摘要
根据天津市2013年12月1日—2013年12月24日气象监测数据,先进行温度、湿度、风力因素对PM2.5浓度影响的相关性分析及定性分析,绘制温度、湿度和PM2.5浓度的二维趋势分布图及气体扩散风向浓度分布图;再运用BP神经网络模型对天津市2013年12月25日—2014年1月9日PM2.5浓度进行仿真预测,最终得到精确预测值。结果表明:温度及风速因素与PM2.5的浓度成反比,湿度因素与PM2.5的浓度成正比,而且通过BP神经网络模型对于"离散样本"、"气象参数不确定性"的实际天气情况可以得到较高的预测精度。
PM2. 5hasn't receive much attention until 2013 when people start to understand it would bring dreadful impacts to human health. According to the meteorological monitoring data for PM2. 5from December 1,2013 to December 24,2013 in Tianjin,this paper firstly analyzes the impact of temperature,humidity and the power of wind on PM2. 5. The study draw the chart of tendency in two dimension to analyze PM2. 5concentrations. In the end,using the mathematical model of BP Neural Networks,it gets the accurate result of PM2. 5concentrations from December 25,2013 to January 9,2014 in Tianjin for convenient prediction based on weather forecast,which provides theoretical basis and realistic reference for controlling and reducing PM2. 5concentrations in the air.
出处
《环境科学与管理》
CAS
2016年第6期121-125,共5页
Environmental Science and Management
关键词
大气污染
污染防治
PM2.5
BP神经网络
预测
air pollution
pollution prevention and control
PM2.5
BP neural networks
prediction
