Research On Optimization And Control Strategy For Ship Dynamic Positioning System

Dynamic positioning system can resist external interferences (wind, wave and ocean current etc) and keep ship steadily in given position and yaw angle in command of control system with its own power. It has no influence of ocean depth, positions accurately and rapidly, and operates expediently. Thrust allocation is an important part of the dynamic positioning system. A rational and effective method of solving thrust allocation to the semi-submersible rig based on the GA optimal algorithm is proposed.The strategy of thrust allocation and the theory of genetic algorithm are introduced firstly. Secondly, based on that work, mathematical model of thrust allocation is built and the optimization objectives of algorithm and various constraints are proposed according to the actual needs of the system. Finally, the improved parallel hybrid genetic algorithm is designed for the vessel with eight azimuthing thrusters.The meaning of applying parallel mechanism of hybrid optimization algorithm is that it can not only preserve the optimal solutions which make the continuation of effected generation of elite solutions be possible, but also keep non-elite solution involved in the evolution of the next generation in case of the introduction of external parallel evolution mechanism. By limiting the scope of the current solution, it can keep some non-elite solution. Simultaneously, the diversity of population is kept by using the ideas of Nicked Pareto Genetic Algorithm to insert every generation continually. The improved algorithm achieved a good balance between diversity and convergence. The algorithm is simple in structure and easy to implement.Finally, the proposed algorithm is applied to optimization problems in the two kinds of typical simulation environment, and the simulation is done. By choosing different parameters, the performance of the algorithm is compared and analyzed.Results show that the proposed algorithm is effective for solving thrust optimization allocation problem and good results can be achieved.