A method and system for short-term precipitation forecasting based on probability distribution

Abstract

The invention belongs to the technical field of precipitation forecasting, and discloses a method and system for short-term precipitation forecasting based on probability distribution. The system for short-term precipitation forecasting based on probability distribution includes: a cloud image collection module, a meteorological data collection module, a data transmission module, A main control module, a precipitation probability generation module, a precipitation estimation module, a precipitation characteristic determination module, and a data display module. The present invention can take advantage of multi-source data through the precipitation estimation module to obtain rainfall data with higher precision and more confidence; at the same time, through the precipitation characteristic determination module, the rainfall and rain pattern characteristic matrix and the time characteristic matrix of the rainfall station are used to obtain the data from the time The amount of rainfall and the movement of the rainstorm center are described in terms of time and space, so that the mathematical matrix is ​​used to quantitatively describe the characteristics of the temporal and spatial distribution of rainfall. Then, based on the mathematical matrix that quantitatively describes the rainfall of each event, the characteristics of rainfall are studied, thereby improving the Study the accuracy and comprehensiveness of rainfall characteristics.

Description

technical field [0001] The invention belongs to the technical field of precipitation forecasting, in particular to a method and system for short-term precipitation forecasting based on probability distribution. Background technique [0002] Weather forecasting (measurement) or weather forecasting (measurement) is the use of modern science and technology to predict the state of the earth's atmosphere at a certain location in the future. Weather forecasting primarily uses the collection of large amounts of data (temperature, humidity, wind direction and speed, air pressure, etc.) and then uses current understanding of atmospheric processes (meteorology) to determine future air changes. The satellite cloud images of my country's Central Meteorological Observatory are taken from meteorological satellites such as "Fengyun-1". Using satellite cloud images for analysis can improve the accuracy of weather forecasts. Weather forecasts are generally divided into three types in terms...

AI Technical Summary

Problems solved by technology

[0004] (1) The existing short-term precipitation forecast uses one type of data alone to estimate rainfall, which may easily cause deviations in the estimation results due to errors in observation data, which is not conducive to providing accurate data support for geological disaster monitoring, flood control and drought relief, etc. ; At the same time, in the actual rainfall process, the rainfall characteristics of the area where the rainfall station is located may be different from other areas of the urban area, that is, the actual rainfall process of the urban area does not match the rainfall characteristics determined by the above method, resulting in the above-mentioned The Accuracy of the Rainfall Characteristic Research Method is Poor

[0005] (2) In the existing technology, the traditional algorithm is used for denoising, which cannot completely suppress the noise, resulting in the inability to preserve and realize the characteristic peak points of the original signal, which reduces the denoising effect

[0006] (3) In the prior art, traditional algorithms are used in the image processing process of the earth cloud image, which cannot effectively avoid removing details such as edges and corners, which reduces the clarity of image details and cannot obtain reliable and accurate data

[0007] (4) In the existing technology, the traditional algorithm is used for probability forecasting, which cannot effectively improve the fitting effect, and cannot significantly improve the single-station forecasting skill of the numerical model, reducing the accuracy of the forecast

Images