Research On Active And Passive Vibration Control Method Of The Six Degrees Of Freedom Vibration Isolation Platform Experiment System

Stewart platform with six degrees of freedom which used as a vibration reduction mechanism that was applied in spacecraft precision measuring, the stand or fall of its damping performance has become the main topics in orbit measurement techniques. Scholars at home or abroad recognized the Stewart parallel mechanism with six degrees of freedom as the most effective damping device currently. The most prominent advantage of Stewart were: high stiffness, accurate positioning, strong carrying capacity, good dynamic characteristics, et al. In consideration of the safety and reliability in the real application of Stewart, the passive vibration control method was widely used and immortal achievements have been obtained. But for years the active control of vibration used in Stewart platform has become one of the important topics, and the researchers have made some achievements. In order to consider reliability and performance of vibration reduction of Stewart, the active control method used with piezoelectric actuators were applied to control the performance of the Stewart platform, and the active control algorithm was established, the method can attenuation the vibration effectively.The paper was based on project(9140A2011QT48) weapons and equipment pre-research fund” high resolution general design method of vibration isolation” and the author established Stewart platform with six degrees of freedom based on piezoelectric actuators as active components, the numerical simulation optimization and the experiment were carried out. By studying the hysteresis characteristics of piezoelectric actuators and establish the corresponding control unit and then applied to the optimized passive vibration isolation Stewart platform, then the corresponding control law was established based on chaotic characteristics through the actuator acceleration sensor data collected by discriminant. This topic research results not only has important theoretical and application value to the spacecraft is in orbit vibration control, but also has important reference value for further research in the future. The main innovation points can be summarized as:(1)Because of the hysteresis characteristics of the piezoelectric actuation devices in the application of active vibration control of Stewart, the RBF neural network, FLS-SVM and the combined method was applied to calculate the weighted values of the Preisach hysteresis model, the training sample data were from the FORCs curve. According to the conventional discrete Preisach hysteresis based method to determine the initial delay unit weighted value as the output of hysteresis unit and voltage as the input in the α-β plane. Combined with the quadratic polynomial interpolation method to establish the final piezoelectric hysteresis model. Through the error analysis of the inner and outer ring of the Hysteresis loop, the results show that the combined method of RBF NN and FLS-SVM was the best one and the model can well reflect the actual working status of piezoelectric actuators.(2)Considering of external and internal uncertainty uncertainty in the process of modeling and control of piezoelectric actuator hysteresis model, H∞ robust intelligent control method is introduced, the methods including the mixed sensitivity design and optimization problem and the rules of choosing weighted function. To determine the generalized controlled object of the robust intelligent controller, the mathematical model of controlled object was established and in order to get correct effective generalized controlled object, chaos optimization theory was adopted to global optimize weight function, thus the robust intelligent controller was obtained. Through the contrast, the simulation results show that the initial moment overshoot volume is small, and the oscillation is not obvious, can quickly achieve stability based on the optimized H∞ robust intelligent control method and in the noisy input, the system output changes gently, smaller overshoot and transition time is short, system can be quickly restored to the stable value, has a good inhibitory effect to exert the interference of noise. And the test verified the effectiveness of the model.(3)In order to achieve good damping performance of Stewart platform, the mathematical model of dynamics and kinematics were established, then multi-objective optimization function of the system was established based on the spring stiffness and damping coefficient, within the scope of the design variables, optimization constraints were set and also the corresponding weights. According to the established simulation optimization of conditions, the final passive vibration reduction model was determined. And dynamics simulation was carried out with this model, the natural frequency and frequency response characteristic of the system were obtained and displacement, velocity and acceleration of top platform were calculated under the disturbance of particular. Through experiment contrast, the result show that after optimization, the vibration isolation platform has better damping performance, and can attenuate of the bottom platform vibration disturbance, it not only provided a design basis for the active control, also can guarantee a certain vibration reduction performance if the active control is failure.(4)To solve the problem of feedback control of platform, the acceleration test data of piezoelectric actuators was chaos discriminated with chaos theory, each actuator correlation dimension and saturated correlation dimension was got, and maximum Lyapunov index was obtained to verify the chaos characteristics of time series, also the Kolmogorov entropy of acceleration time series was identified, thus the chaos degree of the system was obtained. According to the results, found that the feedback and control problem is nonlinear control problem, which laid a foundation for the establishment of the subsequent control method.(5)In order to control piezoelectric actuator fast and stable, and considering the chaos characteristics of the control signal, the switch control method combined with RBF neural network was introduced as the active control technology. Then the joint Simulink simulation platform was set up with the dynamics virtual prototype. The simulation was carried out at the resonant frequency and the output characteristic was obtained. Results show that the parallel Stewart platform with six degrees of freedom under active vibration control can effectively damping vibration amplitude. And RCP hardware in the loop simulation test results verified the effectiveness of active vibration control method.