Study On Optimization Technology Of Intelligent Propulsion System Of High-Speed Craft

Simulation of ship propulsion system is of great importance to ship design, due to its wide-range relation to the hull, engine and propeller (pump). Computer optimization can save a large amount of time to calculate repeatedly, can reduce the designer's workload, and can shorten the design cycle of ship propulsion system, reduce the cost, economize resources. It has considerable economic and technological value. At first, the kind of high-speed craft and its current situation are introduced in this paper. Through analyze and compare the characteristics of their performance, three kinds of high-speed craft are selected to be optimized. They are hydrofoil-gliding craft, wave-piercing catamaran (WPC) and frigate, which power equipment of propulsion system are the common screw, pitch-adjustable screw and water-jet propeller. Secondly, after analyzing and synthesizing the theories and methods of variant intelligent control, this paper presents a mathematical model of craft's propulsion system based on MATLAB-SIMULINK and work out the optimization program with MATLAB language. In the program, the optimization methods are multi-point crossover GA, adaptive GA (AGA) and fuzzy-GA, the control methods to propulsion system are fuzzy control method and PID control method. The corresponding design variables are the objective velocity, the control parameters of controller and the performance parameters of propeller. Establish the optimization mathematic model which objective function is the rapidity and the control performance, the total objective function is their product. At last, the results of simulation and optimization are analyzed from the optimization result's adaptability. The results show that: the accuracy and reliability of the simulation and optimization models are both excellent; meanwhile the results of three kinds of optimization methods are analyzed, and discuss the effect of crossover point number to calculation results in common GA, the example indicates that AGA and fuzzy-GA are better than the common GA, and the simulation graph of three-point crossover GA is the best in common GA.