Marine weather forecast data processing method and system and electronic equipment

Abstract

The invention provides a marine weather forecast data processing method, a marine weather forecast data processing system and electronic equipment, and relates to the technical field of weather forecast. The data set and the meteorological historical data are fused and then are used for the generation and updating process of a prediction model, so that the weather forecast data are processed by fully utilizing the time characteristics and the space characteristics of the data under the limited data condition, and the dynamic change of meteorological variables is fully captured; and the accuracy of marine weather forecast is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of weather forecasting, and in particular to a method, system and electronic equipment for processing marine weather forecast data. Background Art

[0002] In the process of marine weather forecasting, due to the limitation of marine environmental factors, it is impossible to set up many fixed observation sites, so non-fixed observation points such as ships are needed to obtain and predict marine weather data. Since the amount of such ship observation data is small and the real-time performance of ship observation data is poor, it is difficult to obtain marine weather data in a relatively wide time period. There is a problem of poor quality of meteorological data, which makes it impossible to fully capture the dynamic changes of meteorological variables, resulting in low accuracy of marine weather forecasts. Summary of the invention

[0003] In view of this, the purpose of the present invention is to provide a method, system and electronic equipment for processing marine weather forecast data. The method constructs a new data set by utilizing the difference data between the forecast data and the analysis data contained in the limited historical meteorological data, and fuses the data set with the historical meteorological data for use in the generation and update process of the prediction model, thereby making full use of the temporal and spatial characteristics of the data to process the meteorological forecast data under limited data conditions, thereby fully capturing the dynamic changes of meteorological variables and improving the accuracy of marine weather forecasts.

[0004] In a first aspect, an embodiment of the present invention provides a method for processing marine weather forecast data, the method comprising: Obtain ship observation data corresponding to the observation point based on the position data of the observation point, and use the ship observation data to obtain weather forecast historical data and weather analysis historical data corresponding to the observation point; Obtaining random points corresponding to the observation points, constructing an observation area corresponding to the observation points using the random points, and constructing a first meteorological data set using meteorological forecast historical data and meteorological analysis historical data contained in the observation area; Determine the difference data between the meteorological forecast historical data and the meteorological analysis historical data in the observation area at the same historical moment, and use the difference data to construct a second meteorological data set corresponding to the observation area; Constructing a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set, and updating the prediction model using the difference data in the second meteorological data set; Obtain the weather forecast data corresponding to the observation point at the current moment, and use the updated prediction model to obtain the weather analysis data corresponding to the weather forecast data.

[0005] Optionally, obtaining ship observation data corresponding to the observation point based on the position data of the observation point, and using the ship observation data to obtain weather forecast historical data and weather analysis historical data corresponding to the observation point, including: Determine the ship observation data collected by the meteorological ship at the observation point based on the position data determined at the observation point; Determine the ocean area corresponding to the meteorological ship and the observation point, and obtain the corresponding meteorological historical data in the ocean area; Based on the historical meteorological data, the meteorological forecast historical data and the meteorological analysis historical data are determined.

[0006] Optionally, obtaining a random point corresponding to the observation point, using the random point to construct an observation area corresponding to the observation point, and using the meteorological forecast historical data and meteorological analysis historical data contained in the observation area to construct a first meteorological data set includes: Acquire multiple random points corresponding to the observation points in the ocean area between the meteorological ship and the observation point; The location data of multiple random points are used to determine the boundary of the area corresponding to the observation point, and the observation area corresponding to the observation point is constructed using the area boundary; Obtain historical weather forecast data and historical weather analysis data corresponding to observation points and random points in the observation area; The first meteorological data set is constructed using historical meteorological forecast data and historical meteorological analysis data.

[0007] Optionally, determining difference data between meteorological forecast historical data and meteorological analysis historical data in the observation area at the same historical moment, and constructing a second meteorological data set corresponding to the observation area using the difference data, includes: Obtaining meteorological forecast historical data and meteorological analysis historical data corresponding to the observation area at the same historical moment, and determining the meteorological forecast value and meteorological observation value corresponding to the observation area based on the meteorological forecast historical data and meteorological analysis historical data; The difference data is determined by using the meteorological forecast value and the meteorological observation value, and after the difference data is transferred to the grid field corresponding to the observation area by using the inverse distance weighted interpolation method, the error field corresponding to the observation area is generated by using the grid field; After the first meteorological data set is updated based on the error field, the second meteorological data set is constructed according to the updated first meteorological data set.

[0008] Optionally, constructing a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set includes: Acquire time series feature data corresponding to the observation area in the first meteorological data set, and construct a time series convolution module using the time series feature data; Acquire scale feature data corresponding to the observation area in the first meteorological data set, and construct a compression excitation network module using the scale feature data; The prediction model corresponding to the observation area is constructed using the temporal convolution module and the compressed excitation network module.

[0009] Optionally, a prediction model corresponding to the observation area is constructed using a temporal convolution module and a compression excitation network module, including: The temporal convolution module is used to construct a temporal convolution network corresponding to the temporal feature data, and the compressed excitation network module is used to construct a compressed excitation network corresponding to the scale feature data; A prediction model corresponding to the observation area is constructed based on the time series convolutional network and the compressed excitation network; wherein the prediction model is trained using historical weather forecast data and historical weather analysis data.

