Design of Permanent Magnet Actuator for Reducing a Mean Deviation of Thrust in Ventilation by Using Kriging Model

Abstract

Recently the need for ventilation has increased with the diversification of medical appliances. An actuator for optimal ventilation has to satisfy performance requirements such as low mean deviation of thrust, high thrust, and smaller size. The performance of actuator in ventilation is closely related to patient`s treatment. Improvement of the performance of actuator is related influence on respiratory controllability of ventilation. In the optimization of the actuator, the use of design of experiments (DOE) is established [1]. Generally in RSM, 2nd polynomial regression (PR) model is used for fitting the data. However the model is not adequate to approximate a nonlinear function due to the low-order PR. Thus the conventional optimization design is not profitable for multi-purpose function. For reducing these drawbacks, Kriging model using stochastic processes is applied to the global optimization [2]. This paper shows the optimal design process of actuator in ventilation. Also, the mean deviation of thrust is successfully minimized considering its average thrust and volume using by DOE and Kriging model.