Research On Piezoelectric FTS And RBF Control For Precision Turning

With the rapid development of optical instruments,medical equipment and aviation electronic technology,the shape requirements of parts with asymmetric cross-section and micro-structure surface are becoming more and more complex,and the accuracy requirements are becoming higher and higher.This requires higher processing accuracy,displacement resolution and frequency response of fast servo tool holder(FTS)system in order to improve the processing efficiency and accuracy in the field of precision machining.However,most of the FTS systems designed at present are difficult to meet the requirements of high frequency,high precision and high stability.Secondly,the spindle speed of the machine tool itself is low,which seriously affects the development of FTS ultra-precision turning technology.How to improve the processing stroke,positioning accuracy and high frequency response of ultra-precision turning FTS system is the key to improve the processing accuracy and efficiency in the field of ultra-precision turning.In this paper,a piezoelectric FTS system based on two-stage lever displacement amplifier mechanism is designed,and a hysteresis model of piezoelectric ceramic actuator is established.A controller based on RBF neural network(RBF neural network)and PID combined control algorithm is designed to improve the processing accuracy of fast servo tool holder.And processing efficiency.Its main contents are as follows:(1)Structure design and optimization of piezoelectric FTS systemA servo tool holder model of two-stage lever amplification mechanism is developed and designed.The influence of structural parameters of flexible hinge on the performance of servo tool holder is analyzed.The magnified displacement,natural frequency and stiffness of servo tool holder are studied in detail through theoretical calculation and simulation research,and the mathematical model of FTS system is established.(2)Hysteretic Nonlinear Modeling of Piezoelectric Ceramic ActuatorsThe output characteristics of piezoelectric actuators are tested,and the hysteresis nonlinearity mechanism of piezoelectric ceramics is analyzed.The hysteresis characteristics of piezoelectric ceramics are studied by establishing a mathematical hysteresis model.RBF neural network algorithm is used to optimize the hysteresis model of piezoelectric actuator to reduce the influence of hysteresis characteristics of piezoelectric ceramics on the system's non-linear output.(3)Design of FTS System ControllerA controller based on RBF neural network and PID combined control algorithm is designed to improve the output displacement accuracy of servo tool holder,complete the control output of the controller to the turning vibration of servo tool holder system,and improve the response time and processing efficiency of the system.The simulation results show that the maximum response time of the system is 0.08 s,which is 0.04 s faster than that of the PID controller,and its optimization ability is 30.8%.At the same time,the output of the system is relatively stable,and the maximum displacement of turning vibration is reduced from 1.37 to 0.02 um.The experimental results show that the maximum error between the actual output displacement and the reference displacement of the platform is 0.29 micron,and the control accuracy is better.The performance of the platform is obviously improved by comparison.At the same time,the reliability and stability in the specific control process are considered comprehensively.