Hysteresis compensation control for reluctance actuator force using neural network

Reluctance actuator has a unique property of small volume, low current and can produce great force. So it is very suitable for high-precision and high acceleration control applications such as the next-generation semiconductor lithography equipment. However, the hysteresis characteristics of reluctance actuator cannot be ignored in high-precision control. One of the major challenges of reluctance actuators is the predictability of the force, which has a nonlinear relationship with respect to the current and position and is directly related to the final accuracy in the nanometer range. Therefore, it is necessary to study the control method for the reluctance force. This paper proposes two hysteresis compensation control configurations for the reluctance force using the multilayer neural network (MNN). The multilayer neural network is used as a learning machine of nonlinearity. The advantage and disadvantage of each method as well as their application conditions are investigated extensively through simulations. The simulations are conducted on the E/I Core reluctance actuator model and the results show that the proposed methods are effective in overcoming the hysteresis and promising in high-precision and high acceleration control applications.