Runoff prediction method and device, storage medium and computer equipment

Abstract

The invention provides a runoff prediction method and device, a storage medium and computer equipment. The runoff prediction method comprises the following steps: acquiring historical hydro-meteorological data and historical climate index data of a target drainage basin, and historical runoff return data and historical runoff data of a plurality of preset historical periods; the historical runoff return data is runoff data of a preset historical period obtained by simulating a plurality of different runoff simulation models according to historical hydro meteorological data and historical climate index data; the historical runoff data are actually measured runoff data of historical periods corresponding to the historical runoff return data; according to historical hydrometeorological data, historical climate index data, historical runoff return data and historical runoff data, training the plurality of initial runoff prediction models to obtain a plurality of runoff prediction models; and inputting the watershed hydro meteorological data, the watershed climate index data and the plurality of runoff return data of the target watershed into a plurality of runoff prediction models to obtain runoff data prediction data.

Description

Technical Field

[0001] This application relates to the technical field of runoff prediction, and particularly relates to a runoff prediction method, device, storage medium, and computer device. Background Art

[0002] Existing runoff prediction technologies drive numerical models of the Earth system by describing and solving mathematical and physical equations to systematically simulate and predict watershed hydrological processes including runoff. This method is based on strong physical mechanisms, but the uncertainty of the results is greatly affected by model structure, parameters, and initial conditions. However, the accuracy of this model in capturing the characteristics of climate and hydrological processes at the regional local scale is low. Therefore, the numerical model of the Earth system cannot accurately predict long-term runoff data, that is, the longer the prediction time corresponding to the prediction result output by the numerical model of the Earth system, the lower the accuracy of the prediction result. Summary of the Invention

[0003] The purpose of this application is to overcome the deficiencies in the prior art and provide a runoff prediction method, device, storage medium, and computer device that can obtain highly accurate runoff prediction data.

[0004] The first aspect of the embodiments of this application provides a runoff prediction method, including:

[0005] Obtain historical hydrometeorological data, historical climate index data, historical runoff reanalysis data of multiple preset historical periods, and historical runoff data of multiple said preset historical periods of the target watershed; wherein, the historical runoff reanalysis data is runoff data of a preset historical period simulated by several different runoff simulation models according to the historical hydrometeorological data and historical climate index data; the historical runoff data is measured runoff data of the historical period corresponding to the historical runoff reanalysis data;

[0006] Train several initial runoff prediction models according to the historical hydrometeorological data, the historical climate index data, the historical runoff reanalysis data, and the historical runoff data to obtain several runoff prediction models;

[0007] Input the watershed hydrometeorological data, watershed climate index data, and multiple runoff reanalysis data of the target watershed into several runoff prediction models to obtain runoff data prediction data.

[0008] The second aspect of the embodiments of this application provides a runoff prediction device, including:

[0009] A historical data acquisition module, configured to acquire historical hydrometeorological data, historical climate index data, historical runoff prediction data for multiple preset historical periods, and historical runoff data for multiple said preset historical periods of a target basin; wherein, the historical runoff prediction data is runoff data for a preset historical period simulated by a plurality of different runoff simulation models based on the historical hydrometeorological data and historical climate index data; the historical runoff data is measured runoff data for the historical period corresponding to the historical runoff prediction data;

[0010] A prediction model acquisition module, configured to train a plurality of initial runoff prediction models based on the historical hydrometeorological data, the historical climate index data, the historical runoff prediction data, and the historical runoff data to obtain a plurality of runoff prediction models;

[0011] Input the basin hydrometeorological data, basin climate index data, and multiple runoff prediction data of the target basin into a plurality of runoff prediction models to obtain predicted runoff data.

[0012] A third aspect of the embodiments of the present application provides a computer-readable storage medium storing a computer program, which when executed by a processor implements the steps of the runoff prediction method as described above.

[0013] A fourth aspect of the embodiments of the present application provides a computer device, including a storage, a processor, and a computer program stored in the storage and executable by the processor, wherein when the processor executes the computer program, it implements the steps of the runoff prediction method as described above.

