Active Control Of Flexible Two-Link Manipulator

With the development of science and technology, lightweight and flexible components have been widely used in mechanism. Meanwhile, the requirements on operating speed and precision of mechanism become higher and higher. Research in this area comes to the so-called flexible multibody dynamic system or rigid-flexible coupling dynamic system. The flexible manipulator is a typical application in this area. When the manipulator moves with a large-scale movement, elastic vibration of flexible arm will occur synchronously. These two motions are coupled and affect each other, which results in the complicated dynamics of system such as strong nonlinearity and coupling etc. For the flexible one-link manipulator, single rigid-flexible coupling exists in the system, i.e., the coupling of large motion of the system with the elastic vibration of flexible arm, so the problem of dynamic modeling and control for the system is relatively easier. But for the flexible multi-link manipulator, the system exist not only the rigid-flexible coupling but also the flexible-flexible coupling between elastic vibrations of each flexible arms, so the work of modeling and control for the system is very difficult. Since flexible manipulator system has lots of applications in high tech engineering areas such as aerospace, aviation and robot, it is of importance, both theoretically and practically, to study its dynamic modeling and active control method.On the other hand, with good mechanical-electric coupling quality, piezoelectric material can make both a sensor and an actuator, which explains its wide application in active vibration control systems. Piezoelectric actuator can make part of a component by means of either adhering or embedding, which is useful for vibration control of flexible components with large-scale movement.This dissertation studies the dynamic modeling and active control of flexible two-link manipulator. This research was funded by the National Natural Science Foundation of China (Grant Nos. 10772112 and 10472065), the Key Project of Ministry of Education of China (Grant No. 107043), the Specialized Research Fund for the Doctoral Program of Higher Education of China (20070248032) and the Key Project of Education Committee of Shanghai (Grant No. 09ZZ17). The main research and achievements are as follows: (1) The dynamic modeling theories and vibration control of flexible manipulators are comprehensively reviewed. The research extent and contents of this dissertation are put forward.(2) The dynamic modeling of flexible two-link manipulator is studied. Numerical simulations with known and unknown large motion of the system are carried out to demonstrate the necessity of active control for the system. The study in this section provides dynamic model for the following control studies for the system.(3) This dissertation discusses the linear optimal control of flexible manipulator, where the two joint torques are only used to drive the system and PZT actuator is not used. Simulation results indicate that, with linear optimal controller, the flexible arm can reach an expected position with suppressed vibration, however the time taken to position is longer than expected.(4) Nonlinear control strategy of the flexible manipulator is studied in this dissertation. The PZT controller and the joint controller are synthetically designed, where the joint controller is used to guarantee the large motion of the system and the PZT controller is used to control the vibration of the two flexible arms. Simulation results indicate that, nonlinear control strategy works well with precise positioning, suppression of vibration and time control.(5) Due to the wide applications of PD control strategy in practical engineering, fuzzy PD control method is studied for the flexible two-link manipulator in this dissertation. Fuzzy controller is used to adjust the control weighting coefficient of PD strategy to compose the expert fuzzy PD controller. The fuzzy PD controller is compared with the classical PD controller. Simulation results indicate that, when the control weighting coefficient is better chosen, the classical PD controller and the fuzzy PD controller can both achieve better control effectiveness, but the control cost of the classical PD controller is larger than that of the fuzzy PD controller. When the control weighting coefficient is worse chosen, the control effectiveness of the classical PD controller becomes worse and instability of control system may possibly occur, whereas the fuzzy PD controller may also achieve a better control effectiveness. This indicates that the fuzzy controller may effectively adjust the control weighting coefficient of PD strategy to obtain a better control effectiveness, it may reduce the parameter adjusting work using the classical PD strategy.The topic of dynamic modeling and control of the piezoelectric flexible manipulator is challenging both in mechanics and in control engineering, where many aspects need further study and more efforts. At conclusion, a summary of work done in this dissertation is given and some problems of interest are also brought forward for future research.