Improved Research And Application On Simple Genetic Algorithm

Genetic algorithm is an adaptive global optimization probability search algorithm that simulates biological genetic and evolutionary processes in the natural environment. GA has strong versatility, adaptability, robust and parallelism, and it has been widely for function optimization, combinatorial optimization, production scheduling problems, automation, robotics, image processing, artificial life, and machine learning and other aspects.Simple genetic algorithm is a prototype of other genetic algorithms and infrastructure, and it not only provides a basic framework to various genetic algorithms, but also has a certain practical value. The paper studies the proportional selection mechanism and elitist strategy in simple genetic algorithm and put forward some improved method. Those methods are applied into solving equation (group) problem, and some results are achieved.The work in paper mainly includes the following aspects:1. The basic principles of genetic algorithms, implementation and study status are briefly introduced.2. The defects of proportional selection operator are analyzed and an improved proportional transformation method is inducted into SGA.3. The paper presents an elitist genetic algorithm base on genetic pool for overcoming the defects of general elitist GA. The basic idea of new method is that a genetic pool is set up for every population, and several better individuals are put into. The simulation results indicate that the new method not only improves the rate of convergence but also avoids the premature convergence.4. Improved GA is applied into solving equation. Numerical experiments show that the method is possessed of feasible, high global convergence and universal, but inferior to traditional methods in the search speed and precision.