基于支持向量回归机的城市空气污染浓度预报模型研究

Prediction Model of Urban Air Pollution Concentration based on Support Cector Regression Machine

城市空气污染浓度预报存在准确性不高,构建基于支持向量回归机的城市空气污染浓度预报模型。采集历史空气污染数据(PM_(2.5)、NO_(2)、SO_(2)、CO和O_(3))和气象数据(温度、湿度、气压和风速),并实施数据缺失处理、数据离群检测与处理以及数据规范化等预处理,结合支持向量回归机构建城市空气污染浓度预报模型,得出模型预报结果。结果表明:所研究模型应用下,复相关系数的平方值更大,更接近1,说明该模型预报准确度更高,更接近真实值。

The accuracy of urban air pollution concentration prediction is not high.The urban air pollution concentration prediction model based on support vector regression machine is constructed.The paper collected historical air pollution data(PM_(2.5),NO_(2),SO_(2),CO and O_(3))and meteorological data(temperature,humidity,air pressure and wind speed),and implement pretreatment such as data missing processing,data outlier detection and processing and data standardization.It combined with support vector regression mechanism to build urban air pollution concentration prediction model and obtain model prediction results.The results show that the square value of the complex correlation coefficient is larger and closer to 1,indicating that the prediction accuracy of the model is higher and closer to the real value.