[0014] Compared with the prior art, the present application first acquires the historical hydrometeorological data, historical climate index data, historical runoff data for multiple preset historical periods, and runoff data for a preset historical period simulated based on the historical hydrometeorological data and historical climate index data of the target basin, and then performs model training based on the historical hydrometeorological data, the historical climate index data, the historical runoff prediction data, and the historical runoff data to obtain a plurality of runoff prediction models, and then corrects and optimizes the multiple runoff prediction data according to the basin hydrometeorological data, basin climate index data, and runoff prediction models of the target basin, and more accurate multiple runoff prediction data can be obtained.

[0015] To more clearly understand the present application, the specific embodiments of the present application will be described below in conjunction with the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flowchart of the runoff prediction method according to an embodiment of the present application.

[0017] Figure 2 Schematic diagram of a runoff prediction model for a runoff prediction method according to an embodiment of the present application.

[0018] Figure 3 Schematic diagram of module connections of a runoff prediction data acquisition device according to an embodiment of the present application.

[0019] 100. Medium- and long-term runoff prediction data acquisition device; 101. Historical data acquisition module; 102. Prediction model acquisition module; 103. Runoff prediction data acquisition module. Detailed implementation manners

[0020] To make the objectives, technical solutions and advantages of the present application clearer, the following will further describe the embodiments of the present application in detail with reference to the accompanying drawings.

[0021] It should be clear that the described embodiments are only a part, rather than all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts fall within the scope of protection of the embodiments of the present application.

[0022] When the following description refers to the accompanying drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. In the description of the present application, it should be understood that the terms "first", "second", "third", etc. are only used to distinguish similar objects and do not have to be used to describe a specific order or sequence, nor should they be construed as indicating or implying relative importance. For those of ordinary skill in the art, the specific meanings of the above terms in the present application can be understood according to specific circumstances. The singular forms of "a", "the" and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. The word "if" / "when" used herein can be interpreted as "when...", "when...", or "in response to a determination".

[0023] In addition, in the description of the present application, unless otherwise specified, "a plurality of" means two or more. "And / or" describes the association relationship of associated objects and indicates that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. The character " / " generally represents an "or" relationship between the associated objects before and after.

[0024] Please refer to Figure 1 , which is a flowchart of a runoff prediction method according to an embodiment of the present application, including:

[0025] S1: Obtain the historical hydrometeorological data, historical climate index data, historical runoff return data for multiple preset historical periods, and historical runoff data for multiple said preset historical periods of the target basin; wherein, the historical runoff return data is the runoff data of the preset historical period simulated by a number of different runoff simulation models based on the historical hydrometeorological data and historical climate index data; the historical runoff data is the measured runoff data of the historical period corresponding to the historical runoff return data.

[0026] Historical hydrometeorological data refers to the hydrometeorological data obtained from the historical data of the target basin, and the hydrometeorological data may include precipitation data, runoff data, etc.; historical climate index data refers to the climate index data obtained from the historical data of the target basin, and the climate index data may include atmospheric circulation index, sea surface temperature index, etc.; historical runoff data refers to the runoff data obtained from the historical data of the target basin.

[0027] It should be noted that the time of the preset historical period is later than the time of the historical hydrometeorological data and historical climate index data, that is, the historical runoff return data is the data simulated based on the historical hydrometeorological data and historical climate index data, and belongs to the future prediction data corresponding to the historical hydrometeorological data and historical climate index data, while the historical runoff data is the measured runoff data of the same historical period as the historical runoff return data, that is, the time of the historical runoff data is later than the time of the historical hydrometeorological data and historical climate index data, and the time of the historical runoff data is at least 3 days later than the time of the historical hydrometeorological data and historical climate index data. For example, the historical hydrometeorological data and historical climate index data are the data from January 1st to March 5th, and the historical runoff return data and historical runoff data are the data after March 9th of the same year. Among them, the historical runoff return data in this embodiment refers to the simulated medium- and long-term runoff data, and is multiple medium- and long-term runoff data simulated by different runoff simulation models based on the historical hydrometeorological data and historical climate index data.

[0028] S2: Train a number of initial runoff prediction models based on the historical hydrometeorological data, the historical climate index data, the historical runoff return data, and the historical runoff data to obtain a number of runoff prediction models.

[0029] In a feasible embodiment, step S2 includes:

[0030] S21: Use the historical hydrometeorological data, the historical climate index data, and the historical runoff return data as input samples, and use the historical runoff data as output samples.

[0031] S22: Divide the input samples and the output samples to obtain input training samples, input verification samples, output training samples and output verification samples.

