基于改进灰狼算法的电动缸摩擦辨识及补偿OA北大核心CSTPCD

Nonlinear friction for motorized cylinders is difficult to identify and adversely affects the tracking accuracy of the posi-tion of the six-degree-of-freedom Electrically Driven Stewart Platform(EDSP),an improved Gray Wolf Optimization algorithm in-troducing Levy flight and Cauchy variation(LCGWO)with Levy flight and Cauchy variation is proposed to identify the friction parameters of the electric cylinder,and a feed-forward compensation method based on the friction model is designed.Firstly,a mathematical model of EDSP considering the nonlinear friction of the electric cylinder is developed and the friction is described using the Lugre friction model.Then,considering that the nonlinear friction of the electric cylinder is difficult to identify,an LCG-WO is proposed to identify the static and dynamic parameters in the friction model at the same time,and compared and analyzed with the general identification method.Finally,a feed forward compensation algorithm based on the Lugre friction model is de-signed and experimentally verified.The experimental results show that compared with other algorithms,LCGWO has better conver-gence speed and prediction accuracy;and the friction compensation of the electric cylinder by the friction feed-forward compensa-tion method based on the friction model can effectively improve the platform's position tracking accuracy.