大气中臭氧含量分析预测的支向量机模型

SVM Forecasting Modeling Based on the Ozone Concentration in Air

以俄亥俄州(O h io)的气象、臭氧监测数据为基础,对一个监测点数据进行了分析处理,运用支持向量机回归方法,对气象指标的多参数样本进行学习,获得精确的支持向量机映射关系,并对臭氧含量进行预测.预测结果的误差较小,符合实际情况,能够较好的解决实际问题,说明支持向量机回归在预测上具有小的结构风险与强的泛化能力.

We analyzed and processed the data with the support vector machine method basing on monitoring of the ozone and the meteorology in Ohio, and we acquired the accurate vector machine mapping relation, and also forecast the ozone content correspondingly in the coming years. The error of the forecast result is very small, that was coincident with the actual facts. So SVM is effective method to resolve some actual problem and has small configuration risk and strong abroad ability.