论文部分内容阅读
利用在湖北省鄂州节水示范基地获得的试验资料,对采用滴灌灌溉的越冬大棚番茄上中下3种叶位叶片蒸腾速率的变化规律进行了分析,得出它与主要气象因子之间存在复杂相关性.针对气象因子之间存在严重多重相关性(最大的方差膨胀因子VIPmax=86.46>10)、传统多元回归分析方法失效的情况,引入PLS(Partial Least-Squares Regression)方法,利用土壤温度、相对湿度、平均气温、大气压、蒸发、太阳辐射(日照)等气象因子建立了预测大棚番茄顶层蒸腾速率的偏最小二乘回归模型.预测结果精度很高,最大相对误差仅为0.48%.该模型具有较强的有效性,为通过常规气象数据推求大棚内作物需水规律提供了一种新的思路.
Based on the experimental data obtained from the water-saving demonstration base in Ezhou, Hubei Province, the change rule of leaf transpiration rate in the middle and lower leaves of tomato in winter greenhouse with drip irrigation was analyzed. It was found that there was a complicated relationship between it and the main meteorological factors Correlation.According to the multiple multiple correlations (the maximum variance expansion factor VIPmax = 86.46> 10) between the meteorological factors and the traditional multiple regression analysis method, PLS (Partial Least-Squares Regression) The partial least squares regression model was established to predict the top transpiration rate of tomato in plastic greenhouse with relative humidity, average temperature, atmospheric pressure, evaporation and solar radiation (sunshine), etc. The prediction accuracy was very high and the maximum relative error was only 0.48% Has a strong validity, and provides a new idea for deducing the law of water requirement of crops in the greenhouse through the conventional meteorological data.