Research On Model Identification And Compensation For Cogging Force Of Iron-core Linear Motor

Represented by the integrated circuit, the ultra-precision machining technology haspenetrated into each field of daily life with the flourishing development of the epoch ofelectronic and information. As one of the primary driving elements for high precisionservo systems, linear motor plays a critical role in manufacturing equipment ofintegrated circuit, and seizes a significant position in both of the industrial productionand scientific research. However, in consequence of the end effect of magnet, coggingforce impacts the performance of linear motor in its process of movement, which limitsthe application of the linear motor in the field of high precision servo. In order toimprove the control precision of linear motor, compensation of the cogging force isindispensable.After the background and importance of development of lithography and scanningequipment are elaborated, which reveals the significance of linear motor’s cogging forcecompensation, the overall plan is provided, including preceding model indentificationand.following compensation controller design. For this purpose, firstly, the coggingforce generating principle is analysised, and the model indentification scheme isdetermined. Least square method is used for parameter estimation for the model ofcogging force based on motor postion. Subsequently, considering the systemcharacteristic, a modified internal model controller and iterative learning controller aredesigned, which contain PID control structure both but separately acts in the currentcontrol loop and the position control loop. Comparation with the traditional controlmethod through simulation ceritifies the method mentioned above.In the next place, taken consideration of the system’s actual demand, the limitationof original global iteration algorithm may lead to performance degradation in certainsituation, although the algorithm’s convergence has been proved. For this reason,aiming at the scanning track requirement in reality, the iterative learning control isoptimized in subsection. The result of the simulation testifies this method’s feasibility.Finally, the construction of both the hardware and software for the experimentalplatform is presented, including the compensators’ implementation procedure based onmotion control board card with DSP and FPGA. The experimental result is achieved inthe end, which verifies the scheme’s effectiveness.