River flow prediction method and device, electronic equipment and computer program product

Abstract

The invention discloses a river flow prediction method and device, electronic equipment and a computer program product, and relates to the technical field of flow prediction. The method comprises the following steps: acquiring first precipitation space-time distribution data of a region where a to-be-measured river is located in a target time period after a current time point from a plurality of weather forecast centers; taking the plurality of first precipitation space-time distribution data as the input of a pre-trained precipitation prediction model for operation, and obtaining second precipitation space-time distribution data of the region where the to-be-detected river is located in the target time period, the precipitation prediction model comprises a first model based on a convolutional neural network and a second model based on a long short-term memory neural network; and taking the second precipitation space-time distribution data as input of a pre-trained river flow prediction model for operation, and obtaining flow time sequence data of each river section in the to-be-measured river in the target time period. According to the method, the flow of the river can be predicted more accurately, and an important scientific support is provided for flood prevention and disaster reduction of the river basin.

Description

Technical Field

[0001] The present invention belongs to the technical field of flow prediction, and in particular relates to a river flow prediction method, device, electronic equipment and computer program product. Background Art

[0002] Accurate prediction of river flow is of great significance for flood prevention and disaster reduction, water resources allocation, ecological and environmental protection, and sustainable economic development.

[0003] At present, the prediction of river flow is mostly based on the numerical weather forecast issued by the Meteorological Forecast Center to predict the flow of rivers in the future. However, due to the complexity of the atmospheric system, the high degree of nonlinearity and the limitations of the forecast system itself, the results of numerical weather forecasts often have certain systematic errors, resulting in a large difference between the predicted results of river flow and the actual flow of the river.

[0004] Therefore, how to provide an effective solution to more accurately predict the flow of a river has become a difficult problem to be solved in the prior art. Summary of the invention

[0005] The purpose of the present invention is to provide a river flow prediction method, device, electronic equipment and computer program product to solve the above problems existing in the prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: In a first aspect, the present invention provides a river flow prediction method, comprising: Obtaining the spatiotemporal distribution data of the first precipitation in the region where the river to be measured is located in a target period after the current time point from multiple meteorological forecast centers; The plurality of the first spatiotemporal distribution data of precipitation are used as inputs of a pre-trained precipitation prediction model for operation to obtain the second spatiotemporal distribution data of precipitation in the area where the river to be measured is located during the target period, wherein the precipitation prediction model includes a first model based on a convolutional neural network and a second model based on a long short-term memory neural network, wherein the first model is used to output a multidimensional vector based on the plurality of the first spatiotemporal distribution data of precipitation, and the second model is used to output the second spatiotemporal distribution data of precipitation by taking the multidimensional vector as input; The second precipitation spatiotemporal distribution data is used as input of a pre-trained river flow prediction model for operation to obtain the flow time series data of each river section in the river to be tested in the target time period, wherein the river flow prediction model is trained using the historical precipitation spatiotemporal distribution data of the area where the river to be tested is located in the historical time period as sample input and the historical flow time series data of each river section in the river to be tested in the historical time period as sample output.

[0007] Based on the above disclosed content, the present invention obtains first precipitation spatiotemporal distribution data of the area where the river to be measured is located in a target period after the current time point from multiple meteorological forecast centers; then uses multiple first precipitation spatiotemporal distribution data as input of a pre-trained precipitation prediction model for calculation, and obtains second precipitation spatiotemporal distribution data of the area where the river to be measured is located in the target period. The precipitation prediction model includes a first model based on a convolutional neural network and a second model based on a long short-term memory neural network. The first model is used to output a multidimensional vector based on multiple first precipitation spatiotemporal distribution data, and the second model is used to use the multidimensional vector as input to output second precipitation spatiotemporal distribution data; finally, the second precipitation spatiotemporal distribution data is used as input of a pre-trained river flow prediction model for calculation, and the flow time series data of each river section in the river to be measured in the target period is obtained. In this way, when predicting river flow, the analysis and processing capabilities of long short-term memory neural networks for time series data and the analysis and processing capabilities of convolutional neural networks for spatial feature data are utilized, and the advantages of the two are combined to construct a precipitation prediction model for precipitation prediction. The temporal and spatial characteristics of precipitation data are comprehensively considered to obtain the spatiotemporal distribution data of precipitation obtained from multiple meteorological forecast centers, and the spatiotemporal distribution data of precipitation in the area where the river to be tested is located can be more accurately predicted based on the spatiotemporal distribution data of precipitation obtained from multiple meteorological forecast centers. This can more accurately predict the flow of the river, providing important scientific support for flood control and disaster reduction in river basins.

