| Abstract |
In the last two decades, modelling and simulation have become a general tool in the product development of mechanical systems. By using computer simulation from the early phase of the design process, many problems that arise from the dynamical interaction of different sub-systems of the product can be avoided with lower computational costs. This thesis proposal develops dynamics for design (DFD) procedure for the optimal design of spatial holonomic and nonholonomic systems using the multibody system (MBS) approach. The goal of the DFD procedure is to obtain an efficient design cycle by integrating the dynamics of interconnected systems, including nonlinearities, multibody system analysis and parameters estimation, with current design methodologies.
The thesis first illustrates the methodology for using the multibody system approach in modeling holonomic and nonholonomic systems with two common generalized coordinates, Reference Point Coordinate Formulation with Euler Angles (RPCF-EA) and Reference Point Coordinate Formulation with Euler Parameters (RPCF-EP). In the method, the definition of holonomic and non-holonomic constraints is illustrated with the derivation of different joint constraints used in MBS. The typical second-order nonlinear differential algebraic equations (DAEs) have been adopted to illustrate in the equations of motion with the RPCF-EA and RPCF-EP.
The study then uses the MBS approach in modeling selected holonomic and nonholonomic systems with particular emphasis on the generalized coordinates suitable for the parameters estimation process. Solving the equation of motion obtained system acceleration and Lagrange multipliers, the acceleration vector can be integrated to determine the coordinates and velocities. Lagrange multipliers vector can be used to determine the generalized reaction forces acting on system bodies. The discipline of parameter estimation deals with the determination of such parameters using a numerical simulation based on a model of the system under consideration. The typical process for determining the parameters with a model of a system is by using optimization methods. The objective functions represent the difference between measured data from an experiment and simulated data from a model.
Furthermore, the computations and analyses are implemented for selected mechanical systems in the thesis. The symbolic manipulation as well as the computational work of solving the obtained DAEs is carried out using a commercial simulation tool. Once the preliminary design of the mechanical structure has been attained, the stress distribution of flexible component in the grating tiling device is examined, including system stability and structural design aspects. Kinematics simulation results include system velocities and positions with respect to simulation time, which can be used in the parameter estimation process. The most important parameter affected selected mechanical system to apply parameter estimation of a multibody system is the method used to measure the degree of freedom. The parameter estimation procedure with multibody system dynamics is used to obtain the optimal design parameters by comparing the simulated model output with experimental measured data.
Finally, the experimental test rigs are constructed to validate the DFD procedures of the systems proposed in the thesis. The analysis takes the form of a critical evaluation of results obtained using the DFD procedure, parameter estimation and multibody system dynamics modeling on applications considered. Areas of success are highlighted, and situations are identified where currently available techniques have limitations. The benefits of an inter-disciplinary and applications-oriented approach to problems of modelling and design are also discussed and the value in terms of cross-fertilization of ideas resulting from involvement in a wide range of applications is expounded. Detailed simulation and experimental results show that the proposed model with a multibody system dynamics approach is accurate and efficient and can be used in the investigations of holonomic and nonholonomic systems. |
