Study On Long Travel Nano-scale Piezoelectric Actuator Control System

The piezoelectric is a kind of information function ceramic material which can make transformation between mechanical energy and electric energy. Due to their characteristics of high displacement resolution, high stiffness, rapid response and noiseless, piezoelectric actuators have been widely used in high-precision positioning devices and tracking systems, such as micro-mechanics, micro-electronics, precision processing, biomedical, robots, and aerospace. However, the piezoelectric actuators suffer from the inherent small output displacement, hysteresis, creep and nonlinearity.These characteristics affect the performance of precision positioning systems severely.This thesis focuses on the research of the characteristics of piezoelectric actuators including hysteresis, creep and nonlinearity. The flexible hinge micro displacement amplification mechanism is designed to amplify the displacement of piezoelectric actuators, the creep models of piezoelectric actuators are established and the creep errors are predicted and corrected by neural network. The main research contents are as follows:Firstly, the inherent characteristics of piezoelectric actuators are studied. Based on the polarization theory, the produce mechanism of hysteresis is analyzed. The characteristics of creep is studied based on the friction between the crystal lattice.Meanwhile, the characteristics of nonlinearity and the repetitive of the displacement are researched and the characteristics of piezoelectric actuators are verified by some experiments.Secondly, a flexible hinge micro-displacement amplification mechanism based on two levers is designed. The piezoelectric ceramic output amplified up to 0.3mm. The mechanism resolution is about 10 nm. The performance of the amplification mechanism is studied and analyzed through theoretical calculation, ANSYS simulation and experiments. The comparison of the creep value is proposed by experiments.Finally, the creep models of piezoelectric actuators are established by neuralnetwork. The connection between the creep value and the time, voltages are described by the creep models. The neural network is optimized by genetic algorithm to improve the prediction accuracy of the neural network. The reverse voltage method is adapted to corrected the creep errors of piezoelectric actuators. The theories as stated earlier are verified by some experiments.Through theoretical research and experimental verification, the research work of this paper has important theoretical significance and practical application value for the high precision positioning of piezoelectric ceramics.