遗传算法求解林分空间结构优化问题  

Genetic algorithm to solve the optimization problem of stand spatial structure

作　　者：卿东升 彭进香 李建军[3] 刘帅[3] 邓巧玲 QING Dongsheng;PENG Jinxiang;LI Jianjun;LIU Shuai;DENG Qiaoling(College of Life Science and Technology,Central South University of Forestry and Technology,Changsha,Hunan 410000,China;School of Information Engineering,Hunan Applied Technology University,Changde,Hunan 415000,China;College of Computer and Information Engineering,Central South University of Forestry and Technology,Changsha,Hunan 410000,China)

机构地区：[1]中南林业科技大学生命科学与技术学院,湖南长沙410000 [2]湖南应用技术学院信息工程学院,湖南常德415000 [3]中南林业科技大学计算机与信息工程学院,湖南长沙410000

出　　处：《森林与环境学报》2022年第4期434-441,共8页Journal of Forest and Environment

基　　金：国家自然科学基金项目“气候变化模式下森林结构多目标优化模型研究”(31570627);湖南省教育厅科学研究项目“基于新媒体技术的数据可视化分析与应用研究”(20A358);湖南省科技计划项目“洞庭湖低质次生林结构调控及重构关键技术”(2015WK3017)。

摘　　要：林分空间结构优化近似非确定性多项式难题,为了提高其优化精度和效率,提出一种基于遗传算法的林分空间结构多目标优化方案。在构建林分空间结构多目标优化数学模型的基础上,将林分中的林木编码,使用1条染色体代表林分空间结构优化过程中的1个非劣解,通过选择、交叉和变异等方案逐步更新染色体结构,解决了林分空间结构多目标优化问题,并以湖南省大围山及乌云界自然保护区中4块方形试验样地中的林木数据进行了仿真试验。结果表明,大围山地区样地D1的林分适应度值从0.057提升到0.076,样地D2的林分适应度值从0.134提升到0.474;乌云界地区样地W1的林分适应度值从0.098提升到0.122,样地W2的林分适应度值从0.099提升到0.373。各样地林分适应度值均有不同程度的提升,表明遗传算法在求解林分空间结构多目标优化的问题上是有效的。The optimization of stand spatial structure is approximate to the nondeterministic polynomial problem.To optimize accuracy and efficiency,a multi-objective optimization scheme of stand spatial structure based on a genetic algorithm is proposed.Based on the multi-objective optimization mathematical model of stand spatial structure,the trees in the stand were coded,and a chromosome was used to represent a non-inferior solution in the process of stand spatial structure optimization.Then,the chromosome structure was gradually updated through selection,crossover,and variation.Finally,the multi-objective optimization problem of stand spatial structure was solved.Additionally,simulation experiments were carried out on the forest data of four square experimental plots in Dawei Mountain and Wuyunjie Nature Reserve in Hunan Province.According to the experimental results,the stand fitness value of sample plot D1 in the Daweishan area increased from 0.057 to 0.076,while that of sample plot D2 increased from 0.134 to 0.474.The stand fitness value of sample plot W1 in the Wuyunjie area increased from 0.098 to 0.122,while that of sample plot W2 increased from 0.099 to 0.373.In conclusion,the stand fitness values of various plots have been improved to varying degrees,which shows that the genetic algorithm is effective in solving the problem of multi-objective optimization of stand spatial structure.

关 键 词：林分空间结构 遗传算法 多目标优化 林分空间结构优化 智能算法 

分 类 号：S718.5[农业科学—林学] TP18[自动化与计算机技术—控制理论与控制工程]