[0032] S23: Training the plurality of initial runoff prediction models according to the input training samples and the output training samples to obtain a plurality of candidate runoff prediction models.

[0033] S24: Validating and optimizing the plurality of candidate runoff prediction models according to the input validation samples and the output validation samples to obtain the plurality of runoff prediction models.

[0034] In this embodiment, in order to improve the prediction accuracy of each runoff prediction model, the historical hydrometeorological data, the historical climate index data, the multiple historical runoff report data and the historical runoff data are divided into training samples and verification samples; wherein, the training samples are used to train the initial runoff prediction model, and the verification samples are used to verify and optimize the trained initial runoff prediction model to obtain a runoff prediction model with high prediction accuracy.

[0035] S3: Input the watershed hydrometeorological data, watershed climate index data and multiple runoff return data of the target watershed into several runoff prediction models to obtain runoff data prediction data.

[0036] The basin hydrometeorological data refers to the hydrometeorological data of the target basin in the recent period of time, and the basin climate index data refers to the basin climate index data of the target basin in the recent period of time; wherein, the preset recent period of time can refer to the recent days, the recent weeks or the recent months. Multiple runoff return data increase the reference data of the runoff prediction model for predicting future runoff changes, reduce the impact of the prediction time on the prediction results, and especially obtain high-accuracy medium- and long-term runoff data prediction data. Among them, the medium- and long-term refers to the prediction time of several weeks or months.

[0037] Compared with the prior art, the present application first obtains the historical hydrometeorological data, historical climate index data, historical runoff data of multiple preset historical periods of the target watershed, and the runoff data of the preset historical period simulated according to the historical hydrometeorological data and historical climate index data, and then performs model training based on the historical hydrometeorological data, the historical climate index data, the historical runoff return data and the historical runoff data to obtain several runoff prediction models, and then corrects and optimizes the multiple runoff return data based on the watershed hydrometeorological data, watershed climate index data and flow prediction model of the target watershed, so as to obtain multiple runoff prediction data with higher accuracy.

[0038] Among them, the prediction result range can be increased through multiple runoff prediction data, or the average value, maximum value, median value, etc. of multiple runoff prediction data can be taken to obtain the final runoff prediction data with the highest accuracy from multiple runoff prediction data.

[0039] Please refer to Figure 2 , in a feasible embodiment, each of the initial runoff prediction models includes a long short-term memory network encoder, a first fully connected layer, a repeat input layer, a long short-term memory network decoder, a time distribution layer, and a second fully connected layer. Among them, the long short-term memory network encoder is the LSTM Encoder, the first fully connected layer is the Dense_1, the repeat input layer is the RepeatVector, the long short-term memory network decoder is the LSTM Decoder, the time distribution layer is the TimeDistributed, and the second fully connected layer is the Dense_2.

[0040] As Figure 2 shown, during training, multiple historical runoff reanalysis data of the target basin are the original members of the runoff forecast, and the historical runoff data are the new members of the runoff forecast; during application, multiple runoff reanalysis data of the target basin are the original members of the runoff forecast, and multiple runoff prediction data are the new members of the runoff forecast.

[0041] The step S2: training several initial runoff prediction models according to the historical hydrometeorological data, the historical climate index data, the multiple historical runoff reanalysis data, and the historical runoff data to obtain several runoff prediction models includes:

[0042] S201: Input the historical hydrometeorological data and the historical climate index data into the long short-term memory network encoder for encoding to obtain the historical hydrometeorological time series vector and the historical climate index time series vector output by the long short-term memory network encoder.

[0043] Among them, the role of the long short-term memory network encoder is to encode the input sequence into a state with a fixed shape, and this state contains all the information of the input sequence. Among them, the historical hydrometeorological data and the historical climate index data are respectively the input sequences of the long short-term memory network encoder.

[0044] S202: Input the monthly time series of the historical runoff reanalysis data into the first fully connected layer to obtain the reanalysis data time series information output by the first fully connected layer.

[0045] Among them, the first fully connected layer is used to integrate the features of the input data. For example, when inputting the monthly time series of the historical runoff reanalysis data into the first fully connected layer, the first fully connected layer integrates the features of the monthly time series of the historical runoff reanalysis data to obtain the reanalysis data time series information.

[0046] S203: Input the historical hydrometeorological time series vector, the historical climate index time series vector, and the time series information of the return data into the repeated input layer for time dimension repeated expansion processing to obtain the time series expansion information output by the repeated input layer.

