Modeling And Control For Hysteresis In Precision Positioning System

With the rapid development of micro/nano technology, piezoelectric ceramics, themetal alloy of shape memory, etc. as representatives with the characteristics ofintelligent materials for high precision, fast response and great driving force arewidely applied in many precision fields. However, the hysteresis nonlinearity in theseintelligent materials not only can make the control precision of the system reduced,but also may cause system unstable. Hysteresis is a kind of unconventionalnon-smooth nonlinearity, and its complexity mainly displays in multi-valued mappingand memory. In order to eliminate the effects of hysteresis nonlinearity of the system,it is needed to find a model to describe the behavior of hysteresis and add the controlscheme into system to obtain the expected control effect.The dissertation is organized as follows:1. The dynamic hysteresis nonlinear model is established with putting forward amodified hysteresis operator describing characteristics of hysteresis.In numerous hysteresis models, Preisach model is the most widely used one.Therefore, a dynamic hysteresis model based on a Preisach-type operator and a linearsystem is established in this paper. This model not merely can describe hystereticcharacteristics well, but also has a simple structure and few parameters to identify.2. An improved particle swarm algorithm is proposed to identify parameters ofmodel.An improved Particle Swarm Optimization (PSO) algorithm is put forward. PSOalgorithm is a kind of emerging evolutionary computation technology based on swarmintelligence theory. In order to enhance the effect of optimization, an improved PSOalgorithm is proposed with a velocity updating formula. The velocity has two weightsvarying with the optimized results and makes the optimized process had fasterconvergence and better precision.3. A discrete sliding mode control scheme is employed in this paper, andcompared with other control schemes. It is difficult to get the expected control effect from the conventional controlscheme for hysteresis system, so in order to reduce the influence of hysteresis in thesystem, a sliding mode control scheme is designed based on a hysteretic inverseoperator. Firstly, it is need to introduce a hysteretic inverse operator to compensate theinfluence of hysteresis. Then the sliding mode controller is designed. Finally, someexperiments are performed by using software. Compared with experimental results ofinverse compensation PID algorithm and open loop control algorithm, the slidingmode combining with the inverse compensation control strategy for the hysteresisnonlinearity system has higher control accuracy.