Search Results

This paper presents an approach for efficiently evaluating motorcycle main frame strength using external loads predicted from measured strain data in our development process. The loads are calculated by simple matrix inversion, and can be used as boundary conditions of static analysis that resembles actual phenomena. The advantage of this method is that it allows relatively precise reproduction of actual boundary conditions without the data usually needed for dynamic simulation such as tire and suspension characteristics which often take large amount of time and man-hour to obtain. Although this approach is simple and common practice, there are a lot of things to be concerned for gaining useful results in a broad range of stages in the motorcycle main frame development process. How we effectively make use of this approach is going to be introduced here.

This paper describes a simple engine model at idling and it applies particularly to idle speed control. Through linearization in the neighborhood of the nominal operating points (650 rpm), the engine is expressed as a reduced-order constant coefficient state variable (2 state) model. It was produced through the system order-reduction method. The strategy for controlling idle speed uses the Linear Quadratic and Integral (LQI) optimal control theory. The tracking controller was designed using a state variable engine model, and the performance index was minimized. Since state variables are artificially introduced, they are not directly accessible. Therefore, they must be estimated in accordance with a stored dynamic model (i.e. observer), in which the engine dynamic behavior is estimated on the basis of a state variable model which represents the engine's internal states, in determining controlling values.