Research Of High-Speed And High-Precision Positioning Problems For Macro/micro Driver In The Integrated Circuit Industry

With the development of integrated circuit photoetching, integrated circuit testing andpackaging and surface mount technology (SMT) in the integrated circuit industry, therequirements for the stroke, speed, acceleration and precision of positioning system are moreand more strict, which challenge the design of the high-speed and high-precision positioningplatform and control strategy. Due to the high-speed and high-precision positioning platformis widely used in the key equipments in the integrated circuit industry, the exploring of theresearch on the fundamental and essential problems and key technique of the long stroke andhigh precision positioning platform is of great important theoretical significance andapplication value.In this thesis, the object of study is the macro/micro driving positioning platform whichdriven by linear motor and smart material-based actuator. According to the production task'scharacteristic of the integrated circuit industry, as for the macro mechanism—the linear motorplatform, the iterative learning control strategy of trajectory tracking of the linear motorplatform and the high-speed and high-precision point-to-point positioning technique arestudied. As for the micro mechanism—the smart material-based actuator, because thehysteresis phenomenon exists in the whole working range of the actuator, which seriouslyinfluence the positioning accuracy of the system, the control strategy to mitigate hysteresisphenomenon is also studied. This study focuses mainly on the following aspects:1. As for the iterative learning control strategy of trajectory tracking of the linear motorplatform, the linear motor model is considered as a linear discrete-time system with parameteruncertainty, and the problem of robust monotonically convergent iterative control for a classof uncertain linear discrete-time systems with non-zero constant initial error is studied. Thelearning law under consideration is an anticipatory ILC (iterative learning control). Based ona simple quadratic performance function, a sufficient condition for robust monotonicconvergence of the proposed learning algorithm is presented in terms of linear matrixinequality. Finally, the simulation and experiment of the linear motor platform are given toshow the effectiveness of the proposed scheme.2. According to the characteristic of equipment's point-to-point repeat motion in the integrated circuit industry, a hybrid controller which combines a gain scheduled P controllerand A-type ILC controller is designed. The gain scheduled P controller is used to improve thesystem speed and suppress the non-repetitive disturbances, while the A-type ILC is used tocompensate for the effects of the repetitive disturbances. Then, the stability analysis is givenin the time domain. Finally, in the point-to-point motion experiments, the control algorithm isverified in the linear motor platform. The results show the superior performance of the hybridalgorithm over a standard PID controller.3. Using the backstepping technique, an adaptive inverse controller is designed to mitigatethe effects of hysteresis phenomenon for a class of uncertain dynamic non-linear systemspreceded by unknown hysteretic nonlinearities, where the hysteresis is described byPrandtl-Ishlinskii (P-I) model. First, the continuous P-I hysteresis model is decomposed into adiscrete P-I operator and a small-bounded error. Then, the inverse discrete P-I operator isconstructed to eliminate the hysteresis effects, and the bounded error is estimated by theadaptive controller. Next, two adaptive controllers are designed: the first controller can ensurethe closed-loop system is uniformly ultimately bounded; the second controller can ensure theclosed-loop system is globally stable. The simulation results demonstrate the effectiveness ofthe proposed scheme.4. An adaptive inverse control scheme is proposed for a class of uncertain nonlinearsystems preceded by unknown hysteretic nonlinearities which are described byKrasnosel'skii-Pokrovkii model. In order to overcome the difficulty of constructing analyticinversion of the hysteresis model, the pseudo-inversion is introduced to compensate thehysteresis nonlinearities and the error caused by the hysteresis model and pseudo-inversion isestimated online by the adaptive controller. To avoid the possible chattering caused by thesign function, a smooth adaptive controller with the hyperbolic tangent function is designed.The proposed controller ensures the closed-loop system is uniformly ultimately bounded. Thesimulation results show that the effectiveness of the proposed scheme.In the end, the research achievement is summarized and some considerable questions andthe direction of study on high-speed and high-precision Macro/Micro driving positioner arepresented.The dissertation is supported by the State Key Program of National Natural Science...