Research On Vibration Control Of Planar 3-RRR Flexible Parallel Manipulator Based On Acceleration Feedback

Due to its lightweight,high positioning accuracy and fast running speed,the planar 3-RRR flexible parallel manipulator is widely used in the fields of precision positioning,material handling,and macro-micro combination.However,the continuous vibration generated by moving it to the singular area at a certain speed makes an impact on the system positioning accuracy and leads to the fatigue damage.The acceleration sensor is easy to install with no impact on the structural characteristics.Acceleration feedback characterizes with increasing damping and enhancing robustness for the system.Therefore,for the vibration control problem of the planar 3-RRR flexible parallel manipulator,the control algorithms based on acceleration feedback and vibration control experiments were studied in this paper.Firstly,based on the inverse kinematics and the interpolation,the discrete computation formulas of the circular and sinusoidal trajectory were deduced to realize the motion planning of the parallel manipulator.The zero degrees attitude angle workspace and the distribution regularities of inverse Jacobian singularity of the parallel manipulator were respectively obtained by the numerical method and Jacobian matrix method.The analyzing condition of the inverse singularity was given for the self-excited vibration phenomenon.Based on the Lagrange method,assumed modal method and independent coordinate method,a rigid-flexible coupling dynamic model reflecting the relationship between the multi-input and multi-output of the planar 3-RRR flexible parallel manipulator system was established.Secondly,according to the summary of experimental data,the types and the induced mechanism of vibration were analyzed.The cause of self-excited vibration of the planar 3-RRR flexible parallel robot was also found.With the analysis of self-excited vibration characteristics,the feasibility of using acceleration feedback control was proved,and the acceleration feedback proportional control algorithm and the RBF neural network adaptive control algorithm based on acceleration feedback were studied.Finally,the detection system of acceleration signal was built on the experimental platform of the planar 3-RRR flexible parallel manipulator.The acceleration feedback signal after filtering and phase-shifting was applied to the vibration control experiments.The experimental results show: when the period of phase-shifting is accurate and the acceleration feedback gain is set up properly,the acceleration feedback proportional control algorithm can realize the self-excited vibration control at different quadrants,and guarantee the position accuracy after the vibration stops.When the initial value of the parameters of the neural network is adjusted well,the RBF neural network adaptive control algorithm based on acceleration feedback can also effectively suppress the self-excited vibration of the parallel manipulator.However,due to time consuming learning for the optimal parameters,its controlling performance is worse than the former.