Application of Support Vector Machines to the Tracking Control of Piezoelectric Actuators
暂无分享,去创建一个
In this paper, a hybrid approach associated the Preisach concept with support vector machines (SVM) is brought forward to identify and predict the nonlinear behavior of piezoelectric actuators (PA). Preisach concept is used to construct mesh nodes in the Preisach plane and determine the final output displacement of PA. SVM is trained based on the mesh nodes and provides favorable generalization ability in the Preisach plane. Experimental results validate the proposed method.
[1] Vladimir Cherkassky,et al. The Nature Of Statistical Learning Theory , 1997, IEEE Trans. Neural Networks.
[2] Dongwoo Song,et al. Modeling of piezo actuator’s nonlinear and frequency dependent dynamics , 1999 .
[3] R. Ben Mrad,et al. A model for voltage-to-displacement dynamics in piezoceramic actuators subject to dynamic-voltage excitations , 2002 .
[4] Musa Jouaneh,et al. Modeling hysteresis in piezoceramic actuators , 1995 .