摘要
提出基于LM_BP神经网络的总辐射日曝辐量计算模型,模型以地外日射日曝辐量、日照百分率、日最高温度和日相对湿度等4个气象因子作为输入向量,以日曝辐量作为输出向量,利用气象站实测的气象数据进行模型实验,结果表明:该模型计算结果较为精确,且泛化性能好,可以推广应用。
In this paper, a model for daily irradiation exposure of global radiation calculation based on the LM_BP neural network was put forward. The input vector was the four meteorological factors, which included daily radiance exposure of extraterrestrial solar radiation, percentage of daily sunshine, daily maximum temperature and daily mean relative humidity. And the output vector was the daily radiance exposure of global radiation. A test was carried on the model by using the meteorological data measured by weather stations. The result of the test showed that the data calculated by the model is accurate and the model had good generalization ability. The model can be further popularization and application, which is of great significance for the calculation and utilization of daily radiance exposure in the un-gauged basins.
出处
《太阳能学报》
EI
CAS
CSCD
北大核心
2013年第7期1202-1205,共4页
Acta Energiae Solaris Sinica
关键词
BP神经网络
LM算法
总辐射日曝辐量模型
BP neural network
Levenberg-Marquardt
daily radiance exposure of global radiation model
