基于改进遗传算法的温室环境动态优化控制  

Dynamic optimal control of greenhouse environment based on improved genetic algorithm

作　　者：晋春 毛罕平[1] 马国鑫[1] 王奇瑞 石强 JIN Chun;MAO Hanping;MA Guoxin;WANG Qirui;SHI Qiang(School of Agricultural Engineering,Jiangsu University,Zhenjiang,Jiangsu 212013,China;School of Electronics and Information,Jiangsu University of Science and Technology,Zhenjiang,Jiangsu 212003,China)

机构地区：[1]江苏大学农业工程学院,江苏镇江212013 [2]江苏科技大学电子信息学院,江苏镇江212003

出　　处：《江苏大学学报（自然科学版）》2022年第2期169-177,共9页Journal of Jiangsu University：Natural Science Edition

基　　金：国家自然科学基金重点资助项目(61233006);江苏省研究生科研与实践创新计划项目(KYCX17_1787)。

摘　　要：针对一种经济效益最优目标的温室环境混合整数变量优化控制问题,提出一种带有工程经验规则的改进遗传算法(IGA)进行有效、实用求解.首先采用分段常数法离散化控制变量,形成一种有限维参数的非线性数学规划(NLP)问题;在标准遗传算法(SGA)直接求解该NLP问题的基础上,采用精确罚函数处理状态变量路径约束;引入工程经验规则,并采用精英保留、多种群并行进化、整数变量取值等改进措施,以提高算法性能.仿真结果显示,相较于SGA,该方法获得更优的性能指标和控制品质,表明了所提方法的有效性、实用性.To effectively and practically solve the dynamic economic optimal control problem of greenhouse environment with mixed integer variables,an improved genetic algorithm(IGA)with engineering constraint rules was proposed.Based on piecewise constant method for discretizing the control variables,the optimal control problem was transformed into nonlinear programming(NLP)problem with finite-dimension parameters,and the standard genetic algorithm(SGA)was used to solve the NLP problem.A precise penalty function was used to deal with the state variable path constraints.The engineering constraint rules and some improvement measures of elite retention,multi-population parallel evolution and integer variable setting were used to improve the algorithm performance.The simulation results show that compared with SGA,IGA obtains better performance indexes and control quality,which proves the effectiveness and practicability of the proposed method.

关 键 词：温室环境 动态优化控制 改进遗传算法 经济效益最优 非线性规划 

分 类 号：S625.5[农业科学—园艺学]