Air quality monitoring station position recommendation method based on high-order graph convolutional network

Abstract

The invention discloses an air quality monitoring station position recommendation method based on a high-order graph convolutional network, and belongs to the technical field of environment detection,and the method comprises the steps: collecting external environment feature data of a whole urban area and air quality distribution data of an established observation station node area; based on theair quality distribution inference model corresponding to the urban area, predicting the air quality AQI distribution of any area where no observation station node is established by using the air quality distribution data of the established observation station node area and the external environment characteristic data of the whole urban area; and recommending a new air quality monitoring station location according to the air quality AQI distribution of any area where the observation station node is not established and an information entropy minimization greedy algorithm based on relevance between urban space-time diagram nodes. Based on the original urban monitoring equipment, a plurality of optimal station positions with practical application values is recommended to newly build the monitoring equipment, so that the precision of predicting the air quality distribution of the whole city by the air inference model can be improved to the greatest extent.

Description

technical field [0001] The invention relates to the problem of site selection and planning of observation sites in urban areas in the field of environmental detection, in particular to an air quality monitoring site location recommendation method based on a high-order graph convolution network. Background technique [0002] In recent years, with the economic growth, environmental problems have become increasingly prominent, and the air pollution problem is receiving unprecedented attention and attention. Urban air quality such as carbon monoxide (CO), carbon dioxide (CO) 2 ), hydrocarbons (HC), nitrogen oxides (NOx), and solid particulate matter (PM2.5, PM10) and other pollutant concentrations are closely related to people's health. [0003] In order to timely and accurately reflect the atmospheric environmental quality and development trend, accurate air quality monitoring equipment is required. However, the cost of these monitoring devices and subsequent maintenance cost...

AI Technical Summary

Problems solved by technology

The knowledge-driven method uses mathematical models and physical knowledge to solve the site selection problem through calculation and simulation; however, in order to achieve a steady state, the simulation process not only requires complex system programming, but also consumes a lot of computing power, which is impractical in modeling The assumption and simplification of the model will further reduce the efficiency of the model, so the research direction has turned to the data-driven model

However, these methods ignore the influence of complex external environmental characteristics such as traffic and land use on the distribution of air quality, and the research methods cannot be based on established observation sites.

Images