COMPUTATIONAL METHODS FOR LIFE PREDICTION OF MECHANICAL COMPONENTS OF DYNAMIC SYSTEMS

A methodology for life prediction of mechanical components, due to the effect of dynamic loads, is presented. The methodology deals with spatial, constrained mechanical systems that undergo nonsteady gross motion and elastic deformation. The method employs state-of-the-art methods of rigid body dynamic analysis, finite element structural analysis, coupled gross motion-elastic deformation dynamics, and computer-based fatigue analysis. For the purpose of life prediction, these techniques are linked together to construct a new methodology that provides a computer-automated, analytical approach to component life evaluation, to complement conventional fatigue experiments.;To be concrete in developing and evaluating the proposed computational scheme, a vehicle system is used to illustrate the method. Equations of motion for vehicle suspension systems that contain nonlinear spring-dampers and flexible control arms are formulated. Deformation modes are used to reduce the number of degrees of freedom in the finite element model of components, so model coordinates are incorporated in the equations of motion as generalized coordinates. Realizing that simulating mechanical systems is computationally demanding, the formulation introduces relative coordinates for suspension dynamic modeling, in conjunction with Cartesian coordinates for the master body. The number of variables and equations of motion is thus reduced. Dynamic stress analysis is carried out using both an uncoupled gross motion-elastic deformation method and a more accurate approach that takes into account the full coupling between large displacement and elastic deformation. With the aid of vehicle operational scenarios, stress/strain histories of mechanical components during service life are simulated. A local strain approach is used to calculate cumulative fatigue damage at notches, which is used in assessing fatigue life.;Numerical results for both dynamic stress analysis and fatigue life prediction are presented. Components that are the most vulnerable to fatigue failure are identified and their cumulative damage due to dynamic loads is estimated.