基于SVM-ACO算法的光环境优化调控模型  

Optimal Control Model of Light Environment Based on SVM-ACO Algorithm

作　　者：王智永[1] 张海辉[1] 胡瑾[1] 辛萍萍[1] 张珍[1] 张斯威 张盼[1] Wang Zhiyong;Zhang Haihui;Hu Jin;Xin Pingping;Zhang Zhen;Zhang Siwei;Zhang Pan(College of Mechanical and Electrical Engineering, Northwest Agricuhure and Forestry University, Yangling 712100, China)

机构地区：[1]西北农林科技大学机械与电子工程,陕西杨凌712100

出　　处：《农机化研究》2018年第10期160-165,共6页Journal of Agricultural Mechanization Research

基　　金：国家自然科学基金项目(31501224,31671587);农业科技创新与攻关项目(2016NY-125);农业科技创新计划项目[NC1504(3)]

摘　　要：设施光环境优化调控是提高作物产量与品质的关键,而光饱和点决定作物利用光的能力。因此,如何根据温度、CO_2浓度变化实现光饱和点的动态获取是设施光环境调控技术发展的重要问题。针对上述问题,本文提出基于支持向量机—蚁群算法(SVM-ACO)的黄瓜光环境优化调控模型。通过多因子嵌套试验获得不同光光量子通量密度、CO_2浓度、温度组合条件下的光合速率值,利用支持向量机算法建立光合速率模型,设计基于连续蚁群寻优算法获取光饱和点并以其为调控目标,建立全范围温度、CO_2浓度下的光环境优化调控模型。调控模型可实现光饱和点的动态获取,且模型决定系数为0.995,均方根误差为15.73,为设施光环境高效精准调控提供了理论依据。Light environment control facilities is the key to improve the yield and quality of crops,and the light saturation point and determine the crop light utilization ability,therefore,according to the change of CO2 concentration temperature and light saturation point is an important problem in the development of dynamic access control technology of light environment facilities.In view of the above problems,this paper proposes an optimized control model of cucumber light environment based on support vector machines ant colony algorithm（SVM-ACO）.Through multi factor nested test of different light photon flux density and CO2 concentration and temperature conditions of photosynthetic rate value,using support vector machine algorithm of nonlinear fitting to establish photosynthetic rate model,the design of continuous ant colony algorithm optimization method based on light saturation point,and then to the light saturation point as control target,establish the optimal regulation model of light environment of the full range of temperature and CO2 concentration.The control model can achieve the dynamic extraction of light saturation point,and the model has a coefficient of 0.995 and a root mean square error of 15.73.It provides a theoretical basis for efficient and precise regulation and control of the facility light environment.

关 键 词：光环境优化 调控模型 光合速率 支持向量机 蚁群算法 

分 类 号：S625.52[农业科学—园艺学]