Study On Modeling And Control Technology Of Piezoelectric Actuated Micro-displacement Stage

Precision positioning technology is one of the key technologies on precision manufacturing, precision measuring and precision actuating. Based on the "Research on intelligent control methods for piezoelectric actuated nano-displacement stage" (Nature Science Fund Project of Zhejiang Province, No: 502021), we absorbed the original thoughts, new theories and technologies of correlative subjects to study some key technologies of piezoelectric actuated flexure micro-displacement stage by using theoretical analysis, computer simulation and experiment in this dissertation. The main work includes hysteresis modeling technology of piezoelectric actuator, designing and modeling technology of flexure hinge micro-displacement and residual vibration eliminating and steady state error controlling technology. The details were studied as follows:In chapter 1: The background and significant of the research was introduced firstly. Then, the current research situations and the deficiency on the key technology of the piezoelectric actuated flexure micro-displacement stage were expatiated. At last, the study contents, implementation, and novelty of the dissertation were proposed.In charter 2: The main parts of the piezoelectric actuated flexure micro-displacement stage were given out firstly. Then, the characteristic of multi-layer piezoelectric actuator was introduced, and it's static and dynamic hysteresis were studied by experiment. Thus, the hysteresis modeling technologies were studied, the Preisach hysteresis modeling technology and its modeling method were emphatically analyzed, and the two dimension linear interpolation and neural network were used to calculate Preisach function. At last, the merits and demerits of single parallel four bar mechanism and double parallel four bar mechanism researched.In chapter 3: Statistical learning theory, support vector machine and its regression arithmetic were studied, a new method was proposed to calculate Preisach function by using the least squares support vector machine regression (LS-SVMR), and a hybrid hysteresis model was build based on Preisach and least squares support vector machine regression. In order to testify the availability of proposed hybrid hysteresis model, contrast experiment were carried out among two dimension linear interpolations Preisach model, neural network Preisach model and the proposed LS-SVMR Preisach hybrid model, the experiment results revealed the LS-SVMR Preisach hybrid model to be a most accurate one.In chapter 4: Firstly, the relationship among force, displacement and stress of single parallel four bar mechanism and double parallel four bar mechanism was compared, a compound parallel four bar mechanism was proposed by using the merits and discarding the demerits of two classical parallel four bar mechanism, the proposed compound parallel four bar mechanism has the merits of non-parasite displacement, little concentrated stress and big object displacement, finite element simulation proved it to be true. Then, an analytical dynamic model of flexure compound parallel four bar mechanism was studied based on Lagrange Equation, finite element simulation and experiment proved the analytical dynamic model to be correct. At last, a two dimension flexure hinge micro-displacement stage was design by using the principal of proposed compound parallel four bar mechanism, and the effect on precision of machining error was analyzed.In chapter 5: Firstly, the basic conception of input shaping controller was introduced, and its design method was analyzed. According to the characteristic of piezoelectric actuated flexure micro-displacement stage, a four impulse input shaping controller was design to eliminate residual vibration. Then, based on four impulse input shaping controller and LS-SVMR Preisach hybrid model, a double feedforward controller was design to apply for open control, LS-SVMR Preisach hybrid model was used to quicken system response, and four impulse input shaping controller was used to eliminate residual vibration. Lastly, for satisfying the requirement of high precision micro-positioning, a double feedforward PID parameter neural network self-turning time separate closed loop hybrid controller was proposed, this hybrid controller can eliminate residual vibration and reduce steady state error to enhance positioning precision of piezoelectric actuated flexure micro-displacement stage.In chapter 6: A measure and control system of piezoelectric actuated flexure micro-displacement stage was build up by using LabVIEW and MATLAB, and the characteristic experiment of stage was carried out. Then, effect factors on positioning precision of micro-displacement stage were analyzed, and some notice were proposed to enhance the positioning precision of piezoelectric actuated flexure micro-displacement stage.In chapter 7: All the work in this dissertation was summed up, and the future research works have been forecast.