摘要
对雾霾的监控预测已成为城市重点监测指标,传统雾霾监测方法使用的是高斯预测模型,在预测过程中会使用大量的参数以及逻辑关系,在大数据分析的趋势下已经不能承接海量相关大数据,针对上述问题,提出一种基于大数据分析的雾霾污染预测方法。建立数据K模型承接海量数据,对大数据特点趋势加以分析,使用组合算子对雾霾污染度预测,摒弃逻辑关系束缚以及大数据的驱使性,对特征数据进行类别掌控完成预测结果。通过对比实验的方式对提出的方法进行检验,实验结果表明:(1)基于大数据分析的雾霾污染预测方法能够在大数据环境下对雾霾污染度进行高精度预测;(2)与传统方法相比较能够承接大数据的变化;(3)摒弃逻辑关系的制约可以实现随时预测;(4)极大的缩减预测成本。
Monitoring for fog in cities has become a key indicator.Traditional fog monitoring method that using Gaussian model will use a large number of parameters in the process of forecasting and logical relationship.Under the trend of big data analysis cannot undertake massive forecasting data,aiming at these problems,put forward a kind of smog pollution prediction method based on large data analysis.Data K model is set up to undertake huge amounts of data and trend analysis of the characteristic of big data and use combination operator of smog pollution degree measurement,abandon the logical relationship between bound and big data driven,categories of features is control of the predicted results.By means of contrast experiment to test method,the experimental results show that（1） haze pollution prediction method based on large data analysis in large data environment for high precision fog haze pollution forecast;（2） compared with the traditional method can accept the change of the big data;（3） instead of the logical relationship,we can improve the restriction to realize every prediction to;（4） greatly reduced cost.
出处
《环境科学与管理》
CAS
2017年第10期122-126,共5页
Environmental Science and Management
关键词
大数据分析
雾霾污染
污染预测
big data analysis
smog pollution
pollution forecast
