摘要
随着嵌入式系统的发展,如何在嵌入式控制终端上实现不同根温、气温、CO2浓度下光饱和点的动态获取,成为设施农业发展的重要问题。以水培生菜为试验材料,设计多因子嵌套试验方案,基于随机森林算法建立面向嵌入式控制终端的光调控目标值模型,研究随机森林算法参数对模型的影响,并对模型进行验证分析。结果表明:该模型可实现基于嵌入式控制终端的光饱和点的动态获取,模型决定系数为0. 9955,均方根误差为5. 677,平均绝对误差为5. 3475,可为面向嵌入式系统的光环境高效精准调控提供理论依据。
With the development of embedded system,the dynamic acquisition of light saturation points under different root temperature,air temperature and CO 2 concentration on the embedded control terminal has become an key issue for the facility agriculture.In this study,a multi-factor nested test was designed and performed on the hydroponic lettuce,and an optical control target value model for embedded control terminals was established based on the random forest algorithm.The influence of random forest algorithm parameters on the model was discussed,and the model was verified and analyzed.The model determination coefficient was 0.9955,root-mean-square error was 5.677 and average absolute error was 5.3475.The results showed that the model could not only achieve dynamic acquisition of light saturation points on embedded control terminals,but also provide a theoretical basis for efficient and accurate regulation of light environment for embedded systems.
作者
完香蓓
简丽蓉
辛萍萍
单慧勇
胡瑾
WAN Xiang-bei;JIAN Li-rong;XIN Ping-ping;SHAN Hui-yong;HU Jin(College of Mechanical and Electrical Engineering,Northwest Agriculture and Forestry University,Yangling 712100,China)
出处
《上海农业学报》
2019年第2期42-47,共6页
Acta Agriculturae Shanghai
基金
国家自然科学基金(31501224)
中央高校基本科研业务费(2452017124)
农业新品种新技术新设施引进项目(201703080)
关键词
光饱和点
水培
随机森林算法
光调控
嵌入式系统
Light saturation point
Hydroponics
Random forest algorithm
Light environment regulation
Embedded system
