Deep neural network-based coastal region sea fog visibility forecasting method
A deep neural network and visibility technology, applied in biological neural network models, neural architectures, forecasting, etc., can solve problems such as the limitations of complex nonlinear changes, and achieve the effect of improving the level of forecasting
Pending Publication Date: 2022-05-17
青岛市气象台 +1
View PDF0 Cites 3 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Among statistical models, traditional regression models such as linear regression and stepwise regression can intuitively display the relationship between predictors and predictors, but they have limitations in solving complex nonlinear changes
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0040] The present invention will be further described below in conjunction with the accompanying drawings.
[0041] A sea fog visibility forecasting method based on a deep neural network is characterized in that the method includes the following steps: 1) establishing a forecast factor and a visibility sample set, 2) establishing and screening a visibility forecasting DNN model, and 3) forecasting sea fog visibility.
[0042] 1) Establish a sample set of predictors and visibility, such as figure 1 shown, including the following steps:
[0043] Step 1.1: Select the coastal area that needs to be forecasted, obtain the data of meteorological observation stations within this geographical range, and extract the visibility data from it, and divide it according to the standard of visibility value 4km into 5 levels, as a visibility label set;
[0044] Step 1.2: Obtain the reanalysis data within the above geographical range, and extract multiple relevant meteorological elements for ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a shore area sea fog visibility forecasting method based on a deep neural network. Establishing a visibility label set according to data of a meteorological observation station in a shore area, acquiring a plurality of forecast factor data sets from reanalysis data, and matching the forecast factor data sets with the visibility label set to establish a forecast factor and visibility sample set; establishing a deep neural network structure, and training a nonlinear relation model of the sea fog visibility in the shore area and a plurality of related forecast factors; sea fog visibility forecasting is realized through a plurality of forecasting factor data forecasted in a numerical mode. According to the method, the advantages of the deep neural network method in solving the complex nonlinear problem are utilized, the nonlinear complex relationship between a plurality of forecasting factors and visibility is fully established based on observation data and reanalysis data, and the method is applied on the basis of numerical mode forecasting, so that the method has a wide application prospect. The invention provides a new algorithm and technology for forecasting the sea fog visibility in the shore area, so as to further improve the sea fog visibility forecasting level in the shore area.
Description
technical field [0001] The invention relates to a method for forecasting sea fog visibility in coastal areas based on a deep neural network, and belongs to the field of meteorological technology. Background technique [0002] Sea fog is one of the important marine meteorological disasters. Sea fog is a weather phenomenon that is affected by the ocean and occurs on the sea, islands or coastal areas so that the horizontal visibility of the atmosphere is less than 1 km. When sea fog occurs, the horizontal visibility in the sea or coastal areas will be reduced, seriously affecting the transportation, military activities, fishery production and agricultural production activities in the sea and coastal areas, and it is also an important cause of various accidents in the sea and coastal areas. There are many sea fog areas along my country's coastal waters. Today, when maritime activities such as shipping, fishery, and production are becoming more and more frequent, the forecast of...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06N3/04G06N3/06G06Q10/04G06F16/29
CPCG06N3/061G06Q10/04G06F16/29G06N3/045
Inventor 时晓曚刘树霄
Owner 青岛市气象台