[0010] Optionally, updating the prediction model using the difference data in the second meteorological data set includes: constructing revised forecast data corresponding to the observation area using the difference data in the second meteorological data set; The revised forecast data is used to update the meteorological analysis historical data, and the updated meteorological analysis historical data and meteorological forecast historical data are used to update the prediction model.

[0011] Optionally, obtain the meteorological forecast data corresponding to the observation point at the current moment, and use the updated prediction model to obtain the meteorological analysis data corresponding to the meteorological forecast data, including: Obtaining the meteorological forecast data corresponding to the observation point at the current moment, and using the meteorological forecast data to update the first meteorological data set and the second meteorological data set; After updating the prediction model using the updated first meteorological data set and the second meteorological data set, obtaining the time series characteristic data and the scale characteristic data corresponding to the meteorological forecast data according to the updated prediction model; The meteorological analysis data corresponding to the meteorological forecast data is determined based on the time series characteristic data and scale characteristic data.

[0012] In a second aspect, the present invention provides a marine weather forecast data processing system, the system comprising: A data acquisition unit, used to obtain ship observation data corresponding to the observation point based on the position data of the observation point, and to obtain weather forecast historical data and weather analysis historical data corresponding to the observation point using the ship observation data; A first meteorological data set generating unit, configured to obtain a random point corresponding to the observation point, construct an observation area corresponding to the observation point using the random point, and construct a first meteorological data set using the meteorological forecast historical data and meteorological analysis historical data contained in the observation area; A second meteorological data set generating unit is used to determine the difference data between the meteorological forecast historical data and the meteorological analysis historical data in the observation area at the same historical moment, and to construct a second meteorological data set corresponding to the observation area using the difference data; A prediction model building unit, used to build a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set, and update the prediction model using the difference data in the second meteorological data set; The weather forecast data processing unit is used to obtain the weather forecast data corresponding to the observation point at the current moment, and to obtain the weather analysis data corresponding to the weather forecast data using the updated prediction model.

[0013] In a third aspect, an embodiment of the present invention further provides an electronic device, including a processor and a memory, wherein the memory stores computer-executable instructions that can be executed by the processor, and the processor executes the computer-executable instructions to implement the steps of the marine weather forecast data processing method provided in the first aspect.

[0014] In a fourth aspect, an embodiment of the present invention further provides a storage medium storing computer executable instructions. When the computer executable instructions are called and executed by a processor, the computer executable instructions prompt the processor to implement the steps of the marine weather forecast data processing method provided in the first aspect.

[0015] An embodiment of the present invention provides a method, system and electronic device for processing marine weather forecast data. In the process of processing marine weather forecast data, the method first obtains ship observation data corresponding to the observation point based on the position data of the observation point, and uses the ship observation data to obtain weather forecast historical data and weather analysis historical data corresponding to the observation point; then obtains a random point corresponding to the observation point, uses the random point to construct an observation area corresponding to the observation point, and uses the weather forecast historical data and weather analysis historical data contained in the observation area to construct a first meteorological data set; then determines the difference data between the weather forecast historical data and the weather analysis historical data in the observation area at the same historical moment, and uses the difference data to construct a second meteorological data set corresponding to the observation area; then constructs a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set, and uses the difference data in the second meteorological data set to update the prediction model; finally, obtains the weather forecast data corresponding to the observation point at the current moment, and uses the updated prediction model to obtain the meteorological analysis data corresponding to the weather forecast data. This method uses the difference data between the forecast data and the analysis data contained in the limited historical meteorological data to construct a new data set, and fuses the data set with the historical meteorological data for the generation and update process of the prediction model. It makes full use of the temporal and spatial characteristics of the data to process the meteorological forecast data under limited data conditions, thereby fully capturing the dynamic changes of meteorological variables and improving the accuracy of marine meteorological forecasts.

[0016] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention are realized and obtained by the structures particularly pointed out in the description, claims and drawings.

[0017] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, preferred embodiments are given below and described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the specific implementation methods of the present invention or the technical solutions in the prior art, the drawings required for use in the specific implementation methods or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some implementation methods of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0019] Figure 1 A flow chart of a method for processing marine weather forecast data provided by an embodiment of the present invention; Figure 2A flowchart of step S101 in a method for processing marine weather forecast data provided by an embodiment of the present invention; Figure 3 A flowchart of step S102 in a method for processing marine weather forecast data provided by an embodiment of the present invention; Figure 4 A flowchart of step S103 in a method for processing marine weather forecast data provided by an embodiment of the present invention; Figure 5 A flowchart of step S103 of a method for processing marine weather forecast data provided by an embodiment of the present invention, in which a prediction model corresponding to the observation area is constructed according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set; Figure 6 A flowchart of step S503 in a method for processing marine weather forecast data provided by an embodiment of the present invention; Figure 7 A flowchart of updating the prediction model using the difference data in the second meteorological data set in step S103 of a method for processing marine meteorological forecast data provided by an embodiment of the present invention; Figure 8 A flowchart of step S105 in a method for processing marine weather forecast data provided by an embodiment of the present invention; Fig. 9 A flowchart of another method for processing marine weather forecast data provided by an embodiment of the present invention; Fig.10 A schematic diagram of the structure of a marine weather forecast data processing system provided by an embodiment of the present invention; Fig.11 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention.

