摘要
绿色植物的光合作用受多方面因素的影响,因此,构建融合多种因素的光合速率模型是实现作物生长状态高效预测的关键。试验选取奶油生菜作为研究对象,设置多因素嵌套试验,其中外界环境变量包括光量子通量密度、室温、根温、二氧化碳体积分数等,共获得试验数据648组,利用Pearson分析法对各因素与植物光合速率进行相关性分析,并选定根温、叶温、二氧化碳体积分数、光量子通量密度为建模输入特征,在此基础上构建基于LMBP算法的奶油生菜光合速率模型。结果表明,根温在0.01水平上与植株的光合速率显著相关。与未加入根温的训练结果进行对比,加入根温后模型训练集和测试集精度均有提高;训练集均方根误差为1.46×10^(-3)μmol/(m^(2)·s),平均绝对误差为1.873×10^(-3)μmol/(m^(2)·s),测试集决定系数为0.995 26。基于LMBP算法的奶油生菜光合速率模型考虑了根温对植物光合作用的影响,可以实现光合速率的精准预测。
The photosynthesis of green plants is affected by many factors.Therefore,constructing a photosynthetic rate model integrating multiple factors is the key to achieve efficient prediction of crop growth.In this paper,cream lettuce was taken as the research object,a multi-factor nested experiment was set,and the external environmental variables included photon flux density,room temperature,leaf temperature,carbon dioxide volume fraction,and so on.648 sets of experimental data were obtained.Pearson correlation analysis method was used to analyze the correlation between each factor and photosynthetic rate.Root temperature,leaf temperature,carbon dioxide volume fraction,and photon flux density were selected as the input characteristics for modeling.On this basis,a photosynthetic rate model of cream lettuce based on LMBP algorithm was established.The results showed that root temperature was significantly related to the photosynthetic rate of the plant at the level of 0.01.Compared with the training results of the model without considering the root temperature,the accuracy of the training set and test set of the model after adding root temperature was improved.The root mean square error and the mean absolute error of the training set were 1.46×10^(-3)μmol/(m^(2)·s) and 1.873×10^(-3)μmol/(m^(2)·s),respectively.The determination coefficient of the test set was 0.995 26.The photosynthetic rate model of cream lettuce based on the LMBP algorithm considered the effect of root temperature on plant photosynthesis,and could achieve accurate prediction of photosynthetic rate.
作者
黎雪
LI Xue(Biological Engineering Faculty,Yangling Vocational&Technical College,Yangling 712100,China)
出处
《山西农业科学》
2023年第3期257-263,共7页
Journal of Shanxi Agricultural Sciences
基金
杨凌职业技术学院产学研基地提升项目(JD20-05)
杨凌职业技术学院科技创新项目(ZK21-66)。
关键词
根温
预测模型
光合速率
相关性分析
LM训练法
root temperature
prediction model
photosynthetic rate
correlation analysis
LM training method
