基于MLR和人工神经网络的新疆杨日蒸腾量估算

Estimating daily transpiration of Populus alba var.pyramidalis Bunge based on MLR and artificial neural network

为解决当前旱区防护林主要树种新疆杨日蒸腾量(Tr)估算值的精确度低、估算模型泛化能力差等问题,采用7种气象因子(日照时长、风速、相对湿度、饱和水蒸气压差、最低温、最高温和日均温)的8种组合作为模型输入,构建了传统多元线性回归模型(MLR)和人工神经网络模型(BP和Elman),估算2020年生长季新疆杨Tr值,并对3种模型不同输入组合的估算结果进行比较与评价;同时基于Garson算法量化各气象因子对Tr估算值的相对贡献率。结果显示,BP和Elman模型对Tr估算值的精确度超过73.66%,在不同输入组合下人工神经网络模型估算精确度比MLR模型提高了8.45%~31.33%,其中拓扑结构为6-4-4-1的Elman模型估算值的精确度最高;气象因子饱和水蒸气压差对Tr估算值的相对贡献率最大,相对湿度次之,不同温度变量对Tr估算值的相对贡献率依次为:日均温>最低温>最高温。结果表明,所构建的新疆杨日蒸腾量的估算神经网络模型可提高对干旱地区防护林蒸腾量的估算精确度。

A traditional multiple linear regression model(MLR)and an artificial neural network mod⁃el(Back propagation(BP)and Elman)were constructed using 8 combinations of 7 meteorological factors including sunshine duration,wind speed,relative humidity,saturated vapor pressure difference,minimum temperature,maximum temperature,and average daily temperature as model inputs to solve the problems of low accuracy in estimating the daily transpiration(Tr)of Populus alba var.pyramidalis Bunge,the main tree species of shelterbelt forests in arid areas,and the poor generalization ability of estimation mod⁃els.The Tr value of Populus alba var.pyramidalis Bunge in 2020 growth season was estimated.T he re⁃sults of estimating three different input combinations of models were compared and evaluated.At the same time,the relative contribution rates of various meteorological factors to the estimated Tr values were quanti⁃fied based on the Garson algorithm.The results showed that the accuracy of BP and Elman models in esti⁃mating Tr exceeded 73.66%.Under different input combinations,the estimation accuracy of the artificial neural network model had increased by 8.45%-31.33%compared to the MLR model.Among them,the Elman model with a topological structure of 6-4-4-1 had the highest accuracy of estimation.The relative contribution rate of saturated vapor pressure difference to Tr estimation was the largest,followed by rela⁃tive humidity.The relative contribution rate of different temperature variables to the estimated Tr values was in the increasing order of average daily temperature>minimum temperature>maximum tempera⁃ture.It is indicated that the neural network model for estimating daily transpiration of Populus alba var.py⁃ramidalis Bunge can improve the accuracy of estimating the transpiration of shelter forests in arid areas.It will provide scientific guidance for the sustainable development of shelter forests and the precise regulation of agricultural water resources.