计算机工程与应用 ›› 2011, Vol. 47 ›› Issue (32): 231-235.

• 工程与应用 • 上一篇    下一篇

卫星云图的遗传神经网络集合预测模型研究

金 龙,黄 颖,何 如   

  1. 广西区气候中心,南宁 530022
  • 收稿日期:1900-01-01 修回日期:1900-01-01 出版日期:2011-11-11 发布日期:2011-11-11

Nonlinear ensemble prediction model for satellite image based on genetic neural network

JIN Long,HUANG Ying,HE Ru   

  1. Guangxi Climate Center,Nanning 530022,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2011-11-11 Published:2011-11-11

摘要: 针对目前缺乏卫星云图非线性预测技术理论方法问题,采用类似于数值预报模式的集合预报方法,构建了一种遗传神经网络的卫星云图非线性滚动预测模型。通过对每间隔3小时的红外卫星云图样本序列作自然正交展开,将提取出的时间系数作为云图预报建模的预报分量。考虑降水云系的发展变化,主要是受到云团环境物理量场的影响,利用数值预报模式的物理量预报产品作为各预报分量的预报因子,并采用系统降维计算处理方法,分别建立相应的时间系数遗传神经网络集合预报模型。将预报得出的各时间系数与空间向量合成,得出未来时刻的卫星云图预报图。预报试验结果表明,这种非线性预报模型可以较好地预报未来20~30小时的强降水云团发展、移动的主要特征和变化趋势。实况云图与预测云图的平均相关系数达到0.78。

关键词: 遗传算法, 卫星云图, 神经网络, 非线性, 集合预报

Abstract: A nonlinear rolling prediction model for satellite image has been developed based on genetic neural network using the ensemble prediction method similar to the numerical prediction model,due to lacking of the guidance of a nonlinear prediction theory for satellite image at present.Empirical Orthogonal Function(EOF) method is applied to the samples of infrared satellite image every 3 h in heavy rainfall processes;time coefficients extracted are used as predictands.Since the changes of precipitation cloud system are governed by the physical quantity fields in cloud cluster,the physical quantities prediction products from numerical prediction model are used as predictors,and Genetic Neural Network Ensemble Prediction(GNNEP) models are established for the corresponding time coefficients based on the technique of the reduction of data dimensionality for data interpretation.By integrating the predicted time coefficient and space vector,the future satellite image is obtained.Results show that the nonlinear prediction model can better forecast the main features of the development of heavy rainfall cloud cluster in future 20~30 h.The mean correlation coefficient between the observed and predicted image is 0.78.

Key words: genetic algorithm, satellite image, neural network, nonlinear, ensemble prediction