Urban atmospheric pollution prediction method

Abstract

The invention provides an urban atmospheric pollution prediction method which comprises the following steps: (A1) carrying out navigation in a city to obtain pollution source and pollution concentration data; carrying out grid division on the navigation area; (A2) combining the pollution concentration data with grid division to obtain pollution concentration gridding data of the city; (A3) collecting geographical and meteorological data of the city to obtain a diffusion dilution matrix T; (A4) substituting the pollution concentration gridding data and the diffusion dilution matrix into a narrow smoke cloud dilution matrix to obtain a pollution source intensity matrix with local characteristics so as to obtain pollution source intensity data Q; (A5) obtaining a pollution diffusion concentration C; and (A6) superimposing the predicted concentration after source intensity diffusion on any grid to obtain a gridded atmospheric pollution predicted concentration. The method has the advantages of accurate prediction and the like.

Description

technical field [0001] The invention relates to air pollution, in particular to a prediction method for urban air pollution. Background technique [0002] PM 2.5 It is one of the important pollutants in the atmosphere and has serious harm to the atmospheric environment and human health. Currently about PM 2.5 The monitoring technology has been relatively mature, and now the PM 2.5 Pollution prediction techniques are mainly based on air quality models (WRF-Chem, CMAQ, CAMx, etc.) or feature data machine training methods. [0003] Traditional PM 2.5 Pollution prediction methods require a large amount of monitoring data and require a large amount of investment. The prediction work based on the air quality model requires local source inventory data. Many regions in my country do not have a complete source inventory and the update period is long, resulting in poor local characteristics and timeliness of the prediction results. However, only the big data machine training met...

AI Technical Summary

Problems solved by technology

[0003] Traditional PM 2.5 Pollution prediction methods require a large amount of monitoring data and a large amount of investment

The prediction work based on the air quality model requires local source inventory data. Many regions in my country do not have a complete source inventory and the update period is long, resulting in poor local characteristics and timeliness of the prediction results.

However, only the big data machine training method is considered, and PM is not considered from the environmental point of view. 2.5 Atmospheric pollution processes from emission sources to diffusion, in predicting PM 2.5 Concentrations have limitations and pitfalls

[0004] In order to reduce the cost of environmental monitoring, increase the flexibility of environmental monitoring, and solve the pain points of lack of locality and lack of process in pollution prediction results, this patent provides an urban air pollution prediction method, such as atmospheric PM 2.5 Pollution Prediction Methods

Images