复杂环境下城市区域空气污染浓度实时监测研究

Research on Real-time Monitoring of UrbanAir Pollution Concentration in Complex Environment

为提升空气污染浓度实时监测效果,研究复杂环境下城市区域空气污染浓度实时监测方法。将H城市采集到的数据进行归一化处理,通过CAA算法剔除冗余和无效数据,引入深度卷积神经网络,构建空气污染浓度预测方法,对比预测数据与实时监测数据,误差超过阈值,则再次采集实时监测数据,实现复杂环境下城市区域空气污染浓度实时监测。实验测试证明:H城市冬季污染物浓度取值最高,夏季最低,在工业区域和生活区域均是上午和傍晚的污染物浓度偏高。

In order to improve the effect of real-time monitoring of air pollution concentration,the method of real-time monitoring of urban air pollution concentration in complex environment was studied.The data collected in City H was normalized,redundant and invalid data were eliminated by CAA algorithm,and a deep convolutional neural network was introduced to build an air pollution concentration prediction method.The predicted data and real-time monitoring data were compared.If the error exceeded the threshold,real-time monitoring data would be collected again to realize real-time monitoring of urban air pollution concentration in a mixed environment.Experimental tests show that the concentration of pollutants in H city is the highest in winter and the lowest in summer.In both industrial and living areas,the concentration of pollutants is high in the morning and evening.