个人信息Personal Information
副教授
博士生导师
硕士生导师
性别:男
毕业院校:大连理工大学
学位:博士
所在单位:船舶工程学院
办公地点:综合实验2号楼407房间
电子邮箱:wangwenhua@dlut.edu.cn
A novel adaptive fuzzy PID controller based on piecewise PID controller for dynamic positioning of sandglass-type FDPSO
点击次数:
论文类型:期刊论文
发表时间:2019-09-01
发表刊物:JOURNAL OF MARINE SCIENCE AND TECHNOLOGY
收录刊物:SCIE
卷号:24
期号:3
页面范围:720-737
ISSN号:0948-4280
关键字:Sandglass-type FDPSO; Dynamic positioning; Adaptive fuzzy PID control; Piecewise control; Stability; Thruster model
摘要:A novel adaptive fuzzy proportional-integral-derivative (PID) controller based on the proposed piecewise PID controller is put forward for the sandglass-type floating, drilling, production, storage and offloading unit (FDPSO). First, on the premise of appropriate positioning accuracy, a piecewise linear PID controller considering the directional characteristic between the motions and environmental loads is proposed to conserve energy of traditional linear PID controller. Then, to improve the stability during the transient process and alleviate oscillation induced by piecewise PID controller, the robust adaptive fuzzy PID controller based on the piecewise PID controller is put forward, whose stability is verified by fuzzy Lyapunov method. The proposed fuzzy output variables are designed including motions in both the horizontal and vertical planes to solve the surge-pitch coupling problem and optimize energy consumption. The time-domain simulations with a complete model of all DP components and external forces are carried out. Therein, the thruster model is improved by introducing a virtual thrust command to decrease the obvious phase lag, response delay and amplitude reduction of the traditional thruster model. In addition, the application of wave filter is discussed, especially when the wave-frequency (WF) motions are more prominent and significant than low-frequency (LF) motions. Finally, comparisons between the proposed adaptive fuzzy PID controller and traditional linear PID controller are made and analyzed. Good control effects of the proposed controllers are demonstrated with the assistance of numerical simulations with a series of control parameters.