Research On Visualization Techniques Of Urban Environment Sanitation Data

With the development of urban economy,people's living and consumption level has been improved,but at the same time the production of people's daily garbage is also increasing.The problem of urban garbage is becoming more and more serious,and the phenomenon of "garbage siege" becomes a trend.At present,China's sanitation industry is mainly concentrated in the construction of urban environment sanitation facilities,but is relatively lacking in sanitation information construction.This thesis will study the visualization technology of urban sanitation data,and explore the changes and trends of garbage quantity and the way of garbage disposal in recent years from many aspects,then provide the decision-making basis for the relevant departments of sanitation.The main work and achievements of the thesis are as follows:1.This thesis proposes a prediction method based on the mutual information,combined with the radial basis function network and the error reverse correction,to realize the prediction of the amount of living garbage,and then visualize the selection process of factors affecting the garbage quantity and the garbage quantity data.First of all,the mutual information is used to determine the influencing factors of garbage generation.Then the initial prediction of the garbage generation quantity is obtained through the radial basis function network.Finally,the initial prediction result is corrected by the error correction method,so the final prediction result is obtained.In this thesis,we use the method of force map to simulate the influencing factors of garbage generation,and visualize the amount of garbarge in recent years by the dynamic Choropleth map,then analyze the distribution and trend of garbage generation.The experiment results show that the method of this thesis can predict the amount of garbage well,and analyze the distribution and trend of garbage generation in China by visualization.2.This thesis presents a visual analysis scheme for garbage disposal data,which uses SOM neural network clustering algorithm and combines U matrix,parallel coordinate and Small-Multiple visualization technology.Firstly,the SOM neural network clustering algorithm is used to cluster the processing quantity data of various garbage disposal ways,and the U-matrix is used to visualize the SOM clustering results.Each clustering result are described in detail by means of parallel coordinates.According to the clustering results,Small-Multiple visualization technology is used to describe garbage distribution and distribution trends of all provinces in recent years.In order to help users to further explore the changes in waste disposal ways,multi-view linkage,brush technology and other interactive ways are provided.The experimental results show that this visualization and interactive way can help users analyze the changes in garbage disposal.3.This thesis uses the Gaussian diffusion model of point discharge source to simulate the diffusion of waste gas in waste incineration plant,and the visualization method based on the time wheel(TimeWheel)combined with the histogram is used to achieve the visualization of each pollutant index data about the waste gas in waste incineration plant.In this thesis,the diffusion model of waste gas in waste incineration plant is established by Gaussian diffusion model of point discharge source.According to the established diffusion model,the contaminant concentration contour is drawn based on the Marching Squares algorithm,and the contours are elliptic fitted.Map scatter plot is used to describe the distribution of the waste incineration plants,with different colors mapping different status of waste incineration plant.Dynamic histogram is used to visualize the real-time exhaust monitoring data,and time wheel combined with bar chart is used to visualize the exhaust gas indicators within 24 hours Of the emissions.