摘要
针对现有颗粒扩散模拟方法未考虑多个污染源问题,导致模拟效果不理想,提出基于GIS的大气污染颗粒物浓度扩散模拟分析。按照离散气象站点位置的风向和风速信息构建风场表面,利用烟囱出口位置风速均值等环境条件,计算大气颗粒物污染参数,建立高斯扩散模型;考虑多个污染源问题,转换高斯扩散坐标系与地理坐标系;创建基于GIS的大气颗粒物浓度扩散系统,使用相空间重构把颗粒物浓度序列映射至高维特征矢量空间,形成相点结构训练样本,运用支持向量机回归进行训练获得最优参数值,增强颗粒物浓度扩散模拟可靠性。实验结果表明,上述方法可有效划分颗粒物浓度扩散范围及影响水平,模拟结果精度较高,能够为大气污染整治和有关研究提供数据支持。
The traditional particle diffusion simulation method ignores many pollution sources, resulting in poor simulation effect. Therefore, this paper puts forward a GIS based simulation analysis of atmospheric particulate matter concentration diffusion. The wind field surface was constructed based on the wind direction and speed of discrete weather stations. According to the mean wind speed at the chimney outlet, the atmospheric particulate pollution parameters were calculated to establish the Gaussian diffusion model. Gauss diffusion coordinate system and geographic coordinate system were transformed by the analysis of multiple pollution sources. A GIS based atmospheric particulate matter concentration diffusion system was established. The phase space reconstruction was utilized to map the particle concentration sequence to the high-dimensional feature vector space to form the phase point structure training samples. Support vector machine regression was adopted to train to obtain the optimal parameters, thus enhancing the reliability of particle concentration diffusion simulation. The experimental results show that the method has high simulation accuracy, and can provide a theoretical basis for air pollution control and related research.
作者
赵逸雪
刘鑫
ZHAO Yi-xue;LIU Xin(College of Resources and Environmental Science Institute of Environmental Engineering,Ningxia University,Ningxia Yinchuan 750021,China;Beijing Normal University,Beijing 100875,China)
出处
《计算机仿真》
北大核心
2021年第5期484-487,492,共5页
Computer Simulation
关键词
大气污染
颗粒物
浓度扩散
数值模拟
GIS technology
Air pollution
Particulate matter
Concentration diffusion
Numerical simulation
