Air quality inference method based on space-time matrix decomposition

Abstract

The invention discloses an air quality inference method based on space-time matrix decomposition, and relates to the technical field of risk assessment. The method comprises the steps of 1, collecting air quality monitoring data, and obtaining a time feature matrix and a space feature matrix; and 2, designing an air quality inference model based on space-time matrix decomposition, and inferring the air quality of an unknown area on the basis of the model. According to the air quality inference method based on the space-time matrix decomposition, the air quality of an area without an air quality monitoring station in the area can be reasonably inferred according to the data of the existing air quality monitoring station in the area, and the inference result has good precision and stability.

Description

technical field [0001] The invention relates to the technical field of risk assessment, in particular to an air quality estimation method based on space-time matrix decomposition. Background technique [0002] With the acceleration of urbanization, air pollution has become an important problem that endangers human health. Therefore, it is of great significance to protect human health to obtain urban air quality maps with high spatio-temporal resolution (fine-grained), and to plan travel plans and travel modes according to the guidance of the map. [0003] At present, there are roughly four ways for people to obtain urban air quality data, 1. Monitoring by air quality monitoring stations; 2. Satellite remote sensing; 3. Detection by mobile sensors carried by people or vehicles; 4. Based on existing data , to evaluate the air quality in a certain area by calculation. [0004] Monitoring air quality through air quality monitoring stations has good results, but there are also ...

AI Technical Summary

Problems solved by technology

[0006] These calculation methods also have certain defects. For example, some methods (such as deep recurrent neural network (RNN)) rely on the data provided by a large number of air quality monitoring stations. When the number of air quality monitoring stations is insufficient, the method cannot be used. For another example, some methods (such as differential autoregressive integrated moving average (ARIMA) model) cannot make reasonable inferences for areas without air monitoring stations, and the accuracy and stability of the inferred data are not ideal

Images