[0008] In one possible design, the method further includes: Dividing the area where the river to be tested is located into multiple sub-areas; The step of using the plurality of the first spatiotemporal distribution data of precipitation as input of a pre-trained precipitation prediction model to perform calculations to obtain the second spatiotemporal distribution data of precipitation in the area where the river to be measured is located during the target period includes: The first rainfall time series data of the multiple sub-areas in the target time period after the current time point is used as the input of the precipitation prediction model for operation to obtain the second rainfall time series data of the multiple sub-areas in the target time period.

[0009] In one possible design, the method further includes: Based on the flow time series data of each river section in the river to be tested in the target time period, the flood peak data of each river section in the river to be tested and the time corresponding to the flood peak data are determined.

[0010] In one possible design, the multiple meteorological forecast centers include the European Centre for Medium-Range Weather Forecasts, the China Meteorological Administration, and the U.S. National Centers for Environmental Prediction.

[0011] In one possible design, the method further includes: Acquire a plurality of sample data, each of which includes first historical precipitation spatiotemporal distribution data of the river to be measured acquired from the plurality of meteorological forecast centers; The multiple sample data are used as inputs of a pre-established precipitation prediction model, and the second historical spatiotemporal distribution data of precipitation in the area where the river to be tested is located in the historical period corresponding to the multiple sample data are used as outputs of the precipitation prediction model for training to obtain a trained precipitation prediction model.

[0012] In a possible design, during the training process of the precipitation prediction model, the parameters of the precipitation prediction model are adjusted through a non-inferior hierarchical genetic algorithm.

[0013] In one possible design, the target period after the current time point is 0-240 hours after the current time point.

[0014] In a second aspect, the present invention provides a river flow prediction device, comprising: An acquisition unit, used for acquiring the first spatiotemporal distribution data of precipitation in the region where the river to be measured is located in a target period after the current time point from multiple meteorological forecast centers; A first operation unit is used to use the plurality of the first spatiotemporal distribution data of precipitation as input of a pre-trained precipitation prediction model to perform operation to obtain second spatiotemporal distribution data of precipitation in the area where the river to be measured is located during a target period of time, wherein the precipitation prediction model includes a first model based on a convolutional neural network and a second model based on a long short-term memory neural network, wherein the first model is used to output a multidimensional vector based on the plurality of the first spatiotemporal distribution data of precipitation, and the second model is used to use the multidimensional vector as input to output the second spatiotemporal distribution data of precipitation; The second computing unit is used to use the second precipitation spatiotemporal distribution data as an input of a pre-trained river flow prediction model to perform computing to obtain the flow time series data of each river section in the river to be tested during the target time period, wherein the river flow prediction model is trained using the historical precipitation spatiotemporal distribution data of the area where the river to be tested is located during the historical time period as sample input and the historical flow time series data of each river section in the river to be tested during the historical time period as sample output.

[0015] In a third aspect, the present invention provides an electronic device comprising a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the river flow prediction method as described in the first aspect or any possible design of the first aspect.

[0016] In a fourth aspect, the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, executes the river flow prediction method described in the first aspect or any possible design of the first aspect.

[0017] In a fifth aspect, the present invention provides a computer program product comprising instructions, which, when executed on a computer, causes the computer to execute the river flow prediction method as described in the first aspect or any possible design of the first aspect.

