Method for concept design and optimization of twist beam axles

Autoren

Xiangfan Fang, Kanlun Tan, Jens Olschewski, Lehrstuhl für Fahrzeugleichtbau, Universität Siegen

Jahr

2019

Zusammenfassung

The twist beam axle (TBA) is one of the most common type of rear axles. It was developed at first for the VW Polo based on the rigid axle. In comparison to a rigid axle whose wheels cannot move independently due to the direct connection between the two wheel carriers by a cross member, the relative movement of the two wheels in a twist beam is possible. There, the cross member is placed towards the vehicle front direction and connect the two trailing arms in a position between the wheel center and the body mount. During parallel wheel travel, the motion of a twist beam axle causes only slight changes in track, toe and camber that is similar to a rigid axle. However, during an opposite wheel travel the twist beam axle can provide a similar behavior like a multilink suspension. It is thus called a semi-independent axle.
The performance of TBA contains different axle stiffnesses, kinematic characteristics for both parallel and opposite wheel travel and durability. They are determined by design variables such as hard points, shape of the components in 3-dimensional package and their cross section size as well as the selected materials. All requirements are simultaneously influenced by all concept variables. On the one hand these complex interactions enable an enormous conception potential because the large number of the variables can provide a large variety of concept possibilities. On the other hand, the optimization of the product becomes very difficult. The local improvements of one requirement may have negative effects on the other requirements. That is the main challenge of the concept design of twist beam axles.
In the conventional approach for the concept development phase, engineers start usually with several “empirical” or benchmarking based concepts. At first, these initial concepts are roughly designed with CAD software and then converted into FEM and MBS models. Their stiffness and kinematics are calculated by simulations. A relatively good concept that may have a good optimization potential can be identified. This selected concept will be further optimized via many CAD and CAE iterations until it meets all concept targets.
There are two significant problems in this conventional process. First, many manpower and computing resources must be invested in the evaluations of all initial concepts. Moreover, the quality of the initial concepts is highly dependent on the personal experiences of the engineers. It is possible that even the selected concept could only have a limited potential and might not reach the targets after a long time of optimization. The conventional approach cannot meet the need of current development, which requires shorter time, fewer resources and better results. The development of efficient methods is therefore strongly desired and the target of this current work will be presented below.

Mitglieder-Login