基于KPCA的台风强度神经网络集合预报方法研究

针对影响台风强度前期预报因子较多以及因子的非线性变化特点,首先采用逐步回归方法筛选出部分预报因子,再利用核主成分分析方法在剩余的预报因子中提取包含了原数据较多信息的核主成分与前期选入的预报因子共同作为模型输入.进一步考虑到神经网络集合预报中个体的准确性和差异性的权衡问题,在不同的初始条件下生成若干组神经网络,分别选择每组中性能最优的个体,建立了一种新的非线性神经网络集合预报模型.最后以西北太平洋海域2001-2010年5-10月的台风强度为研究对象进行了预报试验.结果表明,这种神经网络集合预报模型的预报结果符合实际应用的要求,其预报平均绝对误差明显小于同等条件下的神经网络方法和逐步回归预报方法.

A new neural network ensemble forecast method based on KPCA for typhoon intensity

In this paper, a new neural network ensemble forecast model is developed where the stepwise regression method is used to choose forecast factors which are best correlated with the series of typhoon intensity, and the main information is extracted from remaining forecast factor where kernel principal component analysis(KPCA) method is used. Further, the balance of accuracy and difference is considered for the individuals of neural netwoork ensemble forecast, and then the neural network individuals are generated and the best of each group is selected respectivery under different initial conditions. Finally, the typhoon intensity prediction experiment in the northwest Pacific Ocean, from May to October 2001-2010 was conducted. Results show that the mean absolute prediction error of neural network ensemble forecast model significantly less than that of stepwise regression method and single neural network under the same conditions.