[0020] icon: 1010 - data acquisition unit; 1020 - first meteorological data set generation unit; 1030 - second meteorological data set generation unit; 1040 - prediction model construction unit; 1050 - meteorological forecast data processing unit; 101 - processor; 102 - memory; 103 - bus; 104 - communication interface. DETAILED DESCRIPTION

[0021] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be clearly and completely described in combination with the embodiments below. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0022] In the process of marine weather forecasting, due to the limitation of marine environmental factors, it is impossible to set up more fixed observation sites, so non-fixed observation points such as ships are needed to obtain and predict marine meteorological data. Since the amount of such ship observation data is small and the real-time performance of ship observation data is poor, it is difficult to obtain marine meteorological data within a relatively wide time period, and there is a problem of poor quality of meteorological data, which makes it impossible to fully capture the dynamic changes of meteorological variables, resulting in low accuracy of marine weather forecasts. Based on this, the present invention implements a method, system and electronic device for processing marine weather forecast data, which uses the difference data between the forecast data and the analysis data contained in the limited meteorological historical data to construct a new data set, and fuses the data set with the meteorological historical data for the generation and update process of the prediction model, so as to make full use of the time characteristics and spatial characteristics of the data under limited data conditions to process the meteorological forecast data, thereby fully capturing the dynamic changes of meteorological variables and improving the accuracy of marine weather forecasts.

[0023] To facilitate understanding of this embodiment, firstly, a method for processing marine weather forecast data disclosed in an embodiment of the present invention is described in detail. Figure 1 As shown, including: Step S101, obtaining ship observation data corresponding to the observation point based on the position data of the observation point, and using the ship observation data to obtain weather forecast historical data and weather analysis historical data corresponding to the observation point.

[0024] In the process of marine weather forecasting, the location data of the observation point is the starting key information of the entire process. In layman's terms, the observation point can be a non-fixed observation point such as a ship, or a fixed observation point such as an island in the ocean, oil extraction equipment, etc. The location data can be information such as longitude and latitude coordinates obtained through a satellite positioning system (such as GPS, Beidou, etc.). Based on these location data, accurate matching and screening can be performed in the relevant database of ship observation data to obtain the ship observation data corresponding to the observation point at different time points. The ship observation data contains various types of meteorological historical data. Using the location data of the observation point as an index, accurate retrieval can be performed in the relevant database of the ship observation data to obtain the meteorological forecast historical data related to the observation point (such as weather forecast information for the observation point released at different historical times, including temperature, precipitation probability, etc.) and meteorological analysis historical data (such as an analysis report on the past meteorological conditions of the observation point, including climate trend analysis, etc.).

[0025] Step S102, obtaining random points corresponding to the observation points, using the random points to construct the observation area corresponding to the observation points, and using the meteorological forecast historical data and meteorological analysis historical data contained in the observation area to construct a first meteorological data set.

[0026] When obtaining random points corresponding to observation points, a specific random algorithm will be used to obtain several random points within a certain range around the observation points. The distribution range and number of these random points can be set according to actual needs and accuracy requirements. Through the range contained in these random points, an observation area containing the observation point and all random points is constructed. The observation area can be an irregular polygonal area or a specific geometric shape as required. After the observation area is constructed, all historical meteorological forecast data and meteorological analysis data in the observation area will be collected and sorted, and duplicate and invalid data records will be removed. The valid data will be organized according to certain formats and standards to construct the first meteorological data set. This data set contains various meteorological-related information in the observation area over a period of time in the past, providing basic data support for subsequent analysis and modeling.

[0027] Step S103, determining the difference data between the meteorological forecast historical data and the meteorological analysis historical data in the observation area at the same historical moment, and constructing a second meteorological data set corresponding to the observation area using the difference data.

[0028] Since there are few ship observation data in the observation point, it is necessary to expand it, and by determining the difference data between the weather forecast historical data and the weather analysis historical data in the observation area at the same historical moment, the second meteorological data set corresponding to the observation area is constructed. Specifically, after the first meteorological data set is constructed, the weather forecast historical data and the weather analysis historical data in the first meteorological data set are compared and analyzed moment by moment to obtain the difference data corresponding to the two. For example, the difference between the wind speed predicted in the weather forecast at a certain moment and the wind speed recorded in the actual weather analysis is compared, and the difference is summarized, collected and sorted, so as to construct the second meteorological data set corresponding to the observation area.

[0029] Step S104, constructing a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set, and updating the prediction model using the difference data in the second meteorological data set.

