Reliance on predictions based on mathematical models can be justified only if experimental evidence demonstrates that predictions are confirmed by the outcome of physical experiments. The key elements of simulation governance are Verification, Validation and Uncertainty Quantification (VVUQ). There are two major objectives of SimGov: Application of design rules and formulation of design rules. The most fundamental technical requirement is solution verification, which is a prerequisite for both the creation and application of design rules.
In the application of established design rules, data verification and solution verification are very important. The goal is to ensure that the data are used properly and the numerical errors in the quantities of interest are reasonably small. Engineering simulation apps for standardizing recurrent analysis processes under the framework of Simulation Governance must be developed by FEA analyst for users who need not have FEA expertise, must possess built-in safeguards to prevent use outside of the range of parameters for which they were designed; must incorporate automatic quality assurance procedures; and must be deployed with detailed description of all assumptions incorporated in the mathematical model and the scope of application.
To ensure the level of reliability needed for professional use, FEA-based engineering sim apps must incorporate solution verification procedures. A-posteriori estimation of relative errors in the quantities of interest is an essential technical requirement of Simulation Governance.
Successful companies often view this democratization process as a way to leverage the expertise of a few specialists in capturing the institutional knowledge of their organization by developing the procedures and tools needed without compromising the quality and accuracy of the results for the users.
The aim is the standardization of recurring mechanical/structural analyses tasks carried out at the component/assembly level for geometrically similar structures (same topology with variable dimensions) in the field of use. Simulation Apps satisfying the technical requirements of Simulation Governance will ensure the level of accuracy and reliability expected in professional use.
By Smart Engineering Simulation Apps we mean FEA-based software tools for standardization and automation of recurring analysis tasks and process workflows for use by non-specialists. Designed to fit into existing analysis processes of an engineering organization or industry, simulation apps (“sim apps”) capture institutional knowledge, best practices and design rules, can be shared by engineering groups at different geographic locations and produce consistent results by tested and approved analysis procedures. When designed to meet the requirements of Simulation Governance, sim apps for engineering use are “smart” because their embedded intelligence enables accurate, efficient, robust, and reliable simulations with built-in quality assurance, so critical for the non-expert user.
Proper application of numerical simulation procedures requires expertise in computational engineering that is not widely or readily available. Standardization deployed by means of Smart Simulation Apps can leverage this expertise for recurring analysis tasks and process workflows similar to the expertise of specialists in applied mechanics made available through classical engineering handbooks. Because classical handbooks present results for parameterized problems solved by classical methods, they have limitations in model complexity and scope. FEA-based Smart Simulation Apps developed by expert analyst on the other hand, deploy verified solutions obtained by numerical means allowing models of much greater complexity to be deployed for users who do not need to have the same level of expertise in numerical simulation technology.
The p-version of the finite element method was developed during the late 1970s and early 1980s and the proof of superior convergence characteristics was established in 1981.
This is at odds with the so-called traditional "h-version" approach in which the degree of the polynomial P is kept at 1 or 2, and the quality of the approximation is improved by decreasing the characteristic size of the element, which is indicated by the letter h. It has been shown in the field of structural analysis that when using the "p-version" with properly constructed meshes, the rate of convergence is exponential as opposed to the "h-version" in which the rate of convergence is algebraic at best.
In classical FEA implementations of the "h-version", the shape functions used for approximating the displacements within the element subdomain are polynomials of degree 1 or 2. Together with the fact that this method uses isoparametric mapping, then the degree of the polynomial is associated with the number of nodes of each element. So polynomials of degree 2 presents the need for using elements with intermediate nodes (midside nodes), while polynomials of degree 1 uses elements with nodes at the vertices only. The approximation of the solution depends strongly on the mesh used for the discretization of the domain, so it is important to construct good meshes, particularly to 'catch' high stress gradients, which can push to noticeably denser meshes. It follows that the quality of the solution is strongly related to the shape and size of the elements.
In "p-version" implementations, the mapping and the approximation are treated separately. This means that the element mapping to the geometry does not depend on the polynomial order of the functions that approximate the displacements. This allows for larger elements to be used and the quality of the approximation is controlled by increasing the degree of polynomials over a fixed mesh.
StressCheck FEA implements the methodology of the P-version type which refers, as outlined above, to the implementation of the finite element method in which the error of discretization is reduced systematically increasing the polynomial order of element shape function on a fixed mesh, rather than decreasing the size of the element analysis via the FEA H-version.
The basic "p" technology version can provide advantages over the classic formulation H-version. For example the convergence rate is increased with high quality results.