[0018] Beneficial effects: The river flow prediction method, device, electronic device and computer program product provided by the present invention can use the analysis and processing capabilities of long short-term memory neural networks for time series data and the analysis and processing capabilities of convolutional neural networks for spatial feature data when predicting river flow, combine the advantages of the two to build a precipitation prediction model for precipitation prediction, comprehensively consider the temporal and spatial characteristics of precipitation data, obtain precipitation spatiotemporal distribution data from multiple meteorological forecast centers, and more accurately predict the precipitation spatiotemporal distribution data of the area where the river to be tested is located based on the precipitation spatiotemporal distribution data obtained from multiple meteorological forecast centers, thereby being able to more accurately predict the flow of the river, provide important scientific support for flood control and disaster reduction in river basins, and facilitate practical application and promotion. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 A flow chart of a river flow prediction method provided in an embodiment of the present application; Figure 2 A schematic block diagram of a river flow prediction device provided in an embodiment of the present application; Figure 3 A schematic block diagram of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0020] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the present invention will be briefly introduced below in combination with the drawings and the description of the embodiments or the prior art. Obviously, the following description of the structure of the drawings is only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention.

[0021] It should be understood that although the terms first, second, etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are only used to distinguish one unit from another unit. For example, a first unit can be referred to as a second unit, and similarly, a second unit can be referred to as a first unit without departing from the scope of the exemplary embodiments of the present invention.

[0022] It should be understood that the term "and / or" that may appear in this article is merely a description of the association relationship between associated objects, indicating that three relationships may exist. For example, A and / or B can represent three situations: A exists alone, B exists alone, and A and B exist at the same time. The term " / and" that may appear in this article describes another type of association object relationship, indicating that two relationships may exist. For example, A / and B can represent two situations: A exists alone, and A and B exist alone. In addition, the character " / " that may appear in this article generally indicates that the previous and next associated objects are in an "or" relationship.

[0023] In order to accurately predict the flow of a river, the embodiments of the present application provide a river flow prediction method, device, electronic device and computer program product. The river flow prediction method, device, electronic device and computer program product can more accurately predict the flow of a river, providing important scientific support for flood prevention and disaster reduction in river basins.

[0024] The river flow prediction method provided in the embodiment of the present application can be applied to a user terminal or a server. It can be understood that the execution subject does not constitute a limitation on the embodiment of the present application.

[0025] The river flow prediction method provided in the embodiment of the present application will be described in detail below.

[0026] like Figure 1 As shown, it is a flowchart of the river flow prediction method provided in the first aspect of the embodiment of the present application. The river flow prediction method may include but is not limited to the following steps S101-S103.

[0027] Step S101. Obtaining the first spatiotemporal distribution data of precipitation in the region where the river to be measured is located in a target period after the current time point from multiple meteorological forecast centers.

[0028] The multiple weather forecast centers may be, but are not limited to, at least two of the European Center for Medium-Range Weather Forecasts (ECMWF), the China Meteorological Administration (CMA), and the National Centers for Environmental Prediction (CEP). In an embodiment of the present application, the multiple weather forecast centers include the European Center for Medium-Range Weather Forecasts, the China Meteorological Administration, and the National Centers for Environmental Prediction (CEP).

[0029] The spatiotemporal distribution data of precipitation in the area where the river to be tested is located may refer to rainfall data at different locations in the area where the river to be tested is located at different time points or time periods.

[0030] The target time period after the current time point can be set according to actual conditions, such as 0-24h, 0-72h, 0-120h or 0-240h after the current time point.

[0031] Step S102: using a plurality of first spatiotemporal distribution data of precipitation as input of a pre-trained precipitation prediction model for operation, and obtaining second spatiotemporal distribution data of precipitation in the region where the river to be measured is located during a target period of time.

[0032] Among them, the precipitation prediction model includes a first model based on convolutional neural networks (CNN) and a second model based on long short-term memory neural networks (LSTM). The first model is used to output a multidimensional vector based on multiple first precipitation spatiotemporal distribution data, and the second model is used to take the multidimensional vector as input and output the second precipitation spatiotemporal distribution data of the area where the river to be tested is located in the target period.

[0033] In an embodiment of the present application, a precipitation prediction model for precipitation prediction is pre-trained, and multiple sample data can be obtained during training, wherein each sample data includes first historical spatiotemporal distribution data of precipitation of the river to be tested obtained from the multiple meteorological forecast centers (European Center for Medium-Term Weather Forecasts, China Meteorological Administration and National Centers for Environmental Prediction of the United States), and then the multiple sample data are used as inputs of the pre-established precipitation prediction model, and the second historical spatiotemporal distribution data of precipitation in the area where the river to be tested is located in the historical period corresponding to the multiple sample data are used as outputs of the precipitation prediction model for training, thereby obtaining a trained precipitation prediction model.