[0030] During the implementation of this step, the data in the first meteorological data set will be deeply analyzed to extract the time series feature data (such as the changing trend of meteorological elements over time, periodic laws, etc.) and scale feature data (such as the distribution characteristics of meteorological elements at different spatial scales, etc.) corresponding to the observation area. In actual scenarios, the corresponding modeling algorithm (such as the neural network algorithm in machine learning, the time series analysis algorithm, etc.) is selected based on these feature data to build a prediction model corresponding to the observation area. The prediction model can predict future meteorological conditions based on the input meteorological related information. Then, the difference data in the second meteorological data set is used to update and optimize the constructed prediction model. By inputting the difference data as feedback information into the prediction model, the parameters and structure of the model are adjusted, so that the prediction model can better adapt to the actual meteorological conditions and improve the accuracy and reliability of the prediction.

[0031] Step S105, obtaining the meteorological forecast data corresponding to the observation point at the current moment, and using the updated prediction model to obtain the meteorological analysis data corresponding to the meteorological forecast data.

[0032] It can be seen that this method uses the difference data between the forecast data and the analysis data contained in the limited historical meteorological data to construct a new data set, and fuses the data set with the historical meteorological data for the generation and update process of the prediction model. It realizes the full use of the temporal and spatial characteristics of the data to process the meteorological forecast data under limited data conditions, thereby fully capturing the dynamic changes of meteorological variables and improving the accuracy of marine meteorological forecasts.

[0033] Optionally, step S101 of acquiring ship observation data corresponding to the observation point based on the position data of the observation point, and using the ship observation data to acquire weather forecast historical data and weather analysis historical data corresponding to the observation point, such as Figure 2 As shown, including: Step S201, determining ship observation data collected by a meteorological ship at the observation point according to the position data determined at the observation point; Step S202, determining the ocean area corresponding to the meteorological ship and the observation point, and obtaining corresponding meteorological historical data in the ocean area; Step S203, determining weather forecast historical data and weather analysis historical data based on the weather historical data.

[0034] In a specific implementation process, the ship observation data collected by the meteorological ship is determined by the position data corresponding to the observation point, and then the meteorological historical data in the ocean area between the meteorological ship and the observation point is collected, and the meteorological forecast historical data and meteorological analysis historical data contained in the meteorological historical data are used to construct the subsequent first meteorological data set. Optionally, a random point corresponding to the observation point is obtained, the observation area corresponding to the observation point is constructed using the random point, and the first meteorological data set is constructed using the meteorological forecast historical data and meteorological analysis historical data contained in the observation area. Step S102, such as Figure 3 As shown, including: Step S301, obtaining a plurality of random points corresponding to the observation point in the ocean area between the meteorological ship and the observation point; Step S302, using the position data of multiple random points to determine the boundary of the area corresponding to the observation point, and using the area boundary to construct the observation area corresponding to the observation point; Step S303, obtaining weather forecast historical data and weather analysis historical data corresponding to observation points and random points in the observation area; Step S304: construct a first meteorological data set using meteorological forecast historical data and meteorological analysis historical data.

[0035] Specifically, firstly, a plurality of random points are obtained in the ocean area between the meteorological ship and the observation point, and the region boundary corresponding to the observation point is determined using the position data of the random point, so as to construct the observation area corresponding to the observation point using these region boundaries. Then, the meteorological forecast historical data and meteorological analysis historical data corresponding to the observation point and the random point in the observation area are intercepted to construct the first meteorological data set.

[0036] Optionally, the difference data between the meteorological forecast historical data and the meteorological analysis historical data in the observation area at the same historical moment is determined, and the difference data is used to construct a second meteorological data set corresponding to the observation area S103, such as Figure 4 As shown, including: Step S401, obtaining weather forecast historical data and weather analysis historical data corresponding to the observation area at the same historical moment, and determining weather forecast values ​​and weather observation values ​​corresponding to the observation area based on the weather forecast historical data and weather analysis historical data; Step S402, using the meteorological forecast value and the meteorological observation value to determine the difference data, and using the inverse distance weighted interpolation method to transfer the difference data to the grid field corresponding to the observation area, and then using the grid field to generate the error field corresponding to the observation area; Step S403: after updating the first meteorological data set based on the error field, constructing a second meteorological data set according to the updated first meteorological data set.

[0037] The acquisition process of the second meteorological data set is determined based on the difference data between the meteorological forecast historical data and the meteorological analysis historical data in the observation area at the same time. First, the meteorological forecast value and meteorological observation value corresponding to the observation area are determined based on the meteorological forecast historical data and the meteorological analysis historical data. After the difference data is determined using the meteorological forecast value and the meteorological observation value, the difference data is transferred to the grid field corresponding to the observation area using the inverse distance weighted interpolation method, and the error field corresponding to the observation area is generated using the grid field. Finally, after updating the first meteorological data set based on the error field, the second meteorological data set is constructed based on the updated first meteorological data set.

[0038] Optionally, a prediction model corresponding to the observation area is constructed according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set, such as Figure 5 As shown, including: Step S501, obtaining time series feature data corresponding to the observation area in the first meteorological data set, and constructing a time series convolution module using the time series feature data; Step S502, obtaining scale feature data corresponding to the observation area in the first meteorological data set, and constructing a compression excitation network module using the scale feature data; Step S503, using the temporal convolution module and the compression excitation network module to construct a prediction model corresponding to the observation area.

