Prediction model training method and device, prediction model prediction method and device, equipment, program and medium

Abstract

The invention discloses an air quality prediction model training method and device, an air quality prediction model prediction method and device, equipment, a program and a medium, and relates to thetechnical field of computers, in particular to the deep learning technology. The method comprises the following steps: dividing a target monitoring range into a plurality of areas; pre-training an airquality prediction model by adopting a pre-training sample and a pre-training target function, the pre-training sample comprising the measured value; adopting a formal training sample and a formal training objective function to train a pre-trained air quality prediction model, wherein the formal training sample comprises the actual measurement value, and the air quality prediction model is used for predicting the air quality of the plurality of regions according to the space information, the historical information and the environment information. According to the invention, the optimized training target can be achieved by different pre-training samples and target functions.

Description

technical field [0001] This application relates to the field of artificial intelligence, in particular to deep learning technology, and specifically to a training method, prediction method, device, equipment, program and medium for an air quality prediction model. Background technique [0002] While the rapid development of urbanization has brought convenience to people's life, work, and transportation, it has also inevitably caused frequent air pollution worldwide. If the air quality in various areas of the city can be accurately predicted in advance, the potential losses caused by air pollution can be reduced or even prevented through pollution emission control and early intervention. However, due to high construction and maintenance costs, a city can only deploy a limited number of air quality monitoring stations, which cannot cover every area of ​​the city. Therefore, effective forecasting of future air quality in areas that cannot be covered by these monitoring station...

AI Technical Summary

Problems solved by technology

[0003] The prediction of air quality either relies on accurate pollution source information, or the prediction granularity is relatively rough and the accuracy is low, which cannot meet people's needs for urban fine-grained air quality prediction

Images