Ingo Hahn and Jörg C. Sturm, MAGMA Gießereitechnologie GmbH, Aachen Germany
Twenty years after the introduction of simulation software for foundries into the industry, casting process simulation has become an accepted tool for process and design lay-out. Casting process simulation always displays the status quo of its expert user. The user decides if the rigging system or process parameters lead to an acceptable result. Additionally, proposals for optimized solutions have to come from the operator.
One of the biggest benefits of the casting process is also its biggest downfall: Everything happens at the same time and is coupled. Changes in one process parameter impact many casting quality defining features during the process. Multiobjective autonomous optimization offers a way out.
Autonomous optimization uses the simulation tool as a virtual experimentation field and changes pouring conditions, gating designs or process parameters and this way tries to find the optimal route to fulfill the desired objective. Several parameters can be changed and evaluated independently from each other. Autonomous optimization tools take the classic approach of foundry engineers, to find the best compromise and use validated physics. This not only further reduces the need for trial runs to find the optimal process window, but allows the detailed evaluation of many process parameters and their individual impact on providing a robust process.
Obviously, what can be simulated can be optimized. Optimization, therefore, is not a replacement for process knowledge and expertise. Despite beliefs to the contrary, the simulation user of the future needs to know the objectives and goals, and especially the quality criteria that are needed to reach these goals. The questions to ask a program are easy: What is a good gating system? To answer this question, quantitative solutions are required.
An old foundry man’s dream is becoming reality: trial and error is not performed on the shop floor but on the computer. The foundry man defines his optimization goals and can evaluate the best possible solution. He also receives quantitative information about the sensitivity of important process parameters and can assess the robustness of his designs.
The paper will give an overview on the state of the art of virtual autonomous optimization on selected industrial examples.
Please read the complete publication in the linked PDF.