The research focuses on developing a methodology for modeling a pneumatic bias-ply tire with the finite element method for vehicle dynamics simulation. The tire as a load-carrying member in a vehicle system deserves emphasized formulation especially for the contact patch because its representation of mechanics in the contact patch directly impacts the handling and ride performance of a vehicle. On the other hand, the load transfer from the contact patch to the wheel hub is necessary for determining the inputs to a chassis. A finite element (FE) tire model has strong capability to handle these two issues. However, the high cost of computing resources restrains its application mainly in the tire design domain. This research aims to investigate how to balance the complexity of a simplified FE tire model without diminishing its capability towards representing the load transmission for vehicle dynamics simulation. The traditional FE tire model developed by tire suppliers usually consists of an extremely large number of elements, which makes it impossible to be included in a full-vehicle dynamics simulation. The material properties required by tire companies' FE tire models are protected. The car companies have an increasing need for a physical-based tire model to understand more about the interaction between the tire and chassis. A gap between the two sides occurs because the model used for tire design cannot directly help car companies for their purpose. All of these reasons motivate the current research to provide a solution to narrow this gap. Other modern tire models for vehicle dynamics, e.g. FTire or TAME, require a series of full-tire tests to calibrate their model parameters, which is expensive and time-consuming. One great merit of the proposed simplified FE tire model is that determining model inputs only requires small-scale specimen tests instead of full-tire tests. Because much of the usability of a model hinges on whether its input parameters are easily determined, this feature makes the current model low cost and easily accessible in the absence of proprietary information from the tire supplier. A Hoosier LC0 racing tire was selected as a proof of modeling concept. All modeling work was carried out using the general purpose commercial software Abaqus. The developed model was validated through static load-deflection test data together with Digital Image Correlation (DIC) data. The finite element models were further evaluated by predicting the traction/braking and cornering tire forces against Tire Test Consortium (TTC) data from the Calspan flat-track test facility. The emphasis was put on modeling techniques for the transient response due to the lack of available test data. The in-plane and out-of-plane performance of the Hoosier tire on the full-tire test data is used for model validation, not for "calibrating" the model. The agreement between model prediction and physical tests demonstrate the effectiveness of the proposed methodology. / PHD / This research aims to develop a method to build a physically-based tire model less relying on the information of products from tire providers for the purpose of vehicle dynamics simulation. The tire model is a mathematical description of the behavior of tires under various operational conditions. The model is said to be ‘physically-based’ if it is derived from physical laws. In contrast, if the model is termed ‘semi-empirical,’ it means that the model is mainly based on tire measurement data. A physically-based model usually gives more insights to and a better understanding of tire mechanics than a semi-empirical tire model. The tire as a load-carrying member in a vehicle system deserves emphasized formulation especially for the tire-road contact patch because its representation of mechanics in the contact patch directly impacts the handling and ride performance of a vehicle. Therefore, a physically-based tire model is preferred. One kind of physically-based models are developed through the multi-body dynamics (MBD) approach. Various full tire tests are required to identify the parameters associated with the model. Since full tire tests should be conducted on professional tire test machines, the high-cost prevents many users to have a tire model of such kind. The other kind of physically-based models are developed through the finite-element method (FEM). The FEM has strong capability to describe the mechanism of tire-road contact and deformation of the tire body. Also, parameters needed by a finite element tire model are basic material properties of different components of the tire structure, which implies the possibility to acquire parameters through small-scale sample tests instead of full tire tests. However, most of FE tire models are developed for tire design with high complexity, not good for vehicle simulation. This research made efforts to degrade the complexity of the FE tire model and tailor the FE modeling technique suitable for the purpose of vehicle simulation. In addition, the process was designed and implemented for obtaining the necessary parameters associated with the model. A Hoosier LC0 racing tire was selected as a proof of modeling concept without any tire property data provided by tire producers. This research has a practical meaning on building tire models independent of tire companies and at low cost.

No description available.