Urban Agglomerations-based Visual Analysis Of Air Quality Data

With the rapid development of China’s economy,industrialization and urban integration,the concentration of pollutants in the air is increasing and the pollution scope is gradually expanding.The smog has become one of the most serious problems in China,which has affected people’s health and travel seriously.How to solve the problem of smog in the air has become a global problem.In today’s era of big data,urban air quality data are mainly monitored and stored by the Environmental Protection Monitoring Station,which is of great guiding significance for the study and analysis of urban ambient air quality.For these massive multidimensional data,the traditional method is no longer applicable.So we need a new technology to help us deal with and analyze these data,visualization technology can meet our needs,and easy to grasp the hidden rules of data.At present,the application of visual technology is used more and more widely,which has occupied an important position.This paper designs and implements a visual analysis system based on Web platform.It is composed of three visual components,which includes a Voronoi diagram,the threads embedded on the stacked graph and parallel coordinate view.The system takes urban agglomeration as a unit,that is to say,all the cities as a whole to study the dynamic evolution of the city and the interaction between different cities of air quality,the main work of this paper are as follows:(1)The Method of Spatial Display of Urban Clustering.In this paper,the Voronoi diagram is designed to show the clustering results of the urban agglomeration.For the traditional K-Means clustering algorithm,the semantic information is added to the clustering initial center to make the clustering result more stable and better to show the spatial distribution of the urban agglomeration and the evolution in space.(2)The design of a stacked graph and parallel coordinates diagram.In this paper,a visual analysis system is proposed to explore the dynamic evolution of urban agglomeration in time and space.The system includes,a stacked graph with threads embedded to show urban agglomeration in the time evolution process,and a parallel coordinates diagram to show the pollution situation of urban agglomerations.In the design of the stacked graph with threads embedded,the drawing of the threads directly on different strips may lead to the intersection of Bézier curve.In this paper,based on the barycentre method,the strips representing urban agglomerations and the threads representing city transformation are rearranged to reduce the number of cross lines.(3)Urban agglomeration-based visual analysis of air quality data.At present,the research of air quality data is mainly focused on the study of a single city.Based on urban agglomeration,this paper makes a visual analysis of the dynamic evolution of urban agglomeration in time and space and the detailed spatial characteristics of urban agglomeration,which shows that China’s urban air pollution has a certain similarity.Therefore,the remediation of air pollution may need to consider the entire urban agglomeration rather than a single individual city,different government departments should cooperate and so on.Based on the air quality data,the system is clustered by the improved K-Means algorithm according to the similarity of urban pollution proposed in this paper,to study the urban air quality problem by the interaction between visual components.