The Application of the Neuro-Fuzzy Algorithm on the Portable Ship Dynamic Positioning Control System in Waves

摘要

The paper applies the Neuro-Fuzzy algorithm on the portable ship Dynamic Positioning (DP) control system in waves and describes the nonlinear dynamic motion behavior of the barge equipped with the portable outboard DP control system under the external forces in the ocean The algorithm is applied to control the thrusters’ revolutions to optimally adjust the ship position heading and path-tracking in the ocean The technique of the present DP control system is also used in the pipe or cable laying missions Since it doesn’t need the mathematic models to derive the controllers the Fuzzy control or Neural Network algorithm is therefore adopted here The time consumption of building controller can then be decreased largely Besides after the training simulation the neural network PD controller could obtain the suitable dynamic control gains for most sea states which could replace the trial and error method to determine the gains in traditional PD controller Moreover the fuzzy control has the physical intuition in the system adjustment and need not to be trained in many cases before working in the real condition Finally the control system combining the two algorithms would have the advantages of self-tuning function and the training time saving In addition to the waves the external loading due to current wind cable and second-order drifting forces are also included in the calculations The time domain simulations for the six degrees of freedom motions of the barge with the DP system are solved by the 4th order Runge-Kutta method As to the head sea and following sea conditions the neuro-fuzzy algorithm on the DP system indeed worked well In the beam sea and oblique sea cases because of the environmental forces the barge heading varies significantly And in the path tracking cases the results show that the path and heading deviation are limited in the acceptable ranges based on the present control method The original contribution of this paper is to break through the limitations of traditional DP system using a single fixed gain value and consider the coupling motion of the barge with DP system According to the different sea states and the ship motions the parameters of present DP system can be adjusted directly and DP capability is enhanced by the Neuro-Fuzzy algorithm When the dynamic positioning missions are needed the technique of the portable alternative DP system based the neuro-fuzzy algorithm can serve as a practical tool to assist those ships without equipping with the professional DP facilities

獎項日期	2014 1月 6
原文	English
監督員	Ming-Chung Fang (Supervisor)