Robust Speed Controller Design For Permanent-Magnet Linear Synchronous Motor

Based on the project supported by the National Natural Science Foundation of China“Trajectory planning and precision contour control for the direct XY platform based onthe velocity field”(No.51175349), aiming at the problem of model parameter uncertaintiesand external disturbance,H_∞robust control theory is combined with other controlmethods, two synthetic control strategies of H_∞/LTRandμ H∞are developed toimprove the robust performance of linear servo system.H_∞robust control has good disturbance resistance and linear servo systemuncertainty attenuation performances, which can make the system has good stability, butignores the system robust performance. LTR is a simple and effective method which cansolve the designing of multivariable robust feedback system. According to thecharacteristic of controlled object, predetermined system to achieve the performance(targettransfer function), so that the open loop transfer function in series from the designingfeedback controlled and object model are recovered to target transfer function. Hence, anobserver-based H_∞/LTRsynthetic control method is developed in this thesis. The targetloop designed in this method, that was, under the assumption that the states of system wereavailable for feedback, a law of robustH_∞control was designed to satisfy theperformance specifications. And then, the recovery designed in this method, that was, thetransfer function of the target loop were recovered by introducing the designed stateobserver into system.For the system with structured uncertainties, designingH_∞controller is conservative,while μ control theory can just be able to make up for the shortcoming of theH∞controller, but it has too little calculation conditions and slow convergence speed. So usingμ method to assistH_∞robust theory without orthotropic assumption, aμ H∞speedcontroller is designed in this thesis. This method starts with the upper bound of μ valueand condenses the singular value throughH_∞robust control method, so then a controllerwith robust stability and performance is obtained quickly. Simulation results are analyzed that the two proposed controllers, satisfying strongrobustness for model parameter uncertainties and restraint disturbance, can meet therequirement of high precision control in numerically controlled machine tool.