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

Abstract

The invention discloses a method and device for finely estimating urban air quality by fusing multi-source spatio-temporal data. By fusing multi-source spatio-temporal big data, the present invention designs a spatio-temporal integrated feature scanning model for feature extraction, avoiding spatio-temporal Separation and fragmentation of attribute information. At the same time, by improving the deep forest algorithm, a real-time estimation method of fine-grained urban air quality based on the cascaded forest structure is proposed, and the performance of the invented model is verified through the application of specific examples. On the basis of taking into account the time-space correlation of data and not splitting the space-time attribute information, it can more effectively improve the actual performance of the estimation model, make up for the problem of space-time attribute fragmentation in existing methods, and effectively provide The acquisition and processing of urban microscopic and refined data sources make up for the lack of micro-scale research in the current air quality estimation research.

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