[0039] The existing correction algorithms based on fixed sites and machine learning mainly use models based on lightgbm or random forest models, and use various observation data and static environmental feature data of the site for a continuous period of time as input to train the relevant machine learning model to obtain the corresponding prediction model. Since the environment at sea is more complex than that on land, it is difficult to obtain enough fixed observation points, and there is a lack of observation data that is continuous in time and fixed in position to train the model. In this embodiment, the time series feature data corresponding to the observation area in the first meteorological data set is used to construct a time series convolution module, and then the scale feature data corresponding to the observation area in the first meteorological data set is used to construct a compression excitation network module.

[0040] Optionally, step S503 of constructing a prediction model corresponding to the observation area using the temporal convolution module and the compression excitation network module, such as Figure 6 As shown, including: Step S601, using a temporal convolution module to construct a temporal convolution network corresponding to the temporal feature data, and using a compression excitation network module to construct a compression excitation network corresponding to the scale feature data; Step S602, constructing a prediction model corresponding to the observation area according to the temporal convolutional network and the compression excitation network; wherein the prediction model is trained using historical weather forecast data and historical weather analysis data.

[0041] The temporal convolutional network can utilize the 3DTCN (Temporal Convolutional Network) part: it is mainly responsible for capturing the temporal dependencies in the input data. In order to enhance the temporal modeling capability of the model, dilated convolution is introduced in TCN. Dilated convolution can capture a wider range of temporal features without increasing model parameters by expanding the receptive field of the convolution kernel, thereby effectively modeling long-term dependencies. In 3DTCN, by stacking multiple layers of dilated convolution, each layer uses a different dilation rate (such as 1, 2, 4, 8, etc.), so that the model can capture dependencies over a longer time span with fewer layers. In addition, in order to ensure that the prediction results do not rely on future information, 3DTCN uses causal convolution, that is, the convolution operation of each time step only considers the current and past time steps to ensure that the model output is consistent with the time series of the actual weather forecast. In order to improve the training effect of the model, residual connections are introduced between each layer of dilated convolution. This design can effectively transfer the gradient of the deep convolutional network, avoid the common gradient vanishing problem of deep networks, and thus stabilize the training process of the model.

[0042] The SEU-Net network structure can be used to extract multi-scale spatial information from meteorological data. SEU-Net is based on the U-Net structure and combines the Squeeze-and-Excitation (SE) module to further enhance the ability to dynamically adjust the importance of feature channels. The encoder part of SEU-Net consists of a series of convolutional layers and pooling layers. After each convolution operation, the resolution of the feature map is reduced by max pooling to gradually extract spatial features of different scales. In this process, the SE module is inserted after each encoder layer to compress the feature map output by the convolutional layer and dynamically adjust the importance of different channels through the activation function.

[0043] Optionally, the prediction model is updated using the difference data in the second meteorological data set, such as Figure 7 As shown, including: Step S701, constructing revised forecast data corresponding to the observation area using the difference data in the second meteorological data set; Step S702, using the revised forecast data to update the meteorological analysis historical data, and using the updated meteorological analysis historical data and meteorological forecast historical data to update the prediction model.

[0044] The updating process of the prediction model can be implemented in combination with the training process of the model. Specifically, the first meteorological data set is used as a pre-training data set to train the TCN + SEU-Net correction model parameters. The input of the model is the meteorological forecast data of the previous moment, the meteorological analysis data of the previous moment, and the forecast data of the current moment. The output of the model is the meteorological analysis data of the current moment. Subsequently, the forecast data corrected by the error field generated by the observation value and forecast of the mobile observation point is used to replace the meteorological analysis data to obtain the second meteorological data set as the fine-tuning training data set, so as to use the corrected forecast data to update the meteorological analysis historical data, and use the updated meteorological analysis historical data and meteorological forecast historical data to update the prediction model.

[0045] Optionally, the step S105 of obtaining the meteorological forecast data corresponding to the observation point at the current moment and obtaining the meteorological analysis data corresponding to the meteorological forecast data using the updated prediction model is as follows: Figure 8 As shown, including: Step S801, obtaining the meteorological forecast data corresponding to the observation point at the current moment, and using the meteorological forecast data to update the first meteorological data set and the second meteorological data set; Step S802, after updating the prediction model using the updated first meteorological data set and the second meteorological data set, obtaining the time series feature data and the scale feature data corresponding to the meteorological forecast data according to the updated prediction model; Step S803, determining meteorological analysis data corresponding to the meteorological forecast data according to the time series characteristic data and the scale characteristic data.

[0046] In the actual implementation process, in the process of processing the offshore weather forecast data, the weather forecast data corresponding to the observation point at the current moment is obtained, and the first weather data set and the second weather data set are updated using the weather forecast data. Then, the updated first weather data set and the second weather data set are used to update the prediction model, and then the time series feature data and scale feature data corresponding to the weather forecast data are obtained according to the updated prediction model, and then the meteorological analysis data corresponding to the weather forecast data is determined according to the time series feature data and the scale feature data.