[0047] Among them, the role of the repeated input layer is to repeat the input sequence to expand the receptive field of the time step. Specifically, the repeated input layer will repeat the input sequence a specified number of times in the time dimension, thereby expanding the length of the time step to make the length of the time step reach the preset foresight period length. For example, if the foresight period length is set to n months, the repeated input layer will repeat the input sequence a specified number of times in the time dimension until the length of the time step reaches n months. After the repeated input layer expands the length of the time step, the repeated input layer outputs the time series expansion information in the form of a three-dimensional tensor.

[0048] S204: Input the time series expansion information into the long short-term memory network decoder for decoding to obtain the time series features output by the long short-term memory network decoder.

[0049] The role of the long short-term memory network decoder is to decode the intermediate representation information generated by the encoder to generate the target sequence features.

[0050] S205: Input the time series features into the second fully connected layer through the time distribution layer to apply the second fully connected layer to each time step of the time series features to obtain the training prediction data.

[0051] Among them, the role of the time distribution layer is to apply an ordinary layer to each time step of the time series data. In this embodiment, the time distribution layer applies the second fully connected layer to each time step of the time series features. Among them, the time distribution layer will enable the second fully connected layer to process the time series features at each time step of the time series features, rather than after the entire time series features are completed, so that the initial runoff prediction model can better perform training and learning based on the dynamic changes in time of the historical hydrometeorological data, historical climate index data, historical runoff return data, and historical runoff data to obtain training prediction data with high accuracy.

[0052] S206: Adjust the model parameters of the corresponding initial runoff prediction model according to the data differences between each training prediction data and the historical runoff data to obtain a plurality of the runoff prediction models.

[0053] In this embodiment, through the initial runoff prediction model including the long short-term memory network encoder, the first fully connected layer, the repeated input layer, the long short-term memory network decoder, the time distribution layer, and the second fully connected layer, the time series features of the training data can be extracted and dynamically learned, and a runoff prediction model with high prediction accuracy can be obtained.

[0054] In a feasible embodiment, step S204 includes:

[0055] S2041: Construct a loss function based on the training prediction data and the historical runoff data to obtain a data difference.

[0056] Among them, the loss function refers to the function formula for calculating the root mean square error of the training prediction data and the historical runoff data, that is, the root mean square error calculated by it is the data difference.

[0057] S2042: If the data difference is greater than a preset difference threshold, adjust the model parameters of the corresponding initial runoff prediction model to update the training prediction data until the data difference is less than or equal to the difference threshold, and obtain the corresponding runoff prediction model.

[0058] Among them, the initial runoff prediction model after each adjustment of the model parameters can output new training prediction data, and then obtain a new data difference. When the new data difference is less than or equal to the preset difference threshold, it means that the prediction result of the initial runoff prediction model after adjusting the model parameters has a small error from the actual result. Therefore, it can be used as the runoff prediction model.

[0059] In this embodiment, a data difference is obtained by constructing a loss function, and then it is judged whether to continue model training according to the data difference and the difference threshold, and a runoff prediction model with high prediction accuracy is obtained.

[0060] Please refer to Figure 3 , a runoff prediction data acquisition device 100 is provided in the second embodiment of the present application, including:

[0061] A historical data acquisition module 101, configured to acquire historical hydrometeorological data, historical climate index data, historical runoff return data of multiple preset historical periods, and historical runoff data of multiple said preset historical periods in a target basin; among them, the historical runoff return data is runoff data of a preset historical period simulated by several different runoff simulation models according to the historical hydrometeorological data and historical climate index data; the historical runoff data is the measured runoff data of the historical period corresponding to the historical runoff return data;

[0062] A prediction model acquisition module 102, configured to train several initial runoff prediction models according to the historical hydrometeorological data, the historical climate index data, the historical runoff return data, and the historical runoff data, and obtain several runoff prediction models;

[0063] The runoff prediction data acquisition module 103 is used to input the basin hydrometeorological data, basin climate index data, and multiple runoff return data of the target basin into a number of runoff prediction models to obtain runoff data prediction data.

