| 引用本文: | 马黎华,胡笑涛,王 平,等.滴灌玉米叶温的数据驱动模型与不确定性分析[J].灌溉排水学报,2018,37(6):1-8. |
| MA Lihua,HU Xiaotao,WANG Ping,et al.滴灌玉米叶温的数据驱动模型与不确定性分析[J].灌溉排水学报,2018,37(6):1-8. |
|
| 摘要: |
| 【目的】探寻玉米叶温对不同深度土壤含水率、气象因素的响应关系。【方法】采用不同供水条件下的滴灌玉米土箱试验，基于叶温、气象和土层含水率数据设计3种输入项结合线性回归模型和神经网络模型，研究了玉米叶温与环境因素的数据驱动模型及模型的不确定性。【结果】①在叶温变化模拟中，与线性模型相比神经网络模型具有优势，在40%和60%滴灌湿润比处理下的全因素模型的决定系数由0.8提升到了0.9；80%湿润比处理下全因素模型的决定系数由0.5提升到0.7；②单因素不确定性分析中，与叶温变化最密切的气象因素是空气温度，其次是空气湿度和净辐射；在土层含水率的不确定性分析中，30~40 cm土层含水率与叶温变化的响应关系最密切。【结论】结合MC(Monte Carlo method)设计的模型不确定性分析，以d-factor指标量化单影响因素与叶温的响应关系，不同深度土层含水率与叶温的响应关系存在差异，30~40 cm土层是水分响应关键土层。 |
| 关键词: 叶温； 土壤含水率； 气象因素； 模型； 不确定性分析 |
| DOI：10.13522/j.cnki.ggps.2017.0201 |
| 分类号: |
| 基金项目: |
|
| A Data-driven Model and Its Uncertainty Analysis for Estimating Leaf Temperature of Maize under Drip Irrigation |
|
MA Lihua , HU Xiaotao, WANG Ping, WANG Zhenchang
|
|
College of Resources and Environments, Southwest University, Chongqing 400715, China; Key Laboratory of Agricultural Soil and Water Engineering in Arid and Semiarid Areas, Ministry of Education, Northwest A& F University, Yangling 712100, China; College of Water Conservancy and Hydropower Engineering, Hohai University, Nanjing 210098, China
|
| Abstract: |
| 【Objective】 Leaf temperature affects plant physiological development and this paper is to elucidate the responsive change in leaf temperature of maize under drip irrigation to soil moisture and meteorological factors. 【Method】A field experiment was conducted with the maize growing in soil boxes and the soil moisture of the boxes controlled by drip irrigation. During the experiment, leaf temperature, soil moisture and meteorological factors were measured. The change in the leaf temperature with soil water and meteorological factors, as well as its associated uncertainty were analyzed using the artificial neural network (ANN) model and the multivariable linear regression model (MLR). 【Result】①The ANN model is superior to the MLR model for calculating the leaf temperature, with the determining coefficient of the former being r2 =0.8 and of the latter being r2=0.9 when the soil moisture was in the range of 40% and 60%. When the soil moisture increased to 80%, the determining coefficient of the MLR was r2=0.5 and that of the ANN was r2=0.7. ②The uncertainty analysis revealed that for the meteorological factors, the leaf temperature was most sensitive to air temperature followed by net radiation and air humidity, while for soil moisture, the leaf temperature was most sensitive to water content in 30~40 cm. 【Conclusion】 Uncertainty analysis using Monte Carlo method showed that the proposed models are able to quantify the changes in maize leaf temperature with soil moisture and meteorological factors. The d-factor analysis indicated that the impact of soil moisture on leaf temperature is depth-dependent with the leaf temperature most sensitive to 30~40 cm water content. |
| Key words: leaf temperature; soil water content; meteorological factor; model; uncertainty analysis |