[0034] In the process of training the precipitation prediction model, statistical indicators such as relative bias, mean absolute error (MAE), correlation coefficient (CC) and root mean square error (RMSE) can be used to analyze and evaluate the first historical precipitation spatiotemporal distribution data of the European Center for Medium-Range Weather Forecasts, China Meteorological Administration and the National Centers for Environmental Prediction of the United States, and the predicted second historical precipitation spatiotemporal distribution data, and the precipitation prediction model can be adjusted through a non-inferior hierarchical genetic algorithm.

[0035] In one or more embodiments, the area where the river to be measured is located can be divided into multiple sub-areas, such as multiple grids, and the first precipitation spatiotemporal distribution data of the area where the river to be measured is located in the target period after the current time point can be the rainfall time series data of each sub-area in the target period after the current time point (i.e., rainfall data at different time points or time periods). When calculating the second precipitation spatiotemporal distribution data of the area where the river to be measured is located in the target period, the first rainfall time series data of multiple sub-areas in the target period after the current time point can be used as the input of the precipitation prediction model for operation to obtain the second rainfall time series data of multiple sub-areas in the target period.

[0036] Step S103: using the second precipitation temporal and spatial distribution data as input to a pre-trained river flow prediction model for operation, and obtaining flow time series data of each river section in the river to be tested during a target period of time.

[0037] In an embodiment of the present application, a river flow prediction model for flow prediction is also pre-trained. The river flow prediction model can be trained using the spatiotemporal distribution data of historical precipitation in the area where the river to be tested is located in a historical period as sample input and the historical flow time series data of each river section in the river to be tested in the historical period as sample output.

[0038] After obtaining the second spatiotemporal distribution data of precipitation in the area where the river to be tested is located during the target period, the second spatiotemporal distribution data of precipitation can be used as input of a pre-trained river flow prediction model to obtain the flow time series data of each river section in the river to be tested during the target period.

[0039] In one or more embodiments, the accuracy of the river flow prediction model can also be evaluated based on the predicted flow time series data of each river section in the river to be tested in the historical period and the actual flow time series data of each river section in the river to be tested in the historical period.

[0040] In one or more embodiments, after determining the flow time series data of each river section in the river to be tested during the target time period, the flood peak data of each river section in the river to be tested and the time corresponding to the flood peak data can be determined based on the flow time series data of each river section in the river to be tested during the target time period, thereby providing important scientific support for flood prevention and disaster reduction in river basins.

[0041] In summary, the river flow prediction method provided by the present invention obtains the first spatiotemporal distribution data of precipitation in the target period after the current time point in the area where the river to be measured is located from multiple meteorological forecast centers; then the multiple first spatiotemporal distribution data of precipitation are used as the input of a pre-trained precipitation prediction model for calculation, and the second spatiotemporal distribution data of precipitation in the target period in the area where the river to be measured is located is obtained. The precipitation prediction model includes a first model based on a convolutional neural network and a second model based on a long short-term memory neural network. The first model is used to output a multidimensional vector based on the multiple first spatiotemporal distribution data of precipitation, and the second model is used to use the multidimensional vector as input to output the second spatiotemporal distribution data of precipitation; finally, the second spatiotemporal distribution data of precipitation is used as the input of the pre-trained river flow prediction model for calculation, and the flow time series data of each river section in the river to be measured in the target period is obtained. In this way, when predicting river flow, the analysis and processing capabilities of long short-term memory neural networks for time series data and the analysis and processing capabilities of convolutional neural networks for spatial feature data are utilized, and the advantages of the two are combined to construct a precipitation prediction model for precipitation prediction. The temporal and spatial characteristics of precipitation data are comprehensively considered to obtain the spatiotemporal distribution data of precipitation obtained from multiple meteorological forecast centers, and the spatiotemporal distribution data of precipitation in the area where the river to be tested is located can be more accurately predicted based on the spatiotemporal distribution data of precipitation obtained from multiple meteorological forecast centers. This can more accurately predict the flow of the river, provide important scientific support for flood control and disaster reduction in river basins, and facilitate practical application and promotion.

