基于智能算法的大气污染防控知识图谱：研究方法、演化路径与应用展望

大气污染防控是保障人民健康与生命安全重要的研究课题，智能算法是大气污染防控及预测预警的重要工具手段。通过Citespace对Web of Science核心数据库中1998—2021年智能算法在大气污染防控中应用的文献进行可视化分析。结果表明：(1)智能算法在大气污染防控应用历经“数理统计-遥感监测-数字模拟-神经网络算法”四个阶段；(2)研究主题集聚为“神经网络模型”“智慧健康检测系统”和“城市空间的大气污染”三个核心聚类演变路径；(3)应用方式包括“大气污染时空分布”“大气污染因素相关性”“大气污染事件预测”“空气质量优劣评估”与“健康韧性城市应用”五个方面。未来智能算法在大气污染防控中的应用趋势，将体现“高精度预测-全周期评估-健康化理念”的特点。

Knowledge Graph of Air Pollution Control Based on Intelligent Algorithms:Research Methods,Evolution Paths and Application Prospects

Air pollution prevention and control is an important research topic to protect people's health and life safety.Intelligent algorithm is an important tool for air pollution prevention and control,prediction and early warning.In this paper,Citespace is used to make a visual analysis of the literature on the application of intelligent algorithm in air pollution prevention and control from 1998 to 2021 in the core database of Web of Science.The results show that:(1)the application history of intelligent algorithm in air pollution prevention and control includes four stages:mathematical statistics,remote sensing monitoring,digital simulation and neural network algorithm.(2)The research themes are clustered into three core cluster evolution paths:"neural network model","smart health detection system"and"air pollution in urban space";(3)The application methods include five aspects:"spatial and temporal distribution of air pollution","correlation of air pollution factors","prediction of air pollution events","evaluation of air quality"and"application of healthy and resilient cities".In the future,the application trend of intelligent algorithms in air pollution prevention and control will reflect the characteristics of"high precision prediction-full cycle assessment-health concept".