论文部分内容阅读
风速、空气温度、空气湿度、土壤含水量是影响地表显热的主要因素。利用观测数据应用以BP神经网络为主并结合相关系数的分析方法研究显热通量的动态特征,将显热通量与风速、空气温度、空气湿度、土壤含水量做了相关性分析。结果表明空气温度和空气湿度与显热通量相关程度显著,它们的相关系数绝对值都在0.5以上,BP网络权值的点积明显高于其它两个影响因子;风速和土壤含水量与显然通量相关程度普通,它们的相关系数仅略高于0.3。此外,风速、空气温度与显热通量呈正相关关系而空气湿度、土壤含水量与显热通量呈负相关关系。通过对这些影响因子进行分析将有助于提高测定显热通量的精度,利用BP神经网络预测显热通量值可以为气象、生态环境等方面的研究提供可靠的决策数据。
Wind speed, air temperature, air humidity and soil water content are the main factors that affect the surface sensible heat. Based on the observation data, the dynamic characteristics of sensible heat flux were studied based on BP neural network and correlation coefficient. The correlation between sensible heat flux and wind speed, air temperature, air humidity and soil water content were analyzed. The results show that air temperature and air humidity have significant correlation with sensible heat flux, their correlation coefficients are all above 0.5, and the dot product of BP network weights is obviously higher than the other two influencing factors; wind speed and soil water content are obviously The flux correlation is normal, and their correlation coefficient is only slightly above 0.3. In addition, wind speed, air temperature and sensible heat flux are positively correlated with air humidity, soil water content and sensible heat flux have a negative correlation. By analyzing these influence factors, it will be helpful to improve the accuracy of determining the sensible heat flux. Using the BP neural network to predict the sensible heat flux value can provide reliable decision data for meteorological and ecological research.