Design Of Monitoring And Early Warning And Visual Information Service Platform For Heavy Metals In Cultivated Soil Based On LIBS Technology

In recent years,the area of heavy metal pollution in cultivated land soil in China has been increasing,which seriously endangers the economic income of cultivated land,people ’s health and the prosperity and stability of the country.However,the treatment of heavy metal pollution in cultivated land is very difficult,and a large amount of resource investment and long-term treatment practice are needed to find a suitable treatment plan.As an important part of heavy metal pollution control in cultivated soil,it is necessary to control heavy metal pollution in cultivated soil.However,the vast area of cultivated land contaminated by heavy metals,long-term monitoring needs,as well as the processing of large amounts of data in the detection and the prediction of heavy metal element concentrations pose new challenges to the monitoring of heavy metal pollution in cultivated soil.In this paper,according to the current stage of heavy metal pollution in cultivated land soil,the long-term monitoring of heavy metal pollution in cultivated land soil,the prediction of heavy metal pollution,the early warning of regional heavy metal pollution,the storage and processing of large amounts of data,and the visualization of heavy metal pollution information,this paper puts forward the scheme of heavy metal monitoring and early warning and visualization information service platform based on LIBS technology.The main work is as follows :Aiming at the prediction of heavy metals in cultivated soil,the standard samples were used for spectral acquisition.Based on these spectral data,the parameters of decision tree regression,LightGBM and TabNet were optimized by grid search,and the effect of the optimized model on the prediction of heavy metals in cultivated soil was compared.The root mean square error of the TabNet model was 7.91,which was significantly better than the other two algorithm models.TabNet was selected as the prediction algorithm for the prediction of heavy metals in cultivated soil.Aiming at the problem of early warning and visualization of regional heavy metal pollution,the risk screening value and risk control value proposed in the ’ Soil Environmental Quality Agricultural Land Soil Pollution Risk Control Standard(Trial)’given by the Ministry of Ecology and Environment of the People ’s Republic of China were used as the basis for early warning classification.The specific plots were warned,and the heavy metal pollution warning was carried out in the form of the Nerome index.At the same time,the effect of radial basis function interpolation algorithm and cubic spline interpolation algorithm on the spatial interpolation of heavy metal content in cultivated soil was compared.The radial basis function interpolation algorithm was selected to interpolate the heavy metal content of the undetected points in the plot,and the interpolation results were presented in the map in the form of heat map,which realized the visualization of heavy metal information in cultivated soil.Aiming at the problems of large amount of data processing and preservation and related information services generated by large-scale long-term monitoring of heavy metals in cultivated soil,a monitoring and early warning and visual information service platform for heavy metal pollution in cultivated soil was designed.The RDSMySQL database is used as a data storage and management tool,flask,gunicorn and nginx are used as the back-end framework to process the collected spectral data and accept and respond to user requests.According to the audience,the use scenario analysis was carried out,and the elements detection,equipment management,expert consultation,personal center and other pages were designed to complete the visual information service related to the monitoring and early warning of heavy metals in cultivated soil.