Ensemble forecasting method and system based on machine learning algorithm, and medium

Abstract

The invention provides an ensemble forecasting method and system based on a machine learning algorithm and a medium, and relates to the technical field of air quality forecasting, and the method comprises the steps: 1, building training data of a model according to related data including pollutant concentration and weather forecast; 2, establishing a coupling optimization model by utilizing multiple machine learning methods; and 3, taking the obtained training data as the input of the multiple machine learning method coupling optimization model, and obtaining the air quality forecast in the future time period. The influence of meteorological conditions such as temperature, humidity, wind speed, wind direction, rainfall and air pressure on the pollutant concentration can be introduced, meanwhile, multiple machine learning algorithms are coupled, and the forecasting accuracy of the air quality mode is improved.

Description

technical field [0001] The present invention relates to the technical field of air quality forecasting, in particular to an ensemble forecasting method, system and medium based on machine learning algorithms. Background technique [0002] In recent years, the serious problem of air pollution has attracted widespread attention. Reasonable air quality forecasts can help authorities make decisions to limit anthropogenic emissions while guiding the public to avoid peak pollution periods. At present, regional air quality numerical models have become the main means of short-term nowcasting and medium-term forecasting. The uncertainty of numerical prediction mainly comes from the uncertainty of the initial state of the atmosphere and the uncertainty of the forecast model. The nonlinear characteristics of the atmospheric motion determine that whether it comes from the initial field or from the model itself, the extremely small error in the model integration process will be amplifi...

AI Technical Summary

Problems solved by technology

[0004] In view of the above existing technologies, the current mainstream ensemble model algorithm can integrate the advantages and disadvantages of different single models, but it is difficult to consider the influence of meteorological conditions such as temperature, humidity, wind speed, wind direction, precipitation and air pressure on the pollutant concentration, and at the same time affect the single model The evaluation of the historical performance of the model is difficult to be comprehensive and accurate

Images