Analysis And Method Research Of Compensation For Force Ripple Of Iron Linear Motor

In current years,the precision positioning stage is widely applied in various walks of life,for example IC manufacturer,space and aeronautics industry,precision machining industry and so on.In the all motor drive devices,the permanent magnet linear motor is widely applied in the field of high speed and high precision motion control.Because there aren't mechanical transmission to realize the conversion from rotational motion to linear motion in linear motors,the influence of motor system's nonlinearity and friction disturbance are reduced a lot.But meanwhile,it brings the matter of more model uncertainties and stronger external disturbance,for example the force ripple of linear motor.Therefore it's crucial to reduce the influence of above mentioned problem,whether from the aspect of proper structural design or effective system control method.The paper's research background is the lithography machine's dual-stage system.Methods of solving the problem of force ripple are studied in depth in this dissertation.Firstly,the paper analyzed the causes of linear motor force ripple from the motor's structure level theoretically.The model of controlled object was acquired by analyzing results of a lot of experiment of frequency sweep.And the feedb ack control system was designed,which proposes solid foundation for follow-up work.Secondly,the data acquisition experiment was did in existing platform.And the paper presented the cogging force's and ripple's identification models based on the RBF neural network.Besides the identification models were used in compensation experiment by table lookup.And the experiment results proved it's necessary to design other more effective methods of compensation for force ripple.Thirdly,the iteration learning control method was studied by the way of theoretical analysis,simulation and experiment research in the paper.The experiement results proved the effectiveness of ability of iteration learn ing compensation method in suppressing periodic force ripple disturbance.However it can't reduce the impact of force ripple that exhibits nonperiodicity.Finally,the method of disturbance observer was studied in depth and an improved disturbance observer was proposed to solve problems of force ripple.And the experiment results proved the effectiveness and the excellent nature of compensation strategy based on improved disturbance observer.Besides the paper proposed a compensation strategy combining the improved disturbance observer and iteration learning control method,which can increase the system following error and decrease the iteration periods at the same time.