Modal properties identification in motorcycle design : a comprehensive study on structural dynamics and cbar welded joints / Khaironi Binti Mat Jelani

Faculty of Mechanical and Automotive Engineering Technology

Thesis (Master of Science) -- Universiti Malaysia Pahang – 2024

Accurate prediction of motorcycle chassis structure dynamic characteristics, such as natural frequencies and mode shapes, is essential for improving performance and safety. Current modelling methodologies frequently struggle to produce solid predictions, resulting in disparities between simulated and actual behaviors. This research discusses an important issue by identifying the motorcycle chassis structure and analyzing its dynamic parameters. It also presents a method for optimizing the length of element connections using the CBAR element, with the goal of improving structural analysis accuracy and dynamic behavior predictions. Besides, this study proposes a method for identifying the best element connector length using the CBAR element in the motorcycle structure. The investigation of the modal characteristics of motorcycle chassis construction was conducted using experimental modal analysis (EMA) and finite element analysis (FEA). Improved modelling approaches on motorcycle chassis construction are necessary to evaluate the accuracy and reliability of modal characteristics prediction in studies. The next step is to fine-tune the CBAR element connector length in the motorcycle chassis structure for improved correlation of modal characteristics, improving the accuracy of structural analysis and prediction of dynamic behavior. By using MSC Nastran/Patran, a few measurements and modal investigations were conducted utilizing SOL 103. In experimental modal analysis, an impact hammer and a roving accelerometer were used to assess the dynamic behavior of the motorcycle chassis structure under freefree boundary conditions. To simulate the dynamic behavior of the motorcycle chassis, the FE model included CBAR element connections for the welded joints. This model has been modified using Nastran SOL 200 based on testing data. The comparison showed that the error percentage between EMA and FEA was less than 10% for the initial model with updates. Furthermore, updating the model for five mode forms lowered the error percentage by 0.17% over updating for four mode shapes. The study improved CBARbased FE model, which includes improving the length element joining structure, showed positive results. After updating three parameters, the overall error percentage reduced to 2.48%, down from 2.54% with two parameters. Adding a 3 mm gap to the update parameter increased convergence. The investigation improved the design and analysis processes to produce more accurate computer simulations and provided an approach to overcome characteristic differences in computer models.