Research And Application Of Multi-objective Genetic Algorithm Based On Banker Law And Information Entropy

Genetic algorithm was developed from nature, it was proposed by professor J. Holland of University of Michigan in the U.S.A and developed by DR. Goldberg and some other scientists in1980s.It is widely used in the world nowadays owing to its benefit in solve multi-objective problems,although it was proposed for only20to30years.Lots of scientists dedicate themselves to the research of genetic algorithm, and GA is a hot spot of science study today. GA is used in production scheduling, aircraft control, industrial robot design, image processing and other fields of science. And the level of research in GA has also become a great symbol of a country's national strength.This paper first introduced the origin, the development and research status of genetic algorithm, then we discussed the models and merits and demerits of all sorts of multi-objective algorithm. Banker law is a important method of solving the multi-objective problems, its speed is fast but its weak in maintain the diversity of population.Information entropy is from thermodynamics, and is being used in information science now, it can help the population to maintain the diversity of the population. Information entropy is from thermodynamics, and is being used in information science now; it can help the population to keep diversity. In this paper we wish we can combine the two methods together to create a new method of genetic algorithm.In this paper we do the basic genetic operation to the population before we chose out the banker, then we computing the entropy of the population until the value reach the requirement,then we do the banker law to the population. This new method will be applied to the problem of water supply, the research shown that this new method can help the system to work better, it can work quickly and the population can also keep diversity.Figure16table3reference47...