This dissertation is focused on new methods of base dynamic parameters estimation for general mechanisms. Symbolic estimation methods, numerical estimation methods, and experimental estimation methods are presented.; The concepts of mass transfer and moment of inertia transfer through a joint are introduced for symbolic base dynamic parameter estimation. Four propositions are developed for planar joints and successfully extended to spatial joints. A simple equation to predict the number of base parameters for planar mechanisms is also presented. The research leads to a new symbolic base dynamic parameter estimation method that can determine a set of symbolic base dynamic parameters by visual inspection of joint configurations for general mechanisms. This new symbolic method is simpler than other published symbolic methods and is applied to different joint types, including spherical, universal, rotational, and translational. Application examples of the new symbolic method show that it agrees with numerical methods. A discussion of the potential applications of the new symbolic method in machine design is presented.; A MacPherson suspension mechanism is accurately simulated and estimated with singular value decomposition (SVD). The sensitivity of numerical estimation algorithms (SVD) to computational error is discussed. Simulations of spatial mechanisms require very low error to achieve accurate parameter estimation. The numerical estimation results are interpreted using the concepts of mass transfer and moment of inertia transfer, leading to a set of symbolic base parameters. This study lays the foundation for future experimental estimation and design simplification.; A new method for the estimation of base parameters from experimental data is developed. This new method is based on a locally linearized model and the new concept of optimal positions with frequency response analysis techniques. Instead of being driven along ‘optimal trajectories’, the system under consideration is excited by impulse hammer forces at certain optimal operating positions (‘optimal positions’). Base parameters are extracted from the measured system frequency response functions via linear equations. The sensitivity of the least squares solutions of the linear equations to noisy data is suppressed by the optimal operating positions. Optimal operating positions are determined by minimizing the condition number of the coefficient matrix of the linear equations. This method is suitable for general mechanisms, including passive mechanisms. This new approach can estimate all base parameters. Computer simulation and lab experiments on a planar four bar mechanism show that the accuracy of this new method is comparable to the traditional optimal trajectory approach.
展开▼
