Research On Atmospheric Data Fusion Algorithm Based On Multi-source Information

The atmospheric environment is the environment that people rely on for survival.In recent years,people have paid more and more attention to the prevention and control of atmospheric pollution.To effectively prevent and treat a few of air pollution problems,the quality of the air environment must be scientifically evaluated and analyzed before measures can be taken to address the problems.Multi-source information fusion algorithms can perform multi-level and multi-faceted optimization and the process of various types of atmospheric data complete the assessment of atmospheric environmental quality.This paper applies multi-source information fusion technology to analyses and process the atmospheric data.The main research contents are as follows:(1)Combining rough set theory and information entropy theory,an air environment quality evaluation algorithm based on variable precision rough set is proposed.By improving the variable precision rough set theory,the range of the lower approximation is increased to increase the amount of information,and then the weight is constructed and multiplied with the standardized atmospheric data to obtain a comprehensive score,and the comprehensive score of atmospheric environmental quality is compared with the reference table of atmospheric environmental quality classification.It will Obtain the atmospheric environmental quality level.The algorithm can realize the classification of atmospheric environment with different pollution levels,and further divide the atmospheric environment pollution levels under the same level.Through case simulation analysis,the algorithm has high accuracy and has certain reference value for the prevention and control of air pollution.(2)Aiming at the uncertainty of collected data in multi-source information fusion,a multi-source information air data fusion algorithm based on D-S evidence theory is proposed.Firstly,taking the characteristic value of the data obtained by various sensors,obtaining the trust function by calculating the distance between the same kind of data,and setting the threshold to eliminate the abnormal value,and the normal as well as similar data obtained are initially fused.Then,calculating the distance between the heterogeneous data and the eigenvalues of each level,calculating the support function for the obtained distance,performing the basic probability allocation,obtaining the fusion result according to the evidence theory,and finally this thesis makes a decision on the fusion result.According to the simulation analysis,this method can accurately identify the results when there is less evidence,effectively avoid the high conflict problem in DS evidence theory,and obtain the results of atmospheric data fusion decision more accurately.This algorithm is better than a few of traditional algorithms,compared with reducing the amount of calculation and reducing the complexity.