基于主成分分析和BP神经网络的五道沟水面蒸发计算研究

针为探讨五道沟水面蒸发量与气象因子间的关系,准确估算该地区水面蒸发量,选取五道沟1991—2019年水面蒸发量以及气象因子实测资料,基于主成分法分析水面蒸发量的影响因素,并结合BP人工神经网络算法建立了水面蒸发计算模型.结果表明:主成分分析提取了三个主成分,第一主成分为地表温度、饱和差、绝对湿度、平均气温以及水汽压力差...

Calculation of Water Surface Evaporation in Wudaogou Based on Principal Component Analysis and BP Neural Network

In order to explore the relationship between water surface evaporation and meteorological factors in W udaogou region,and accurately estimate its water surface evaporation,this paper selected the observed data of water surface evaporation and meteorological factor in the region during 1991-2019,and analyzed the water surface evaporation based on the principal component method.In addition,water surface evaporation calculation model was established by combining with BP artificial neural network algorithm.The results show that the principal component analysis extracts three principal components.The first principal component is a linear combination of surface temperature,saturation diference,absolute humidity,average temperature and water vapor pressure difference.The second principal component is a linear combination of relative humidity and solar radiation.The third principal component is mainly the influence of wind speed.The input layer of the BP neural network model is 3,and the model input dimension is low.MAE of the true value and the estimated value is 0.18 while RMSE is 0.25,and both of which are less than 1.The model calculation accuracy is high.It can be better used to calculate the actual value of water surface evaporation.