Urban air quality grade predicting method based on multi-field characteristics
A technology for air quality classification and prediction method, applied in the field of air quality prediction, can solve the problems of moderate pollution, heavy pollution, unsuitability, etc., to ensure global convergence, ensure super-linear convergence speed, overcome marker bias and conditional independence hypothetical effect
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Publication Date
- 2014-12-10
- Estimated Expiration
- Not applicable · inactive patent
Smart Images

Figure 1 
Figure 2 
Figure 3
Abstract
Description
technical field
[0001] The invention relates to a method for predicting city AQI grades, in particular to a method for predicting AQI grades of air quality monitoring stations based on multi-field features. Background technique
[0002] Air is a substance that the living things on the earth depend on for survival, and it is an essential substance. Ambient air quality is closely related to people's daily life, and it also plays an important role in the comprehensive evaluation of urban environment. However, with the development of human civilization and economy, air pollution is becoming more and more serious. How to improve air quality and reasonably predict and warn air environment quality is becoming more and more important. According to air quality prediction, people can take corresponding measures such as wearing masks, Try to avoid going out, etc., and protect yourself from air pollutants.
[0003] Traditional air quality prediction methods generally only consider the...
Examples
Embodiment
[0033] Embodiment: the present invention proposes the AQI grade prediction method based on multi-field feature, and flow process is as figure 1 shown. The method is divided into three stages: data preprocessing, training and prediction. Among them, the data flow in the preprocessing stage is represented by a dotted line, the data flow in the training stage is represented by a dotted line, and the data flow in the prediction stage is represented by a solid line.
[0034] The process of data preprocessing stage is as follows: figure 2 As shown, its main steps include:
[0035] 1) For a certain city, collect historical and real-time data and meteorological forecast data for a certain period of time in the future in multiple fields that affect air quality, such as meteorology, traffic, air pollutants, etc.;
[0036] 2) Divide city a into disjoint grids, each grid g=g.wxg.h has the same length g.w and width g.h, use g. c Indicates the center point of the grid g. use g a (w, ...