Research On Adaptive PID-fuzzy Ship Course Controller Optimization

A well-performed method for ship course control is necessary to ensure a safe sailing,reduce course deviation and the unnecessary sailing time so as to save energy.Meanwhile,a ship course control with superior performance is the premise of path following and dynamic positioning.In order to improve the control effect,an improved genetic algorithm and a sub-fuzzy controller are proposed in this thesis to optimize the membership function and scale factor of fuzzy PID controller.Detailed contents include three aspects as follows:On one hand,a MMG module model for a 200-meter PCC(pure car carrier)is established according to the ship maneuvering motion equations of 4-DOF.Then with the help of software Matlab,the simulation experiment of ship turning tests is carried out to verify the reliability of the established mathematical model,by comparing the simulation data with the model test data.On the other hand,a conventional compound type of fuzzy PID controller is designed.A fuzzy controller is added to adjust separately the three main parameters in a PID course controller for a good adaptive ability during the navigation.Specially,a novel strategy is put forward to optimize the membership functions and the three scale factors in fuzzy controller.For the optimization of membership functions,the basic principles for setting the membership functions are concluded,considering which,the comprehensive index is summarized so that whether the settings of membership functions are good or not can be evaluated.Then the improved genetic algorithm with adaptive crossover rate and adaptive mutation rate is introduced and applied to get the global optimal solution for membership function parameters.Meanwhile,according to the mathematical principle of classical genetic algorithm,it is proved theoretically that the global optimal solution is found by the mentioned improved algorithm.For the optimization of scale factors,a two-input/there-output sub-fuzzy controller with the adjusted fuzzy rules is added to adjust the scale factors online.At last,simulations are carried out.The simulation results indicate that the improved fuzzy PID with membership functions optimized and a sub-fuzzy controller performs better than the standard fuzzy PID controller for ship course control regarding the control effect.