[0047] like Fig. 9 The flowchart of another method for processing marine weather forecast data shown in FIG. 1 mainly involves six steps, which are as follows: Step S1, collects reanalysis data of global meteorological elements, historical weather forecast data, and collects ship observation data for a certain range of sea areas.

[0048] Step S2, based on the observation point and the random points around the observation point generated based on the observation point, calculate the local area boundary centered on the observation point, intercept the corresponding reanalysis data and historical forecast data, and construct a pre-training data set.

[0049] Step S3, based on the position of the observation point, calculate the boundary of the local area centered on the observation point, match the corresponding meteorological forecast data, use the inverse distance weighted interpolation method to transfer the error between the observation value and the forecast value to the local area to form an error field, correct the forecast at the corresponding time, and construct a fine-tuning training data set.

[0050] Step S4, construct a weather forecast correction model based on 3DTCN + SEU-Net, in which the 3DTCN part combines the temporal convolutional network (TCN) and the U-Net structure with the Squeeze-and-Excitation (SE) module to achieve effective modeling and correction of the spatiotemporal characteristics of meteorological data. The 3DTCN (Temporal Convolutional Network) part is mainly responsible for capturing the temporal dependencies in the input data. In order to enhance the temporal modeling ability of the model, dilated convolution is introduced in TCN. Dilated convolution can capture a wider range of temporal features without increasing model parameters by expanding the receptive field of the convolution kernel, thereby effectively modeling long-term dependencies. In 3DTCN, by stacking multiple layers of dilated convolutions, each layer uses a different dilation rate (such as 1, 2, 4, 8, etc.), so that the model can capture dependencies with a longer time span with fewer layers. In addition, to ensure that the prediction results do not rely on future information, 3DTCN uses causal convolution, that is, the convolution operation of each time step only considers the current and past time steps to ensure that the model output is consistent with the time series of the actual weather forecast. In order to improve the training effect of the model, residual connections are also introduced between each layer of dilated convolution. This design can effectively transfer the gradient of the deep convolution network, avoid the common gradient vanishing problem of the deep network, and thus stabilize the training process of the model. The SEU-Net part is responsible for extracting multi-scale spatial information from meteorological data. SEU-Net is based on the classic U-Net structure and combines the Squeeze-and-Excitation (SE) module to further enhance the ability to dynamically adjust the importance of feature channels. The encoder part of SEU-Net consists of a series of convolutional layers and pooling layers. After each convolution operation, the resolution of the feature map is reduced by max pooling to gradually extract spatial features of different scales. In this process, the SE module is inserted after each encoder layer to compress the feature map output by the convolutional layer and dynamically adjust the importance of different channels through the activation function.

[0051] Step S5: During the model training process, a pre-training data set consisting of reanalysis data and historical forecast data is used to train TCN + SEU-Net to revise the model parameters. The input of the model is the weather forecast of the previous moment, the reanalysis data of the previous moment, and the forecast data of the current moment. The output of the model is the reanalysis data of the current moment. After the pre-training network, the forecast data corrected by the error field generated by the observation value of the mobile observation point and the forecast is used to replace the fine-tuning training data set consisting of the reanalysis data, and finally the pre-trained revised model is obtained.

[0052] Step S6: Obtain the ship observation data and forecast data of the previous moment, generate the forecast correction data of the previous moment using the method of step S3, input the forecast data and correction data of the previous moment and the forecast data of the current moment into the 3DTCN+SEU-Net correction model trained in step S5, and output the corrected forecast of the local area around the observation point at the current moment.

[0053] The above method transfers the observation value of the ship observation point to the grid field through the inverse distance weight, distributes the correction difference to a local space, obtains a difference field, and superimposes it on the forecast field of the original space to replace the actual data at the nearby moment to correct the forecast data for the next moment. In addition, this method constructs a 3DTCN + SE-U-Net structure, adds an SE module to the original U-Net encoder and decoder, strengthens the spatial feature extraction, and adds a 3DTCN model in front of U-Net to learn the temporal characteristics of continuous forecasts, so that the entire model has a strong ability to learn temporal and spatial features. Therefore, there is no need to obtain the reanalysis data of the current nearby moment to guide the correction of the forecast data, and the correction of the meteorological forecast based only on the mobile observation point is realized; this method does not need to obtain the reanalysis data of the current nearby moment to guide the correction of the forecast data, and the correction of the meteorological forecast based only on the mobile observation point is realized; From the marine weather forecast data processing method mentioned in the above embodiment, it can be seen that the method constructs a new data set in the limited meteorological historical data by using the difference data between the forecast data and the analysis data contained therein, and fuses the data set with the meteorological historical data for use in the generation and update process of the prediction model, thereby making full use of the time and spatial characteristics of the data to process the meteorological forecast data under limited data conditions, thereby fully capturing the dynamic changes of meteorological variables, and correcting the local sea area around the observation point without relying on a large number of fixed observation points. It is more suitable for marine weather forecast correction scenarios and improves the accuracy of marine weather forecasts.

