Research On Modeling And Control For Long Range And High Precision Dual-stage Actuator

The requirement and challenge for the current drive feed system is to achieve long range an d high precision.Although the primary actuator constructed by traditional motors can provide long range,slow response speed and low precision hold back its role in high-precision positioning system.The secondary actuator based on smart materials possesses the apparent advantages for its fast response speed as well as high precision,but the limited range it provides brings about some imperfection.Accordingly,appropriate control schemes should be proposed for dual stage actuator to combine the merits of each actuator,thus,achieving long range and high precision positioning.Main contents and achievements are as follow:(1)Considering a class of nonlinear system with unmatched uncertainties,the radial basis function neural network(RBFNN)is adopted for the online approximation for the non-smooth and multi-valued mapping hysteresis described by Bouc-Wen model,which transforms time-varying hysteresis into the learning of weight matrices.The weight matrix updated law is derived by Lyapunov stability theory.Then a multi-layer sliding mode controller is developed to solve the problem of unmatched uncertainties and the tracking error can be converged into the bound layer designed in advance.(2)Considering a class of nonlinear plants preceded by backlash-like hysteresis,a novel adaptive output feedback control is proposed in this paper based on the Barrier Lyapunov Function(BLF)and RBFNN.Firstly,The BLF is given to overcome the difficulty of full state constraints under the premise of unavailable system state.Then the RBFNN is adopted for online approximation of disturbance-like term separated from backlash-like hysteresis,which removes the effect of hysteresis without any assumption on the bound of disturbance-like term.Finally,the stability of the closed loop system can be guaranteed by Lyapunov theorem and the simulation results validate the effectiveness of the developed control approach.(3)Considering a kind of dual-stage actuator modeled by mass-spring-damping structure,decoupling control strategy is adopted,where primary actuator tracks the reference signal and secondary actuator compensates the tracking error from primary actuator.As long as the tracking error stays in the scope of secondary actuator's travel range,then the output displacement of dual-stage actuator can be stable at the target value.A proximate time-optimal controller is designed for primary actuator in order to have a fast positioning speed.Based on model predictive control,a discrete sliding mode controller,having good robustness for matched or unmatched uncertainties,is proposed to eliminate the chattering in the control signal.Simulation results validate the effectiveness of the developed control approach.