[0042] See also Figure 2 According to a second aspect of an embodiment of the present application, a river flow prediction device is provided, the river flow prediction device comprising: An acquisition unit, used for acquiring the first spatiotemporal distribution data of precipitation in the region where the river to be measured is located in a target period after the current time point from multiple meteorological forecast centers; A first operation unit is used to use the plurality of the first spatiotemporal distribution data of precipitation as input of a pre-trained precipitation prediction model to perform operation to obtain second spatiotemporal distribution data of precipitation in the area where the river to be measured is located during a target period of time, wherein the precipitation prediction model includes a first model based on a convolutional neural network and a second model based on a long short-term memory neural network, wherein the first model is used to output a multidimensional vector based on the plurality of the first spatiotemporal distribution data of precipitation, and the second model is used to use the multidimensional vector as input to output the second spatiotemporal distribution data of precipitation; The second computing unit is used to use the second precipitation spatiotemporal distribution data as an input of a pre-trained river flow prediction model to perform computing to obtain the flow time series data of each river section in the river to be tested during the target time period, wherein the river flow prediction model is trained using the historical precipitation spatiotemporal distribution data of the area where the river to be tested is located during the historical time period as sample input and the historical flow time series data of each river section in the river to be tested during the historical time period as sample output.

[0043] The working process, working details and technical effects of the river flow prediction device provided in the second aspect of this embodiment can be found in the first aspect of the embodiment and will not be described in detail here.

[0044] like Figure 3 As shown, the third aspect of an embodiment of the present application provides an electronic device, comprising a memory, a processor and a transceiver that are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the river flow prediction method as described in the first aspect of the embodiment.

[0045] For specific examples, the memory may include but is not limited to random access memory (RAM), read-only memory (ROM), flash memory, first-in-first-out memory (FIFO) and / or first-in-last-out memory (FILO), etc.; the processor may be but is not limited to a microprocessor of the STM32F105 series, an ARM (Advanced RISC-Machines), an X86 or other architecture processor, or a processor with an integrated NPU (neural-network processing units); the transceiver may be but is not limited to a WiFi (Wireless Fidelity) wireless transceiver, a Bluetooth wireless transceiver, a General Packet Radio Service (GPRS) wireless transceiver, a ZigBee protocol (a low-power local area network protocol based on the IEEE802.15.4 standard, ZigBee) wireless transceiver, a 3G transceiver, a 4G transceiver and / or a 5G transceiver, etc.

[0046] In a fourth aspect of the present embodiment, there is provided a computer-readable storage medium storing instructions for the river flow prediction method described in the first aspect of the embodiment, that is, the computer-readable storage medium stores instructions, and when the instructions are executed on a computer, the river flow prediction method described in the first aspect is executed. The computer-readable storage medium refers to a carrier for storing data, which may include but is not limited to a floppy disk, an optical disk, a hard disk, a flash memory, a USB flash drive, and / or a memory stick, etc., and the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0047] The fifth aspect of this embodiment provides a computer program product comprising instructions, which, when executed on a computer, causes the computer to execute the river flow prediction method as described in the first aspect of the embodiment, wherein the computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices.

[0048] It should be understood that certain details are provided in the following description to facilitate a complete understanding of the example embodiments. However, it should be understood by those of ordinary skill in the art that the example embodiments can be implemented without these certain details. For example, the system can be shown in a block diagram to avoid obscuring the example with unnecessary details. In other examples, well-known processes, structures, and techniques may not be shown in unnecessary detail to avoid obscuring the example embodiments.