[0054] Corresponding to the offshore weather forecast data processing method provided in the above-mentioned embodiment, the embodiment of the present invention provides an offshore weather forecast data processing system, such as Fig.10 As shown, the system includes: The data acquisition unit 1010 is used to obtain the ship observation data corresponding to the observation point based on the position data of the observation point, and to obtain the weather forecast historical data and weather analysis historical data corresponding to the observation point using the ship observation data; The first meteorological data set generating unit 1020 is used to obtain random points corresponding to the observation points, construct the observation area corresponding to the observation points using the random points, and construct the first meteorological data set using the meteorological forecast historical data and meteorological analysis historical data contained in the observation area; The second meteorological data set generating unit 1030 is used to determine the difference data between the meteorological forecast historical data and the meteorological analysis historical data in the observation area at the same historical moment, and construct a second meteorological data set corresponding to the observation area using the difference data; A prediction model building unit 1040 is used to build a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set, and update the prediction model using the difference data in the second meteorological data set; The weather forecast data processing unit 1050 is used to obtain the weather forecast data corresponding to the observation point at the current moment, and to obtain the weather analysis data corresponding to the weather forecast data using the updated prediction model.

[0055] From the marine weather forecast data processing system mentioned in the above embodiment, it can be seen that the system constructs a new data set in the limited meteorological historical data using the difference data between the forecast data and the analysis data contained therein, and fuses the data set with the meteorological historical data for use in the generation and update process of the prediction model, thereby making full use of the temporal and spatial characteristics of the data to process the meteorological forecast data under limited data conditions, thereby fully capturing the dynamic changes of meteorological variables and improving the accuracy of marine weather forecasts.

[0056] The marine weather forecast data processing system provided in the embodiment of the present invention has the same implementation principle and technical effects as those of the aforementioned marine weather forecast data processing method embodiment. For the sake of brief description, for matters not mentioned in the system embodiment, reference may be made to the corresponding contents in the aforementioned marine weather forecast data processing method embodiment.

[0057] This embodiment also provides an electronic device. The structural diagram of the electronic device is as follows: Fig.11 As shown, the device includes a processor 101 and a memory 102; wherein the memory 102 is used to store one or more computer instructions, and the one or more computer instructions are executed by the processor to implement the steps of the above-mentioned marine weather forecast data processing method.

[0058] Fig.11The electronic device shown further includes a bus 103 and a communication interface 104 , and the processor 101 , the communication interface 104 and the memory 102 are connected via the bus 103 .

[0059] The memory 102 may include a high-speed random access memory (RAM), and may also include a non-volatile memory, such as at least one disk storage. The bus 103 may be an ISA bus, a PCI bus, or an EISA bus. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Fig.11 Only one bidirectional arrow is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0060] The communication interface 104 is used to connect to at least one user terminal and other network units through a network interface, and send the encapsulated IPv4 message or IPv4 message to the user terminal through the network interface.

[0061] The processor 101 may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method can be completed by the hardware integrated logic circuit or software instructions in the processor 101. The above processor 101 can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The disclosed methods, steps and logic block diagrams in the embodiments of the present disclosure can be implemented or executed. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The steps of the method disclosed in conjunction with the embodiments of the present disclosure can be directly embodied as a hardware decoding processor for execution, or a combination of hardware and software modules in the decoding processor for execution. The software module may be located in a storage medium mature in the art, such as a random access memory, a flash memory, a read-only memory, a programmable read-only memory, or an electrically erasable programmable memory, a register, etc. The storage medium is located in the memory 102, and the processor 101 reads the information in the memory 102 and completes the steps of the method of the above embodiment in combination with its hardware.

[0062] An embodiment of the present invention further provides a storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps of the marine weather forecast data processing method in the aforementioned embodiment are executed.

[0063] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices, equipment and methods can be implemented in other ways. The system embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0064] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0065] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0066] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that can be executed by a processor. Based on this understanding, the technical solution of the present invention can essentially or in other words, the part that contributes to the prior art or the part of the technical solution can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0067] Finally, it should be noted that the above-described embodiments are only specific implementations of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The protection scope of the present invention is not limited thereto. Although the present invention is described in detail with reference to the above-described embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the above-described embodiments within the technical scope disclosed by the present invention, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for processing marine weather forecast data, characterized in that: The method comprises: Acquire ship observation data corresponding to the observation point based on the position data of the observation point, and acquire weather forecast historical data and weather analysis historical data corresponding to the observation point using the ship observation data; Acquire a random point corresponding to the observation point, use the random point to construct an observation area corresponding to the observation point, and use the meteorological forecast historical data and the meteorological analysis historical data contained in the observation area to construct a first meteorological data set; Determine difference data between the weather forecast historical data and the weather analysis historical data in the observation area at the same historical moment, and construct a second weather data set corresponding to the observation area using the difference data; constructing a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological dataset, and updating the prediction model using the difference data in the second meteorological dataset; Obtain the meteorological forecast data corresponding to the observation point at the current moment, and use the updated prediction model to obtain the meteorological analysis data corresponding to the meteorological forecast data.

2. The method for processing marine weather forecast data according to claim 1, characterized in that: Acquiring ship observation data corresponding to the observation point based on the position data of the observation point, and acquiring weather forecast historical data and weather analysis historical data corresponding to the observation point using the ship observation data, including: Determining the ship observation data collected by the meteorological ship at the observation point according to the position data determined at the observation point; Determine the ocean area corresponding to the meteorological ship and the observation point, and obtain corresponding meteorological historical data in the ocean area; The weather forecast historical data and the weather analysis historical data are determined based on the weather historical data.