[0064] In a feasible embodiment, each of the initial runoff prediction models includes a long short-term memory network encoder, a first fully connected layer, a repeated input layer, a long short-term memory network decoder, a time distribution layer, and a second fully connected layer;

[0065] The prediction model acquisition module 102 includes:

[0066] The time series vector extraction sub-module is used to input the historical hydrometeorological data and historical climate index data into the long short-term memory network encoder for encoding to obtain the historical hydrometeorological time series vector and historical climate index time series vector output by the long short-term memory network encoder;

[0067] The return data time series information acquisition sub-module is used to input the monthly time series of the historical runoff return data into the first fully connected layer to obtain the return data time series information output by the first fully connected layer;

[0068] The time series expansion sub-module is used to input the historical hydrometeorological time series vector, historical climate index time series vector, and the return data time series information into the repeated input layer for time dimension repeated expansion processing to obtain the time series expansion information output by the repeated input layer;

[0069] The time series feature acquisition module is used to input the time series expansion information into the long short-term memory network decoder for decoding to obtain the time series features output by the long short-term memory network decoder;

[0070] The training prediction data acquisition sub-module is used to input the time series features through the time distribution layer into the second fully connected layer to apply the second fully connected layer to each time step of the time series features to obtain training prediction data;

[0071] The prediction model acquisition sub-module is used to adjust the model parameters of the corresponding initial runoff prediction model according to the data difference between each training prediction data and the historical runoff data to obtain a number of the runoff prediction models.

[0072] In a feasible embodiment, the prediction model acquisition sub-module includes:

[0073] The data difference acquisition unit is used to obtain the data difference between the training prediction data and the historical runoff data according to the loss function;

[0074] A model parameter update unit, configured to adjust model parameters of a corresponding initial runoff prediction model if the data difference is greater than a preset difference threshold, so as to update the training prediction data until the data difference is less than or equal to the difference threshold, and obtain the corresponding runoff prediction model.

[0075] In a feasible embodiment of sterol, the prediction model acquisition module 102 includes:

[0076] A sample acquisition sub-module, configured to use the historical hydrometeorological data, the historical climate index data, and the historical runoff return data as input samples, and use the historical runoff data as output samples;

[0077] A sample division sub-module, configured to divide the input samples and the output samples to obtain input training samples, input validation samples, output training samples, and output validation samples;

[0078] A model training sub-module, configured to train the several initial runoff prediction models according to the input training samples and the output training samples to obtain several candidate runoff prediction models;

[0079] A model verification and optimization sub-module, configured to verify and optimize the several candidate runoff prediction models according to the input validation samples and the output validation samples to obtain the several runoff prediction models.

[0080] It should be noted that when the runoff prediction device provided in the second embodiment of the present application executes the runoff prediction method, only the above-mentioned division of each functional module is used for illustration. In practical applications, the above functions can be allocated to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the runoff prediction device provided in the second embodiment of the present application and the runoff prediction method in the first embodiment of the present application belong to the same concept, and the implementation process thereof is detailed in the method embodiment and will not be elaborated here.

[0081] A third aspect of the embodiments of the present application provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the steps of the runoff prediction method described above are implemented.

[0082] A fourth aspect of the embodiments of the present application provides a computer device, including a storage, a processor, and a computer program stored in the storage and executable by the processor. When the processor executes the computer program, the steps of the runoff prediction method described above are implemented.

[0083] The device embodiments described above are merely illustrative. The components 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 to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this application. Those of ordinary skill in the art can understand and implement it without creative efforts.

[0084] Those skilled in the art should understand that the embodiments of the present application can be provided as a method, a system, or a computer program product. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present application can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0085] The present application is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the selected functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks. These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured article including an instruction device, and the instruction device implements the selected functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0086] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the selected functions in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0087] In a typical configuration, a computing device includes one or more processors (CPUs), an input / output interface, a network interface, and memory.

[0088] The memory may include non-permanent memory in the form of computer-readable media, random access memory (RAM), and / or non-volatile memory such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0089] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology for information storage. The information can be computer-readable instructions, data structures, program modules, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile discs (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices, or any other non-transitory media that can be used to store information that can be accessed by a computing device. As defined herein, computer-readable media does not include transitory media such as modulated data signals and carrier waves.

[0090] It should also be noted that the term "comprising", "including", or any other variant thereof is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but also other elements not expressly listed, or elements that are inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the statement "comprising an..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that comprises the element.

