| The positioning method adopted by the traditional marine oil drilling platform is anchoring positioning,but the anchoring positioning has its own defects.As the water depth increases,the positioning accuracy will continue to decrease.The positioning method adopted in this thesis is dynamic positioning,which is fast and accurate and is not affected by water depth.The dynamic positioning relies on the power of the platform itself to resist external disturbance signals such as wind,waves and current under the guidance of the control system,so that the platform is finally stabilized near the target position and direction.The thrust distribution system is an important part of the dynamic positioning system.It is responsible for distributing the force and torque signals assigned by the control system to the propeller.In this thesis,the nonlinear drilling platform model is taken as the research object,and the control system based on multi-dimensional Taylor network(MTN)optimal controller is designed.Based on the genetic algorithm,an effective method to solve the thrust distribution problem is proposed.Then,relaying on the visual simulation technology and matlab simulation data,the 3D visual simulation of the dynamic positioning of the drilling platform based on multi-dimensional Taylor network optimal control is completed.The main contents of this thesis include the following points:1.Establish a mathematical model of the drilling platform.Analyze the influence of external disturbances such as wind,waves and currents on the platform,and establish corresponding mathematical models.2.Design multi-dimensional Taylor network optimal controller.Since the high frequency signal does not affect the final effect of the dynamic positioning,the Kalman filter is designed to filter the input signal.This thesis introduces the multi-dimensional Taylor network theory and designs a multi-dimensional Taylor network optimal controller,and uses the simplex method to optimize its parameters.Considering the parameter uncertainty of the actual drilling platform,the controller robustness simulation experiment was carried out.The results show that the multi-dimensional Taylor network controller has better robustness.3.Thrust distribution.The appropriate thruster installation and configuration strategy is selected,the thrust distribution is regarded as a nonlinear programming problem,the appropriate objective function and constraints are established,and the thrust distribution system is optimized by hybrid genetic simulated annealing algorithm.4.3D modeling technology.The appropriate software is selected to construct the 3D solid model of the drilling platform,and the model is optimized by LOD technology,texture mapping technology and BSP blanking technology.5.3D visual simulation application.Introduce the 3D visual simulation software Vega Prime and build a virtual marine environment with it.MFC is used for secondary development of Vega Prime application;the server/client mode of socket is used to transmit data;OpenGL is used to display real-time data information;the user interface is designed and optimized.Finally,3D visual simulation system of dynamic positioning is completed.Comparing the attitude changes of platforms under different controllers,simulation results show that multi-dimensional Taylor network optimization controller has the best control effect. |
