动学效应的动定位型在线识算.PDF

28 11 Vol. 28 No. 11 2011 11 Control Theory Applications Nov. 2011 :2011 , , , (1. , 200215; 2. ( ) , 200125) : , . , (unscented Kalman filters, UKF) . , . , . , . : : U661.33 : A Online dynamic-model recognition for dynamically positioning vessel by dynamic effect NI Fei , ZHAO Yan-zheng , YE Jun , ZHU Ting (1. Shanghai Jiaotong University, School of Mechanical Engineering, Shanghai 200215, China; 2. Shanghai Zhenhua Heavy Industry Co., Ltd, Shanghai 200125, China) Abstract: Due to the nonlinearity in a vessel model, and the immeasurable disturbance force as well as thruster forces, it is difficult to recognize the vessel model online. An online vessel-model-recognition algorithm based on unscented Kalman filters(UKF) and dynamic effect is proposed. By solving the vessel model associated with external loading forces during dynamical positioning, this algorithm continuously generates the dynamic effect which approaches the sensor feedback. Based on this principle, an effective and adaptive position control can be performed by the nominal vessel model, external and thruster forces. Simulation results in position control validate the effectiveness of the proposed algorithm. Key words: vessel dynamic positioning; online vessel model recognition; unsent Kalman filter; dynamic effect 1 (Introduction) . , , . . , , ,