The main goal of the paper is to design and implement a framework based on the cellular automata (CA) method, which is dedicated to numerical simulations of microstructure evolution in metallic materials under thermal and mechanical processing. Major assumptions and implementation details of the proposed solution involving classes containing dedicated fields and methods are discussed. Finally, the cellular automata framework (CAF) is tested for selected case studies supported by the Windows Workflow Foundation (WWF) approach. Particular attention is put on modelling simple grain growth, static recrystallization and phase transformation phenomena occurring at the microstructure level. Obtained results of simulations as well as performance characteristics are also presented in the paper. As a result, the CA framework, which supports design of complex algorithms with flexible data flow and reusable components is proposed.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Computer simulation programs based on fundamental physics laws become more and more sophisticated and accurate in description of material behaviour under deformation conditions. That contributes to the general knowledge about the microstuctural phenomena that occur during metal processing, e.g. hardening, recovery, and both dynamic and static recrystallization. Therefore there is a need to take explicitly into account microstructure features i.e. grains, grain orientations, inclusions, cracks, different phases etc. during the simulations. These microstructural features and their local interactions under loading can result in a very elevated material properties. That is a reason why many researchers are working towards creation of digital material representation (DMR), which provides detailed description of material structure. The DMR is than an input data for the multi scale simulation of large scale industrial processes as well as for the very local investigation of deformation mechanisms. Available approaches for the DMR are usually in-house codes, which are developed individually and are based on various computational methods e.g. Voronoi Tesselation, Sphere Growth, image processing, Monte Carlo or Cellular Automata. However, character of the in-house codes is a limiting factor in application in real industrial conditions, where for the sake of simplicity a user friendly automatic software is required. That is the reason why the main goal of this research is to create an easy to use, user friendly software library that can be used not only to create the DMR with interesting features represented explicitly but also can be used to study micro scale phenomena during deformation i.e. recrystalization, phase transformation, inhomogeneous flow due to inclusions distribution. Such a software combined with the macro scale analysis methods i.e. finite element (FE) method and with optimisation algorithms can be used to redesigned the material microstructure in order to obtain specific material properties after manufacturing stage. CA Framework is created by using object oriented design, which facilitates its further application in external software. It is composed of several classes responsible for representation of space, automaton, cells, configuration and internal variables. Main assumptions regarding the CA Framework and its functionality will be presented and discussed in the paper. Examples of wide range of generated microstructures for one and two phase materials as well as microstructures with different shapes and size of inclusions will be shown as well. Finally obtained digital microstructures will be used as input data for the FE simulations to investigate e.g. influence of microstructural features on final material properties.
PL
Niniejsza praca przedstawia projekt oraz implementację framework'u programistycznego dedykowanego do wspomagania rozwoju algorytmów opartych o technologie automatów komórkowych. Główny cel pracy skupia się na optymalizacji architektury framework'u tak, aby osiągnąć jednocześnie wysoką efektywność obliczeniową oraz elastyczność w możliwościj projektowania nowych algorytmów. Obecnie różnego algorytmy oparte o automaty komórkowe mogą być definiowane na bazie zaproponowanego framework'u, począwszy od prostej gry w życie, a skończywszy na skomplikowanych algorytmach zarodkowania i rozwoju mikrostruktur materiałowych. Framework połączony jest ze specjalnym modułem wizualizacji otrzymanych wyników zaimplementowanym za pomocą technologii OpenGL. Rezultaty uzyskane dzięki połączeniu modułów zostały również przedstawione i omówione w niniejszym artykule.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.