An Automated Approach to Reduce the Computational Effort for System Level Dynamics Analyses

The operational and organizational efficiency of a company is measured by its ability to manage multiple resources simultaneously while maintaining high levels of reliability and quality. Enhancing this capability provides a clear competitive advantage in the market by improving productivity and, consequently, meeting increasing customer demands. In the context of dynamic analyses, studying the vibrations of aircraft engine components often requires simulations that can last for hours or even days, with computational costs directly tied to the complexity of the component. These simulations yield accurate results only if the preparation of the analyzed models is carried out meticulously, adhering to specific rules for effective modeling. Unfortunately, pre-processing is a delicate and time-consuming phase in the design and verification process of a component. Moreover, certain sub-phases are mechanical and repetitive, making them prone to unintentional errors that increase operational costs. In the industrial world, the process is further complicated by the use of different software tools; Ansys Workbench and MSC Nastran are among the most widely utilized. While both are highly effective in their respective domains, integrating them poses challenges, especially when discrepancies in results arise, caused by different modeling approaches. The need to automate these processes led to the development of a tool designed to alleviate users from repetitive and standardized tasks. This tool introduces innovative approaches to modeling contact interactions between component parts, effectively replacing previous methods. By testing the tool—entirely programmed in Python—on complex models, significant time savings were observed compared to manual processes. In addition to minimizing human errors, the tool addresses several critical issues inherent in the design standard procedures. To simplify the implementation, certain assumptions were made, focusing on the most frequent cases and limiting the application to static axisymmetric frames. As a result, the tool will require further integration and improvements to expand its capabilities in the future.