Genetic Algorithm And Its Applied Research In Nonlinear Mathematical Models Of Aircraft Engines

With the development of morden aero-propulsion system,The simulation technology has became one of the most important means in aeroengine design. Aeroengine nonlinear mathematical models,as the basis of engine performance simulation,are not convergence favorably in entire flight scope,because of the solution limitations of nonlinear equation and the engine balance equations.In this thesis,genetic algorithm (GA) is used to solve aeroengine nonlinear model problems at first time,fast GA and GA application meathods are studied,the main work and the conclusion of the thesis are summarized as follows:1. Based on the analysis of GA,FORTRAN and C GA software are disigned.2. In order to improve GA compute efficiency,gene supplement GA is put forward and a new kind of evolution orientation GA is designed. Example test shows they can increase the GA compute efficiency 0.5 and 2~4 times respectively.3. The thesis brings forward Random Number Sequence Shuffling Technology in a creative way,it can reduces the Linear Congruential Generator pertinency and discrete grid phenomena prominently.4. In C & C++ GA program,using the parameter num equal to population number will make GA compulate more efficiently.5. Converting nonlinear equations solution to GA optimum problem,and then the problem is discused in GA model,fitness,coding,operator & parameter,convergence critericm multi solutions 6 aspects. At last the thesis set up a general GA for nonlinear equations.6. According to detailed analysis of nonlinear equation,the thesis creates a Adaptive Variable Search Scope GA for nonlinear equation.7. In this thesis,GA is initially applied to aeroengine nonlinear mathematical model,the author designs GA GA act as the base of Newton-Raphson (mixed algorithm 1) and GA alternating with Newton-Raphson (mixed algorithm 2) three kinds of aeroengine balance equation solution methods.8. The theoretical analysis and performance simulation results show that the aeroengine nonlinear mathematical mode not only can maintain the same efficiency with current model,but also can achieve convergence in entire flight scope when balance equation solved in mixed algorithm 2.