPUs.

Conclusions


All the key stages of computer aided design and engineering in the product lifecycle management framework was illustrated in the form of modules. The modules were incrementally introduced and their linking is emphasised in the work flow of a CAED system.

References

  1. Chiodi, M., 2011. An innovative 3D-CFD-Approach towards Virtual Development of Internal Combustion Engines, Vieweg+Teubner Research, Germany.
  2. Salome Platform Documentation Version 7.6.0.
  3. Moaveni, S., 2009. Engineering Basics, Cengage Learning, New Delhi.
  4. FreeCAD Documentation Version 0.15.
  5. Stark, J., 2015. Product Life Cycle Management, Springer, London.
  6. Elmer Solver Manual Version 8.0.
  7. EES Manual Version 9.904D.
  8. AutoLISP Developer’s Guide Version 2013.
  9. Ambrosius, L., 2015. AutoCAD Platform Customization: AutoLISP, Wiley, Indianpolis.
  10. Saha, A. A., et al. 2013. Advanced CFD Modeling using GeForce GPUs, International Journal of Mechanical Engineering & Computer Applications, 1, pp. 61-69.

 

http://www.engineersjournal.ie/wp-content/uploads/2015/10/aaeng-1.pnghttp://www.engineersjournal.ie/wp-content/uploads/2015/10/aaeng-1-300x300.pngDavid O'RiordanMech3D
  Author: Auro Ashish Saha, Department of Mechanical Engineering, Pondicherry Engineering College, Pondicherry, India. Email: asaha@pec.edu   Abstract In this article we discuss all the necessary ingredients described in the form of modules that will enable an engineering graduate to familiarise him/herself and become conversant with the paradigms of computer aided engineering (CAED) and design philosophy practised...