| In the theoretical field of parallel manipulators, the conjecture to establish the unified analysis theory is still not solved at present. In the framework of the unified analysis theory, the topology model and parameter model can be communicated, which can improve the analysis efficiency. Under the support of Chinese National Science Foundation, the unified analysis for the parallel hip joint simulator is studied and the unified analysis theory is initially founded. The main research works can be described as follows.(1) Based on the rigid finite element theory and flexible finite element theory, the topology model and parameter model are integrated in the physical meaning which obtains the unified parameter model of the parallel hip joint simulator. Based on the unified parameter model, the forward and inverse postures, forward and inverse velocities, forward and inverse accelerations are analyzed. In addition, the reconstruction curve of motion rule is obtained according to the hip motion rule of human which is recommended by the international standard: ISO14242-1:2002(E). Using the data of reconstruction curve, the numerical analysis of kinematics is carried out. In order to check the results of numerical analysis, a virtual prototype of the parallel hip joint simulator is established and the virtual experiment is carried out. The results of numerical analysis and virtual experiments verify the correctness and efficacy of the unified parameter model.(2) Considering the function of the parallel hip joint simulator and the international standard of artificial hip prosthesis, the task space with three dimensional rotations is determined. By the Monte Carlo method, the work space is solved and a boundary identification method is presented. In addition, a 6S-14 P method is proposed to judge whether the workspace includes the task space. Based on the workspace requirement and low-velocity requirement, the structure parameters are optimized. The numerical analysis for driving velocity of parallel hip joint simulator with optimal parameters is carried out. The numerical results show that the maximum velocity reduces 23.2%, which denotes that the low-velocity driving optimization can effectively decrease the driving velocity and improve the reliability for the parallel hip joint simulator.(3) On the base of the gradient vector in the unified parameter model, the unified stiffness model of the parallel hip joint simulator is established. According to the unified stiffness model, the numerical stiffness analysis is carried out. The stiffness analysis results show that the stiffness of γ direction is significantly lower than the stiffness of a and β direction, so the three rotation stiffness are anisotropic in the task space of the parallel hip joint simulator. In order to check the numerical stiffness results, the virtual experiment is carried out. Comparing the numerical results and the experimental results, the maximum relative error is 29.8% and the minimum relative error is 18.2%. The numerical and experimental stiffness results are all 10 times higher than the allowable stiffness index, so the stiffness of the parallel hip joint simulator is qualified.(4) The bifurcation equation is established by the unified kinematics model of the parallel hip joint simulator, and the judgment condition of the bifurcation is obtained by the static bifurcation equation according to the singular theory, then the unified stability model can be established. Based on the unified stability model of the parallel hip joint simulator, the numerical stability analyses for single parameter and double parameter are carried out. The numerical results show that controlling the approximate uncontrolled degree of freedom when the parallel hip joint simulator enters into the uncontrolled bifurcation domain can improve the stability of the parallel hip joint simulator. In addition, the numerical analysis for control accuracy is implemented. The numerical results of control accuracy show that when the control accuracy is increased from 0.001 mm to 0.01 mm, the maximum of uncontrolled domains under single parameter and double parameter conditions are reduced 0.038 and 0.027, respectively, which provides another theory to improve the stability of the parallel hip joint simulator.(5) Based on the finite element theory, the rigid-flexible elastic dynamics model of the parallel hip joint simulator is established and the numerical analysis of natural frequency and impact parameter are carried out. The numerical results of natural frequency show that the distribution of frequency in the task space is symmetrical. In addition, the minimum of the first order natural frequency is 2.3 which is much higher than the working frequency(1Hz) of the parallel hip joint simulator, which denotes that the parallel hip joint simulator would not occur resonance. The numerical results of impact parameters show that the sensitivity of impact parameter H is the highest, which denotes that the length of middle branch is the main impact factor of the natural frequency. In addition, the numerical results are checked by the the natural frequency experiments.(6) The numerical solution of dynamics is established based on the temporal finite element theory, and the numerical stability for solving differential dynamics model is analyzed. According to the differential dynamics model of the parallel hip joint simulator and the temporal finite element method, the numerical analysis of PD controller and optimal controller are carried out. In addition, the co-simulation model is established in order to check the correctness of the numerical results. The numerical results and simulation results show that the temporal finite element method can correctly solve the differential dynamics model. In addition, the numerical results and simulation results denote that the optimal controller is more accurately complete the given task, which provides a theoretical base for choosing the controllers. The optimal controller is selected to make the control experiment. The control experiment results show that the error range between the motion curve and the desired curve is ±3.5%, which denotes that the parallel hip joint simulator can complete the given task under the optimal controller.Finally, the content of this thesis is summarized and the expectation of related technology development is presented. |
PDF Full Text Request