Font Size: a A A

The Implementation Of An Improved Genetic Algorithm And Its Application In The Analysis Of Tobacco-growing Soil

Posted on:2013-10-16Degree:MasterType:Thesis
Country:ChinaCandidate:Y SunFull Text:PDF
GTID:2233330371483511Subject:Software engineering
Abstract/Summary:PDF Full Text Request
Tobacco growing industry is one of the main planting industry in China.Choosing to usedifferent types of soil will make the tobacco yield and qulity be separated in the plantingprocess. We improve the traditional genetic algorithm and apply the algorithm designing andimplement a tobacco growing system.During the process of improving the traditional genetic algorithm,in this paper we madethe following works:1. After each selection and crossover operation we add a Gaussian mutation to improvethe standard genetic algorithm; In an adaptive crossover probability way to improve thestandard genetic algorithm and reducing the probability of genetic process fell into a localoptimum solution.2. Encoding the oxygen content in soil,water content,PH value and temperaturevalue,in real numbers. The determination of coding and population size use statistical datawhich is given by the past experience. We concern each code string as an individual in a group.to determine the polutation size by generation gap G. Adding oxygen content,moisturecontent,PH value and temperature value to the design of fitness function. And apply varianceto the fitness function for evaluating the detection. According to the fitness assessment we usethe genetic algorithm modified to operate. Make the fitness change rate as the out of the endconditionsDuring the design an implement of tobacco growing system, we mainly achieve to theestablishment of the tobacco optimal environment for the growth table, the soil factor talbe.And we also design and implement the genetic optimization module. Optimizing resultscomparation module and the graphical interface module. Among which,the results optimizationmodule of system require datas provide by the tobacco optimal growthcondition table. And thegenetic optimal module need to add the soil impact factor as input. Application results interface exhibit module is a module that show optimization results to the graphics windows, convinentfor user-friendly interaction.We had run the application system several times and statistical results that returned.According to the statistic analyse, we find out a reasonable tobacco planting program. And wegive some measures to miprove the stability of application system and ultimately sovle thetobacco planting programs of choice by improved genetic algorithm.In this paper,we also analysis the dispoints of the algorithm improving and applicationsystem implementing so as to give some advises to the follow-up comments we hope whichwould play a role of guide in the following research.
Keywords/Search Tags:Genetic algorithm, Fitness function, Gaussian mutation, Adaptive crossover probability, Soil analysis, Tobacco growing application
PDF Full Text Request
Related items