Weather situation typing method and air pollution condition prediction method and device

Abstract

The invention discloses a weather situation typing method and an air pollution condition prediction method and device. The weather situation typing method comprises the steps of selecting a typing factor of a typing region from historical meteorological data as to-be-clustered data; filtering the to-be-clustered data; clustering the to-be-clustered filtered data and establishing a parting model; iteratively updating the parting model; and parting the forecast weather situation by using the parting model. According to the weather situation typing method provided by the invention, the initial cluster center is determined in a more reasonable manner, so that clustering errors can be effectively reduced, the similarity between classes is increased, and the difference between different classesis obvious; the similarity measurement index is improved, the similarity of each row and each column is calculated, the local difference condition is fully considered, the meteorological field distribution characteristics are fitted, the parting result practicability is enhanced, and the parting result is accurate.

Description

technical field [0001] The present application relates to the technical field of environmental monitoring, in particular to a classification method for weather conditions, a method and a device for predicting air pollution conditions. Background technique [0002] Pollution events are closely related to large-scale weather conditions. The evolution of weather conditions determines the spatial distribution and time-varying characteristics of pollutants, and directly affects the duration and degree of pollution processes. The analysis and summary of the pollution weather situation is very important for understanding the characteristics of the weather system and the establishment of forecast ideas. Carrying out classification research on the pollution weather situation, identifying typical pollution weather situations, and analyzing the transmission and diffusion of pollutants under different weather types is very important for air quality early warning. Forecasting is importan...

AI Technical Summary

Problems solved by technology

[0003] In the prior art, the K-means clustering method is often used to classify the weather situation, and the random method is used to generate the initial cluster center when selecting the initial cluster center. The initial cluster center has very important significance in the subsequent clustering algorithm. The current The cluster center generated by the random method cannot represent the sample situation

The measurement method used to judge the similarity between the sample and the cluster center in the clustering algorithm considers the overall similarity between the two, while in the existing research on the classification of the weather situation, the difference in the distribution of the meteorological field in the local space is not considered. consider

Images