In the field of Uncertainty Management, carhs gmbh has the following expertise and provides the following services:
Model Validation
Model validation is the first fundamental step towards the creation of significant CAE models. Stochastic model validation requires multiple tests to be performed, in addition to a Monte Carlo Simulation. It is only on the basis of repeated tests and stochastic simulation, that it is possible to quantify the levels of confidence in a numerical model.
Stochastic material property data-bases
A stochastic simulation with ST-ORM, in order to be significant, requires the knowledge of the statistical properties of materials and tolerances that form integrating part of the numerical model. To this end, carhs can help to establish corporate data-bases where similar information is stored, and subsequently fed into ST-ORM.
In nature, most of the uncertainty appears in the form of random, or stochastic fields. A random field is a random variable with both spatial and temporal dependency. Good examples of random fields are the surface of the sea, or the pattern of filaments in a composite fabric. Experience with ST-ORM shows that modeling a stochastic field as a simple random variable can lead to gross errors in the response. In particular, it has been found that the initial spatial distribution of imperfections can have macroscopic influence on the response of a metal sheet to stamping. ST-ORMÕs version 2.0 can handle two types of 2D random fields. Random fields constitute the central technology of realistic modeling and enables to channel materials science into the FEM.
Stochastic Design Improvement (SDI)
SDI constitutes the new CAE paradigm of the future. Being based on simulation, it replaces and outperforms all contemporary forms of design such as sensitivity or parametric studies, and Multi-Disciplinary Optimization. In SDI, it is the user who specifies the performance he wants his system to exhibit. With SDI, the behavior of the system is naturally robust and physically significant, rather than numerically optimal. In contrast to MDO and other DOE-type techniques, the cost of SDI does not depend on the number of design variables.
Stochastic Design Generation
Contemporary so-called topology optimizers constitute a means of generating a design, based on an initial volume of material, a set of forces and constraints. While material is gradually removed, these tools arrive at a final topology/shape, which is claimed to be optimal. Not only is this "optimalityÓ never verified, the approach fails in the presence of any form of uncertainty. carhs approaches the task of Design Generation by using ST-ORM in conjunction with CAD tools and proprietary mesh morphing tools. In this manner, ST-ORM stochastically generates a set of design alternatives by randomly modifying a parametrized geometry, while the mesh morphing tool is used to guarantee a correct mesh in each case. The combination of ST-ORM, CAD and mesh morphing tools, enables to easily replace any commercial "topology optimizerÓ, with the great advantage that material and load uncertainties may be taken into account already in the design phase.
Robust Design and Product Quality Management
Quality and robustness are two inseparable characteristics that products should meet. With SDI techniques, robustness and quality may easily be assessed and controlled in a unified manner.
Stochastic Process Simulation
With the advent of EASi-PROCESS, it is now possible to build complete processes, including CAD, CAE, assembly modeling, etc. and to evaluate them in a stochastic context by using ST-ORM. ST-ORM can access and modify any component of the process, and allow therefore to pinpoint the critical paths or process components while contemplating the ubiquitous uncertainty.
High Performance Visualization
Developed jointly with SGI, a special volumetric-based visualization tool allows to display and animate simultaneously all the Monte Carlo samples of an entire structure involved in a ST-ORM analysis. For example, if a stochastic car crash simulation is executed, it is now possible, thanks to this technique, to pinpoint the buckling initiation points, or to find out what are the most likely and unlikely states of the structure at any moment in time. The technique has been shown to provide unprecedented levels of insight and understanding.
Data-Mining
ST-ORM is interfaced with SGI's data-mining tool MineSet. When ST-ORM generates huge amounts of data, MineSet can provide very powerful data-mining techniques to identify trends in the data, and enables to present the results on a Power Screen on in a Virtual Reality cave.