Research On DP Ship Motion Control Method Based On Kalman Filtering Technology

It is very difficult for a ship to maintain its position in the complicated and changeable ocean,The generation of Dynamic Positioning(DP)technology can effectively solve the problem of ship positioning,track and regional control,and it is one of the indispensable equipment for deep-sea resource exploration and development,deep-sea scientific investigation and other operations.When the ship is working on the sea,it is affected by the wind,wave and current,and then produces the coupled motion response including the first-order high-frequency wave motion and the second-order low-frequency drift motion.Before the signal enters the control loop,it is necessary to filter out the first-order high-frequency wave motion and noise signal from the coupled motion,and estimate the ship's three degree of freedom motion state by decoupling,then control the only second-order low-frequency drift motion response.The filtering effect of wave motion will directly affect the performance of dynamic positioning system,such as:Dynamic Positioning accuracy,system power consumption,propeller wear,etc.Therefore,wave motion filtering and state estimation technology has become the difficulties,hot spot and one of key technologies in the research of ship Dynamic Positioning technology.First of all,this paper introduced the coordinate system,kinematics model,high frequency model,low frequency model and marine environment interference force model of DP ship motion in detail,so as to prepare for the filtering and state estimation of DP system.Secondly,aiming at the nonlinear model of DP ship,the wave force structure and its influence in the Dynamic Positioning system are analyzed,and the Nonlinear Passive filter and relevant observation equations are designed by using Lyapunov theory.The dynamic characteristic of the error is studied,the passivity and stability of its subsystems H1 and H2 are analyzed,the selection method and numerical requirements of the filter parameters are determined,and the convergence of the filter is verified,Then,based on the ship's dynamics and kinematics equation,a set of ship Dynamic Positioning filtering and state estimation state equation based on Nonlinear Passive filtering method is designed,and a simulation model of ship Dynamic Positioning system's three degree of freedom motion filtering and state estimation system is built,the simulation results show the effectiveness and authenticity of the Nonlinear Passive filter results.Thirdly,the control system of DP motion filtering based on Kalman filtering is established,an improved DPS three degree of freedom decoupling filtering method based on Kalman filtering technology is proposed,the establishment method of sensor noise covariance matrix and measurement covariance matrix parameters is given,and a DPS state observer based on position and yaw angle output is constructed to filter the nonlinear passive estimation.By comparing the filtering and state estimation results of Nonlinear Passive filter and Kalman filter,it is proved that Kalman filter has better filtering effect,more accurate estimation results of high frequency motion,smoother force(moment)output by the controller,and more stable motion track and speed of DP ship.Finally,PID control method is used to design a three degree of freedom motion controller for DP ship,and the propulsion system of DP ship is modeled.Then,direct allocation method and pseudo inverse method are selected to thrust allocation.The applicability of the two methods is verified by simulation.Then,the pseudo inverse method is used as the thrust distribution method,the Nonlinear Passive filter and Kalman filter are used as the state observer respectively.It is verified that the results of the first-order wave motion filtering seriously affect the stability of the thrust output and rotation azimuth of the azimuth propeller,the accuracy of DP control and the energy consumption of the propulsion system after thrust allocation.It is proved that the Kalman filter is effective and accurate,and it is also proved that the wave motion filtering and the state estimation technology are important and necessary for the study of the dynamic positioning system of surface ships.