台风最大风速预测建模仿真与研究

Modeling and Simulation of Maximum Wind Speed Prediction for Typhoon

研究及时并准确地预测台风最大风速,对于气象防灾减灾有重大意义。针对影响台风最大风速的因素较多、众多影响因素与台风最大风速之间存在非线性关系、不同地区数据样本集间差异性很大等问题,采用仿射传播聚类方法与稀疏贝叶斯回归模型相结合,建立了一种新的台风最大风速预测的混合模型。上述模型首先采用仿射传播聚类算法快速客观地对训练样本进行聚类,并将其划分为若干个子类,再用稀疏贝叶斯回归模型对各子类样本分别建立稀疏高效的子模型。试验结果表明,上述混合模型不仅能够满足台风最大风速预测的绝对误差的预定要求,而且预测的绝对误差明显小于同等条件下的其它回归模型预测的绝对误差。

The maximum wind speed of typhoon is predicted timely and accurately, which has great significance to the meteorological disaster prevention and mitigation. In the prediction model for the maximum wind speed of ty- phoon, the number of the input variables is very large and the output variable is a nonlinear function of the input vari- ables, and the sample data have great differences in different regions. A sparse Bayesian mixture model based on af- finity propagation clustering is presented. Firstly, the training samples are clustered into several sub- classes by af- finity propagation clustering algorithm quickly and objectively. Then, the sub - models are trained by sparse Bayesian regression technique with corresponding sub - class samples efficiently. The experimental results show that the sparse Bayesian mixtures regression model based on affinity propagation clustering meets the predetermined requirements on absolute error, and it is superior to other regression models.