[0049] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the protection scope of the present invention. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A river flow prediction method, characterized in that: include: Obtaining the spatiotemporal distribution data of the first precipitation in the region where the river to be measured is located in a target period after the current time point from multiple meteorological forecast centers; The plurality of the first spatiotemporal distribution data of precipitation are used as inputs of a pre-trained precipitation prediction model for operation to obtain the second spatiotemporal distribution data of precipitation in the area where the river to be measured is located during the target period, wherein the precipitation prediction model includes a first model based on a convolutional neural network and a second model based on a long short-term memory neural network, wherein the first model is used to output a multidimensional vector based on the plurality of the first spatiotemporal distribution data of precipitation, and the second model is used to output the second spatiotemporal distribution data of precipitation by taking the multidimensional vector as input; The second precipitation spatiotemporal distribution data is used as input of a pre-trained river flow prediction model for operation to obtain the flow time series data of each river section in the river to be tested in the target time period, wherein the river flow prediction model is trained using the historical precipitation spatiotemporal distribution data of the area where the river to be tested is located in the historical time period as sample input and the historical flow time series data of each river section in the river to be tested in the historical time period as sample output.

2. The river flow prediction method according to claim 1, characterized in that: The method further comprises: Dividing the area where the river to be tested is located into multiple sub-areas; The step of using the plurality of the first spatiotemporal distribution data of precipitation as input of a pre-trained precipitation prediction model to perform calculations to obtain the second spatiotemporal distribution data of precipitation in the area where the river to be measured is located during the target period includes: The first rainfall time series data of the multiple sub-areas in the target time period after the current time point is used as the input of the precipitation prediction model for operation to obtain the second rainfall time series data of the multiple sub-areas in the target time period.

3. The river flow prediction method according to claim 1, characterized in that: The method further comprises: Based on the flow time series data of each river section in the river to be tested in the target time period, the flood peak data of each river section in the river to be tested and the time corresponding to the flood peak data are determined.

4. The river flow prediction method according to claim 1, characterized in that: The multiple meteorological forecast centers include the European Centre for Medium-Range Weather Forecasts, the China Meteorological Administration and the US National Centers for Environmental Prediction.

5. The river flow prediction method according to claim 1, characterized in that: The method further comprises: Acquire a plurality of sample data, each of which includes first historical precipitation spatiotemporal distribution data of the river to be measured acquired from the plurality of meteorological forecast centers; The multiple sample data are used as inputs of a pre-established precipitation prediction model, and the second historical spatiotemporal distribution data of precipitation in the area where the river to be tested is located in the historical period corresponding to the multiple sample data are used as outputs of the precipitation prediction model for training to obtain a trained precipitation prediction model.

6. The river flow prediction method according to claim 5, characterized in that: During the training process of the precipitation prediction model, the parameters of the precipitation prediction model are adjusted through a non-inferior hierarchical genetic algorithm.

7. The river flow prediction method according to claim 1, characterized in that: The target period after the current time point is 0-240h after the current time point.

8. A river flow prediction device, characterized in that: include: An acquisition unit, used for acquiring the first spatiotemporal distribution data of precipitation in the region where the river to be measured is located in a target period after the current time point from multiple meteorological forecast centers; A first operation unit is used to use the plurality of the first spatiotemporal distribution data of precipitation as input of a pre-trained precipitation prediction model to perform operation to obtain second spatiotemporal distribution data of precipitation in the area where the river to be measured is located during a target period of time, wherein the precipitation prediction model includes a first model based on a convolutional neural network and a second model based on a long short-term memory neural network, wherein the first model is used to output a multidimensional vector based on the plurality of the first spatiotemporal distribution data of precipitation, and the second model is used to use the multidimensional vector as input to output the second spatiotemporal distribution data of precipitation; The second computing unit is used to use the second precipitation spatiotemporal distribution data as an input of a pre-trained river flow prediction model to perform computing to obtain the flow time series data of each river section in the river to be tested during the target time period, wherein the river flow prediction model is trained using the historical precipitation spatiotemporal distribution data of the area where the river to be tested is located during the historical time period as sample input and the historical flow time series data of each river section in the river to be tested during the historical time period as sample output.

9. An electronic device, characterized in that: It comprises a memory, a processor and a transceiver which are communicatively connected in sequence, wherein the memory is used to store a computer program, the transceiver is used to send and receive messages, and the processor is used to read the computer program to execute the river flow prediction method as described in any one of claims 1 to 7.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or the instruction is executed by a computer, the river flow prediction method according to any one of claims 1 to 7 is implemented.

Images