Adaptive Model Following Control For Test Turntable

Test turntable is a kind of important inertial navigation test equipment, it is essential to further improve its performance for the improvement of inertial devices and system performance. In general, there are the following main factors which influence the improvement of system performance: (1) The error of the nominal model of the controlled object and the actual situation, which is introduced mainly by the parametric uncertainties and high frequency unmodeled dynamics of the controlled object; (2) The diversification of system parameters, the diversification of load caused by unbalanced torque plays a major role during the operation; (3) Dynamic torque disturbances, which include load imbalance and friction torque.The actual test turntable is studied in this thesis. For parameteric uncertainties and time-variant characteristics of the controlled object in the test turntable, and considering the disturbance torque and noise, the following three kinds of adaptive model following control (AMFC) scheme are respectively proposed.Firstly, an adaptive model following control method is proposed for suppressing the load torque and friction torque disturbances. An absolutely ideal model is selected as reference model in this method, which can assure the system output completely track the input. Taking into account the realizability of adaptive control method in engineering, a simple local adaptive control method is presented. The method is characterized by the adaptive model following control which is designed only for motor and its drive system, but still the classical control is introduced as position feedback loop, which greatly simplifies the adaptive control algorithm and provides the basis for the implementation of engineering. Simulation results show that the method can also inhibit the adverse effects which are caused by load changes and torque disturbances to the system.Secondly, considering the influences of noise on the performance of turntable control system, an adaptive model following control method is proposed. The low pass filter is chosen as reference model in this method, which can inhibit the noise from influencing the servo performance of the system. Simulation results show that this method can suppress interference, without affecting the control performance of the system, and improve the tracking performance of test turntable.Thirdly, with regard to the parameteric uncertainties and time-variant characteristics of the controlled object in the test turntable, an adaptive compound control algorithm is proposed to improve the servo performance of system. An absolutely ideal model is chosen as reference model in this method, which not only can simplify the adaptive control algorithm, but improve the tracking performance of system. Simulation results show that the method can assure the capability of rapid tracking the input for high-precision turntable system, meantime effectively suppress the influence of the time-variant parameters on the system performance.This dissertation gives several kinds of adaptive control methods. The introduce of state filters can avoid the dependence of the control algorithm on the all state variables of the actual system. The adaptive control law can be integrated by the input and output of system, which establishes the foundation for the practical application in turntable control.Finally, for the realization of the adaptive model following control in test turntable, based on PC/104, this dissertation gives the digital control hardware system, angular measurement and feedback system, and achieves the control algorithm of the software programming by the C language based on DOS operating system.