Urban air quality refined estimation method and device based on fusion of multi-source spatio-temporal data

Abstract

The invention discloses an urban air quality refined estimation method and an urban air quality refined estimation device based on the fusion of multi-source spatio-temporal data. The method achievesthe feature extraction through the fusion of multi-source spatio-temporal big data and the design of a spatio-temporal integrated feature scanning model, and avoids the separation and segmentation ofspatio-temporal attribute information. Meanwhile, by improving a deep forest algorithm, a fine-grained urban air quality real-time estimation method based on a cascade forest structure is provided, and the performance of the model is verified through specific example application. On the basis of well considering the space-time relevance of the data and not splitting the space-time attribute information, the space-time attribute information can be obtained; by means of the method, the actual performance of the estimation model is more effectively improved, the problem of space-time attribute segmentation existing in an existing method is solved, an urban microcosmic and refined data source obtaining and processing mode is effectively provided, and the defect that microscale research lacks in current air quality estimation research is overcome.

Description

technical field [0001] The present invention relates to the fields of artificial intelligence applications, atmospheric environment management and monitoring, and in particular to a method and device for finely estimating urban air quality by fusing multi-source spatio-temporal data. Background technique [0002] With the acceleration of the urbanization process, many subsequent urban problems need to be solved urgently, among which the urban air quality problem bears the brunt. However, in the ever-expanding urban space, it is difficult for the limited distribution of urban air quality monitoring stations to reflect the urban air quality status dynamically, comprehensively and in real time. Due to the needs of assisting government functional departments in making decisions and guiding public life services, it is urgent to propose methods that can estimate real-time air quality conditions in cities at the micro-scale. The most commonly used traditional methods are geostatis...

AI Technical Summary

Problems solved by technology

However, the essence of this type of method is still a linear regression method. It is difficult to fully exploit the nonlinear relationship between multiple influencing factors and air quality, and it is difficult to meet the high temporal and spatial resolution requirements of real-time estimation of fine temporal and spatial granularity.

Images