Temporal And Spatial Correlation Of Urban Air Pollutants

With the rapid development of China's economic and urbanization process,the air pollutants problems continue to be serious,due to the air pollutants generated by industrial emissions,domestic waste gas,entertainment consumption,urban population,transportation,energy consumption,etc.Air pollution not only damages the ecological environment,the physical and mental health of the people,but also has a negative impact on many aspects of the city's investment environment and economic development.The control of air pollution is of great significance to the sustainable development of cities.It has become the height of national strategic development mentioned by governments and a research hot spot for relevant enterprises,scientific research institutions and scholars.Governing air pollution requires to focus on two issues: on the one hand,to determine the source of pollutants,and on the other hand,to develop solutions to eliminate or reduce pollutants.Therefore,exploring the propagation process of air pollutants,studying the causes of pollutants and the correlation between them is the key to the research of air pollutants.However,the spread of urban air pollutants is a complex problem.Affected by weather conditions,geographical conditions and other factors,it is more difficult to study the problem of urban air pollution sources.This paper studies the spread of air pollutants between cities and their temporal and spatial correlation.The main research contents include:(1)Excavate the correlation between urban air pollutant data and propose a method for judging the effective correlation between air pollutants in surrounding cities and air pollutants in target cities.Specifically,the surrounding urban air pollutants are effectively associated with the target city air pollutants,that is,the relevant geographic data,meteorological data and air quality monitoring data meet the following three matching conditions:(a)the location and direction of the surrounding cities matching,that is,the air pollutants of the surrounding cities spread toward the target city;(b)the distance between the surrounding cities and the target city,the wind speed of the surrounding city,the time stamp of the surrounding cities data and the time stamp of the target city data(with delays)matching;c)The surrounding cities wind direction data(with time stamps)matches the target city wind direction data(with delay time stamps).(2)Three effective modes,such as rising trend,decreasing trend and steady trend of urban air pollutant time series data,are proposed.The bottom-up linear segmentation method is used to segment the air pollutant time series data,effectively filtering out the noise in original air pollutant data,and using pattern matching methods to assess the specific impact of various air pollutants in the surrounding cities and target city.In addition,the method proposed in this paper was verified by the data published by http://beijingair.sinaapp.com/ and "China Meteorological Data Network",and the analysis results of the influence of air pollutants in surrounding cities on Nanjing City are given.The contribution of this paper is to use the proposed effective correlation method of air pollutants to find out the data with high degree of correlation,and to screen out high-quality data samples to provide conditions for the training effect and prediction accuracy of the model.Compared with the existing urban air pollutant correlation research methods,the method for judging the effective correlation between the surrounding urban air pollutants and the target urban air pollutants provides a more direct and effective method for determining the associated data.This paper proposes an effective pattern matching method for the trend of air pollutant time series data to solve the problem of data timestamp unsynchronization.It is more reliable to evaluate the degree of influence of air pollutants in the surrounding cities on the target city.