基于模式识别的台风风暴潮灾情等级评估模型研究

A Study on the Pattern Recognition Based Model for Evaluation of Typhoon Storm Surge Disaster Degrees

风暴潮灾害每年都给我国沿海地区造成巨大的经济损失。本文在前人研究的基础上,引入风暴潮灾情等级的模式识别思想,运用感知器方法的理论及其算法,并以广东省湛江地区为例,建立了该地区的台风风暴潮灾情等级判别函数,经还原检验和判别预测获得满意的结果,证明该方法用于台风风暴潮灾情等级判别是可行的。这对于实现风暴潮灾情等级的快速评估,为抢险救灾的人力、财力和物力分配提供依据具有重要意义。

Storm surge disasters cause a great economic loss to the littoral regions in our country every year. Based on the research of other scholars, this paper introduces the idea of pattern recognition into the storm surge disaster degree evaluation for the first time. Using the theory and method of the perception approach and the loss data of Zhanjiang region of Guangdong Province, we build a typhoon storm surge disaster degree discriminant function and get satisfactory results. It proves that the perception approach is feasible in the storm surge disaster degree evaluation. This conclusion is significative to realize rapid evaluation of storm surge disaster degrees and to distribute the manpower and material resources in reason when providing disaster relief.