제 29회 태성에스엔이 CAE 컨퍼런스 2일차 산업군별 발표자료 - 해외해석사례 세션 발표자료입니다.
일부 자료는 대외비로 인해 공개 불가한 점 양해 부탁드립니다.
SDC Verifier enables code checking in ANSYS
ANSYS software is used by engineers to define the behavior of the structures as well as stresses, displacements and other standard FEA results. But for certification, design approval, tenders and insurance structures in different industries all over the world should be designed and built according to Standard Rules and Regulations. For these structural checks engineers need to use ANSYS results together with some hand calculations, spreadsheets or separate dedicated software, this routine is time consuming and not always easy to follow. SDC Verifier enables automatic code checking within ANSYS environment.
Use of standard and rules based design limits compared to direct Finite Element Analysis Results
For the creation of a (structural) design, engineers use combinations of components as plates, standardized beams and stiffeners and connection techniques as welding, bolting or riveting. With automatic recognition of these items and using checks from rules, regulations and standards, these standard components and connections can directly be verified within a FEA model with linear results.
In this paper / presentation the influence of the simplifications of the design rules will be shown and compared with a process of using a much more complicated fully non-linear calculation approach where the influence of all design parameters including fabrication tolerances have to be taken into account.
Critical infrastructure protection using LS-DYNA explicit simulations
Critical infrastructure represents huge variety of possible targets during any terrorist attack. A mechanical protection is an important part of any infrastructure protection. This presentation describes a process of typical protection system development with regards to numerical simulations especially in explicit ANSYS LS-DYNA code. Still real experiments can be significant as some milestones in virtual development. Typically, material research should start this kind of complex multi-physics simulations. Anti-blast barriers, vehicle attack protection systems and natural gas pipeline protection were analyzed in this study.
Data mining and metamodeling for digital twins
A digital twin is “a dynamic virtual representation of a physical object or system across its lifecycle” [Wikipedia]. For this reason a digital twin produces and collects a massive amount of data. The data comes from sensors or is produced by automatic simulation processes. At the end all the data should be used for “understanding, learning and reasoning”. Here the newest techniques of artificial intelligence can and should be used. ANSYS and especially ANSYS optiSLang offers various advanced methods for simulation process management, data mining and metamodeling (ROM). In the presentation we will show how automated simulation processes and artificial intelligence / machine learning can be used to help within digital twins. Therefore we will use practical applications of world-wide operating companies.
Introduction of Multiscale analysis and examples related to Semiconductor package
Semiconductor package has complex structure made by many kinds of materials. For example, substrate is made by many numbers of grass fibers and epoxy resin. Electronic parts are sealed by resin in which many number of fillers are included. Electronic parts and substrate are bonded by means of solder or other conductive adhesive. These materials have complex microstructure as well. These structures have been simplified in order to reduce analysis cost. However there is a possibility to affect accuracy if the right way for simplification does not be executed.
Multiscale analysis approach has a potential for improvement of the problems drastically. In this presentation, we will introduce our original solution for multiscale analysis and show some examples for validation related to semiconductor package.