Fractional-order Iterative Learning Control For Permanent Magnet Linear Synchronous Motor Based On MEEMD Algorithm

Permanent magnet linear synchronous motor(PMLSM)has been widely used in high-precision CNC machine tools and industrial robots due to its characteristics such as large thrust,fast response and high precision.Iterative learning control(ILC)can effectively suppress repetitive disturbances,and theoretically can achieve complete tracking in a limited interval,which is very suitable for servo control systems that perform repetitive tasks.However,in practical engineering applications,it is tough to guarantee the initial conditions of each iteration to be consistent and the external disturbances to be repetitive,which will cause the tracking error accumulation of each iteration and then affect the convergence speed and tracking accuracy of the ILC system.To improve the tracking performance of the PMLSM iterative learning control system when performing periodic tasks,the fractional-order iterative learning(FO-ILC)controller based on the modified ensemble empirical mode decomposition(MEEMD)algorithm is designed in this thesis.Firstly,the structure and working principle of PMLSM are introduced,and the disturbance factors affecting the servo performance of PMLSM are analyzed,the mathematical model of PMLSM is established after simplified under the condition of magnetic field orientation,which lays the foundation for the design of the controller.Secondly,in order to improve the displacement tracking accuracy and dynamic tracking performance of the PMLSM servo system,fractional-order calculus theory is used to improve the first-order PD-type ILC controller,and then PD~?-type FO-ILC controller is designed.The empirical mode decomposition(EMD)algorithm is used to decompose the tracking error generated in the FO-ILC process,and the divergent components are filtered and eliminated to improve the learning performance.However,when the tracking error contains measurement noise,modal mixing will occur if the EMD algorithm continues to be used for decomposition,and the accuracy of the decomposition will be affected.In order to avoid modal mixing,the FO-ILC controller based on MEEMD algorithm is designed.The MEEMD algorithm is used instead of EMD algorithm to decompose the tracking error with measurement noise.First of all,the complementary ensemble empirical mode decomposition(CEEMD)algorithm is used to decompose the measurement noise.Next,the appropriate permutation entropy(PE)threshold is set to filter it out,and then the EMD algorithm is used to decompose the remaining effective tracking error.The MEEMD algorithm can ensure that the components of the decomposition are as close as possible to the actual signal.After the measurement noise and divergent components filtered and eliminated,the learning components can be extracted out and reconstructed as the FO-ILC input signal,which can avoid the accumulation of tracking errors in the ILC process,and improve the convergence speed and tracking accuracy of the system.Finally,the PMLSM control system model is built under the Matlab/Simulink simulation environment,and the FO-ILC program,EMD algorithm and MEEMD algorithm program are written to verify the simulation results.The simulation results are compared and analyzed to verify the effectiveness of the proposed method.