摘要
城市环境空气污染监测与预警性能过差会加快有毒有害物质扩散速度,为了在环境空气进一步恶化前及时采取严格的防治措施,提出基于强化深度学习的城市环境空气污染监测与预警方法。分析气候因素和气象因素,利用气体传感器采集条件感知空气中污染源类型和污染物含量,实现空气污染数据的采集。通过补充缺失数据、清洗失真数据优化原始空气污染数据,并从时间变化和空间变化两种角度提取空气污染数据特征,输入强化深度学习模型,实现城市环境空气污染监测与预警。实验结果表明,所提方法监测与预警效果好。
Poor monitoring and early warning performance of urban environmental air pollution will accelerate the diffusion of toxic and harmful substances.In order to take strict prevention measures in time before the environmental air deteriorates further,a monitoring and early warning method of urban environmental air pollution based on reinforcement deep learning is proposed.By analyzing the climatic factors and meteorological factors,this paper used gas sensors to perceive the types of pollution sources and pollutant content in the air according to the collection conditions so as to realize the collection of air pollution data.By supplementing missing data and cleaning distorted data,the original air pollution data is optimized,and the characteristics of air pollution data are extracted from both temporal and spatial changes,and input into the reinforcement deep learning model to realize urban environmental air pollution monitoring and early warning.The experimental results show that the proposed method has good monitoring and early warning effect.
作者
马吉伟
王靖宇
谢勇
李田
姚志平
Ma Jiwei;Wang Jingyu;Xie Yong;Li Tian;Yao Zhiping(Jilin Meteorological Service Center,Changchun 130062,China;Nanjing University of Information Science and Technology,Nanjing 210000,China;Jilin Emergency Warning Information Dissemination Center,Changchun 130062,China)
出处
《环境科学与管理》
CAS
2023年第10期116-120,共5页
Environmental Science and Management
基金
吉林省科技发展计划项目(20220203196SF)。
