Adaptive Sliding Mode Control And Feedforward Parameter Self-tuning Of Wire Bonding Machine XY Motion Platform

As the size of the chip becomes smaller and smaller and the distance between the solder joints is increasingly shortened,which puts forward higher standards for the positioning accuracy and robustness of the motion platform in the chip packaging equipment.However,the precision XY motion platform is a nonlinear and strongly coupled system.Its mathematical model is often uncertain and the extent to which it is affected by interference is also unknown.When the system has internal variable perturbation or is affected by external interference,the traditional sliding mode variable structure controller is not robust enough to obtain satisfactory performance.The adaptive sliding mode variable structure control introduces an adaptive control method,which greatly improves the system's ability in anti-interference and resistance to variable perturbation.Its good robustness and certain adaptability makes it more suitable for use in control systems based on permanent magnet linear motors.Feedforward control can further improve the performance requirements of the system without destroying the stability of the original system.At the same time,the differential evolution algorithm is used to offline auto-tune the parameters of the feedforward controller,which reduces time cost and improves work efficiency.This thesis is based on the sliding mode variable structure control theory and studies the position closed-loop two-degree-of-freedom control strategy of the permanent magnet linear motor in the XY motion platform.The main research contents are as follows:(1)This thesis investigates the development status of wire bonding machines and the precision XY motion platforms,and in-depth study of the "feedforward + sliding mode control" two-degree-of-freedom control strategy of the XY motion platform.Then combined with the complex wire bonding process flow of the wire bonding machine,it also designs a three-order S-shaped trajectory plan,which lays foundation for the subsequent experimental application.(2)The working principle of permanent magnet linear motor and its nonlinear effects are explained,and the mathematical model of the electric motor is derived through the motor voltage equivalent circuit model and the electromechanical dynamics formula.On the basis of sliding mode variable structure control theory,the causes of chattering and various solutions to weaken chattering are discussed.Combined with the mathematical model of the motor,a sliding mode variable structure controller based on the exponential reaching law is proposed.Finally this control strategy is analyzed experimentally using Matlab software to verify its tracking performance,anti-interference ability and anti-parameter perturbation ability,and provide a comparative analysis object for the subsequent design of a more optimized sliding mode variable structure controller.(3)Based on the analysis of the advantages and disadvantages of the constant velocity approach law,the exponential approach law and the power-order approach law,this thesis designs the power exponential reaching law.At the same time,in view of the shortcomings of the sliding mode variable structure control system based on the exponential reaching law,such as insufficient anti-interference performance,large chattering,and this system is greatly affected by the perturbation of internal parameters,a sliding mode variable structure controller based on the power exponential reaching law is proposed.Finally experimental results show that compared with the sliding mode variable structure controller based on the exponential reaching law,this system chattering phenomenon is reduced and has better anti-interference and parameter perturbation performance.(4)When the internal parameters of the system are perturbed,the sliding mode variable structure controller based on the power exponential reaching law has a larger tracking error.And the switching gain of this reaching law often takes a larger value due to the compensation interference value,so that this system has a greater chattering phenomenon in the steady state.Therefore,in order to reduce the switching gain of the reaching law and improve the adaptive ability of this system,an adaptive method and iterative learning are introduced to compensate the interference items.Then an adaptive sliding mode variable structure control strategy based on the “power exponential reaching law and iterative learning” is proposed.Iterative learning compensates for the periodic interference of the system and adaptive method compensates for its non-periodic interference to ensure that the system has a certain adaptability.The experiment results show that iterative learning and adaptive sliding mode variable structure control algorithm are combined to enhance the robustness of the system and effectively weakens system chattering phenomenon.(5)The system adopting adaptive sliding mode variable structure control strategy based on the “power exponential reaching law and iterative learning” has a corresponding improvement in anti-interference and parameter perturbation ability and its chattering phenomenon has also been further weakened,but this system has a slight lack of position tracking performance.In order to further improve the syatem positioning accuracy,feedforward control is added to the original control strategy.And the differential evolution algorithm is used to auto-tune the speed and acceleration parameters of the feedforward controller offline.The differential evolution algorithm is used to find the optimal feedforward controller parameters,which reduces the cost of manual parameter setting by manpower and improves work efficiency.