The software also provides an error evaluation system with respect to the exact solution (in terms of energy) based on an analytical model that requires a calculation based on a minimum number of three iterations in ascending order of the degree of the polynomial used for the shape functions. Time meshing is considerably cut-down compared to software using the classical approach, and once a user understands how to interface with the software, it provides a rugged tool with key benefits such as:
Initially developed as a 'digital version' of various paper handbooks (such as Peterson, Roark's Formula for Stress and Strain, etc) the software is an instrument that provides in reality much more. The tool can be placed within the SmartApp as it does not need specific training (in the field of finite element analysis) and will serve as a design aid tool.
Everything is set out to allow the user (typically a designer with none or little experience in field of FEA) to benefit from the powerful technology implemented in the StressCheck facility without worrying about the validity of the mesh (as opposed to the H-version method where it can influence on the results).
Moreover the library of case studies included in the CAE Handbook can be easily enriched by users StressCheck data or by EnginSoft and/or ESRD.
The advantages are similar to those of CAE Handbook, whilst the greater versatility in model study allows for not only geometric parameterization but also for greater flexibility in use for the designer, who can count on a greater breadth of models consulted (for example topological parametrization).
For example, it is possible to study a great variety of single joint fasteners, changing the number of plates, the presence or not of bushings, the presence of washers, etc. Applications with SCTB must still for the time being be developed by ESRD, albeit these competences can be shared.
The first webinar of the series addressed the reasons why simulation is mostly performed by specialists; why legacy simulation technologies struggle with standardizing and automating simulation processes due to their inherent complexity and inability to measure solution quality; and how a different numerical simulation approach enables the creation and use of Smart Engineering Simulation Apps and Digital CAE Handbooks that are accurate, efficient, robust, and reliable.
Webinar 2 focused on the functional and technical requirements for the creation and deployment of Smart Simulation Apps. It will be shown that across the simulation functions of knowledge capture, conceptualization, modeling, numerical approximation and prediction, the technology foundation used by Sim Apps should be simple, accurate, efficient, robust and reliable. To ensure the level of reliability expected in professional use, Simulation Apps must incorporate solution verification procedures, an essential requirement of Simulation Governance.
Webinar 3 looked at how industry-specific portfolio of Sim Apps from new generation of simulation tools for analysts and design engineers fit into existing analysis process. How Sim Apps can capture institutional knowledge and best practices to produce consistent results by tested and approved analysis procedures. The systematic approach for selection, creation, testing and deployment was examined and illustrated with an example from a typical structural analysis problem in aerospace applications.
The last Webinar of the series presented examples and reference case studies illustrating the applicability of Simulation Apps for structural problems requiring accurate computation of stresses from detail stress analysis and the computation of fatigue crack propagation life using linear elastic fracture mechanics.
EnginSoft was founded in Italy in 1984 at the forefront of the emerging field of Simulation Based Engineering & Science (SBES). Today the Company retains an international lead in the sector with bases in Italy, France, Germany, UK, Norway, Sweden, Turkey and the USA.
Our biggest asset is our matrix of highly specialist engineers who possess extensive and multidisciplinary expertise in leading edge engineering simulation technologies. A holistic approach to consultancy provides global support for clients in close integration with key partnership companies located in Greece, Turkey, Israel, Portugal, USA, Brasil, Spain and Japan.
Throughout its long and successful history EnginSoft has remained at the forefront of technological innovation with a track record as a 'sector catalyst' for changing the way that SBES and Computer Aided Engineering (CAE) technologies are applied across several industrial sectors. These include: Automotive, Aerospace, Biomechanics, Defence, Energy, Consumer Goods, Civil Engineering and Oil & Gas.
Provision of High Performance Computing (HPC) resources, making EnginSoft the most reliable and knowledgeable partner for product development.
The knowledge and commitment of EnginSoft's engineers' helps clients' identify the most suitable solutions for complex requirements. Leading edge technologies, honed multi-disciplinary methods and access to advanced High Performance Computing (HPC) are all applied to achieve shortened development times and improved resource utilization.
EnginSoft also makes sure that the underpinning expertise is effectively transferred to all customers, who thereby acquire a tangible, strategic advantage.
Via della Stazione, 27
+39 0461 915391
88, Avenue du Genéral Leclerc Bat.A
92100 Boulogne Billancourt
+33 1 41 22 99 30
+49 2234 60386 10
IDEON Science Park
223 70 Lund
+46 46 286 89 00
University of Warwick Science Park
The Venture Centre
Sir William Lyons Rd
Coventry CV4 7EZ
+44 2476 997160
Teknopark İstanbul - Teknopark Bulvarı
Sanayii Mah. No:1/1A 311
34906 Pendik İSTANBUL
+90 216 5041153
© 2016 EnginSoft Group | All rights reserved