3. The method for processing marine weather forecast data according to claim 2, characterized in that: Acquiring a random point corresponding to the observation point, using the random point to construct an observation area corresponding to the observation point, and using the meteorological forecast historical data and the meteorological analysis historical data contained in the observation area to construct a first meteorological data set, including: Acquire a plurality of the random points corresponding to the observation point in the ocean area between the meteorological ship and the observation point; Determine the boundary of the region corresponding to the observation point by using the position data of the plurality of random points, and construct the observation region corresponding to the observation point by using the region boundary; Acquire the weather forecast historical data and the weather analysis historical data corresponding to the observation point and the random point in the observation area; The first meteorological data set is constructed using the meteorological forecast historical data and the meteorological analysis historical data.

4. The method for processing marine weather forecast data according to claim 1, characterized in that: Determining difference data between the weather forecast historical data and the weather analysis historical data in the observation area at the same historical moment, and constructing a second weather data set corresponding to the observation area using the difference data, including: Acquire the weather forecast historical data and the weather analysis historical data corresponding to the observation area at the same historical moment, and determine the weather forecast value and the weather observation value corresponding to the observation area based on the weather forecast historical data and the weather analysis historical data; Determine the difference data using the meteorological forecast value and the meteorological observation value, transfer the difference data to a grid field corresponding to the observation area using an inverse distance weighted interpolation method, and then generate an error field corresponding to the observation area using the grid field; After the first meteorological dataset is updated based on the error field, the second meteorological dataset is constructed according to the updated first meteorological dataset.

5. The method for processing marine weather forecast data according to claim 1, characterized in that: Constructing a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set, including: Acquire the time series feature data corresponding to the observation area in the first meteorological data set, and construct a time series convolution module using the time series feature data; Acquire the scale feature data corresponding to the observation area in the first meteorological data set, and construct a compression excitation network module using the scale feature data; The prediction model corresponding to the observation area is constructed using the temporal convolution module and the compression excitation network module.

6. The method for processing marine weather forecast data according to claim 5, characterized in that: The prediction model corresponding to the observation area is constructed by using the temporal convolution module and the compression excitation network module, including: Using the temporal convolution module to construct a temporal convolution network corresponding to the temporal feature data, and using the compression excitation network module to construct a compression excitation network corresponding to the scale feature data; The prediction model corresponding to the observation area is constructed according to the temporal convolutional network and the compression excitation network; wherein the prediction model is trained using the weather forecast historical data and the weather analysis historical data.

7. The method for processing marine weather forecast data according to claim 1, characterized in that: Updating the prediction model using the difference data in the second meteorological data set includes: constructing revised forecast data corresponding to the observation area using the difference data in the second meteorological data set; The meteorological analysis historical data is updated using the revised forecast data, and the prediction model is updated using the updated meteorological analysis historical data and the meteorological forecast historical data.

8. The method for processing marine weather forecast data according to claim 1, characterized in that: Obtaining the meteorological forecast data corresponding to the observation point at the current moment, and using the updated prediction model to obtain the meteorological analysis data corresponding to the meteorological forecast data, including: Acquire the meteorological forecast data corresponding to the observation point at the current moment, and use the meteorological forecast data to update the first meteorological data set and the second meteorological data set; After updating the prediction model using the updated first meteorological data set and the updated second meteorological data set, the time series feature data and the scale feature data corresponding to the meteorological forecast data are obtained according to the updated prediction model; The meteorological analysis data corresponding to the meteorological forecast data is determined according to the time series characteristic data and the scale characteristic data.

9. A marine weather forecast data processing system, characterized in that: The system comprises: A data acquisition unit, used to obtain ship observation data corresponding to the observation point based on the position data of the observation point, and to obtain weather forecast historical data and weather analysis historical data corresponding to the observation point using the ship observation data; a first meteorological data set generating unit, configured to obtain a random point corresponding to the observation point, construct an observation area corresponding to the observation point using the random point, and construct a first meteorological data set using the meteorological forecast historical data and the meteorological analysis historical data contained in the observation area; A second meteorological data set generating unit, configured to determine difference data between the meteorological forecast historical data and the meteorological analysis historical data in the observation area at the same historical moment, and construct a second meteorological data set corresponding to the observation area using the difference data; A prediction model building unit, configured to build a prediction model corresponding to the observation area according to the time series feature data and scale feature data corresponding to the observation area in the first meteorological data set, and update the prediction model using the difference data in the second meteorological data set; The weather forecast data processing unit is used to obtain the weather forecast data corresponding to the observation point at the current moment, and to obtain the weather analysis data corresponding to the weather forecast data using the updated prediction model.

10. An electronic device, characterized in that: It comprises a processor and a memory, wherein the memory stores computer executable instructions that can be executed by the processor, and the processor executes the computer executable instructions to implement the steps of the marine weather forecast data processing method as described in any one of claims 1 to 8.

Images