Multi-Objective Optimizing For Structure Of Fuel Piston Pump

Multi-objective optimization is a difficult problem and a research focus in the fields of science and engineering. Based on the simple genetic algorithm, researchers have developed many multi-objects optimization genetic algorithms. In this thesis an improved Non-dominated Sorting Genetic Algorithm (NSGAâ…¡) is used for the optimizing for structure parameters of fuel piston pump used in aero-engine. The main contents of the thesis are as follows:First, based on the principle of fuel piston pump which is widely used in aero-engine , geometric algorithm of the pump's main frame is deduced applying theorem of dynamics, and the five important characteristic objective functions are proposed. These objective functions are mutually competitive even incompatible in pump designing.Second, the basic theory and process of Genetic Algorithm are systematically introduced and analysed. The essence of Multi-objective problem is presented, and the main process algorithm of NSGAâ…¡is also introduced in this thesis. In the last of this chapter, the correctness of NSGAâ…¡program is demonstrated by the test function of Deb's design.Last, The fuel piston pump's volume, stress condition, pulsant flux, etc. are selected as objective functions and optimized synchronously by Multi-objective Genetic Algorithm. Compared optimized characteristic with the not ones, NSGAâ…¡is proved effectively, and The result also proves the theory's rationality of real parameters in one kind of aeroengine.