[0091] The above are only embodiments of the present application and are not used to limit the present application. For those skilled in the art, various changes and modifications can be made to the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included within the scope of the claims of the present application.

Claims

1. A runoff prediction method, characterized in that: include: Acquire historical hydrological and meteorological data, historical climate index data, historical runoff return data of multiple preset historical periods, and historical runoff data of multiple preset historical periods of the target watershed; wherein the historical runoff return data is the runoff data of the preset historical period simulated by several different runoff simulation models based on the historical hydrological and meteorological data and the historical climate index data; the historical runoff data is the measured runoff data of the historical period corresponding to the historical runoff return data; Training a plurality of initial runoff prediction models according to the historical hydrological and meteorological data, the historical climate index data, the historical runoff return data and the historical runoff data to obtain a plurality of runoff prediction models; The watershed hydrometeorological data, watershed climate index data and multiple runoff return data of the target watershed are input into several runoff prediction models to obtain runoff data prediction data.

2. The runoff prediction method according to claim 1, characterized in that: Each of the initial runoff prediction models includes a long short-term memory network encoder, a first fully connected layer, a repeated input layer, a long short-term memory network decoder, a time distribution layer, and a second fully connected layer; The step of training a plurality of initial runoff prediction models according to the historical hydrological and meteorological data, the historical climate index data, the plurality of historical runoff return data and the historical runoff data to obtain a plurality of runoff prediction models comprises: Inputting historical hydrological and meteorological data and historical climate index data into the long short-term memory network encoder for encoding, obtaining a historical hydrological and meteorological time series vector and a historical climate index time series vector output by the long short-term memory network encoder; Inputting the monthly time series of the historical runoff return data into the first fully connected layer to obtain the time series information of the return data output by the first fully connected layer; Inputting the historical hydrological and meteorological time series vector, the historical climate index time series vector, and the feedback data time series information into the repeated input layer for repeated expansion processing of the time dimension to obtain the time series expansion information output by the repeated input layer; Inputting the time series extension information into the long short-term memory network decoder for decoding to obtain the time series features output by the long short-term memory network decoder; Inputting the time series features into the second fully connected layer through the time distribution layer, so as to apply the second fully connected layer to each time step of the time series features to obtain training prediction data; According to the data difference between each of the training prediction data and the historical runoff data, the model parameters of the corresponding initial runoff prediction model are adjusted to obtain a plurality of the runoff prediction models.

3. The runoff prediction method according to claim 2, characterized in that: According to the data difference between the training prediction data and the historical runoff data, the model parameters of the corresponding initial runoff prediction model are adjusted to obtain a plurality of the runoff prediction models, including: According to the loss function, obtaining the data difference between the training prediction data and the historical runoff data; If the data difference is greater than a preset difference threshold, the model parameters of the corresponding initial runoff prediction model are adjusted to update the training prediction data until the data difference is less than or equal to the difference threshold, thereby obtaining the corresponding runoff prediction model.

4. The runoff prediction method according to claim 1, characterized in that: The step of training a plurality of initial runoff prediction models according to the historical hydrological and meteorological data, the historical climate index data, the historical runoff return data and the historical runoff data to obtain a plurality of runoff prediction models comprises: Taking the historical hydrological and meteorological data, the historical climate index data and the historical runoff return data as input samples, and taking the historical runoff data as output samples; Dividing the input sample and the output sample to obtain an input training sample, an input verification sample, an output training sample, and an output verification sample; Training the plurality of initial runoff prediction models according to the input training samples and the output training samples to obtain a plurality of candidate runoff prediction models; The plurality of candidate runoff prediction models are verified and optimized according to the input verification samples and the output verification samples to obtain the plurality of runoff prediction models.

5. A runoff prediction device, characterized in that: include: A historical data acquisition module, used to acquire historical hydrological and meteorological data, historical climate index data, historical runoff return data of multiple preset historical periods, and historical runoff data of multiple preset historical periods of the target basin; wherein the historical runoff return data is the runoff data of the preset historical period simulated by several different runoff simulation models based on the historical hydrological and meteorological data and the historical climate index data; the historical runoff data is the measured runoff data of the historical period corresponding to the historical runoff return data; A prediction model acquisition module, used for training a plurality of initial runoff prediction models according to the historical hydrological and meteorological data, the historical climate index data, the historical runoff return data and the historical runoff data, so as to obtain a plurality of runoff prediction models; The runoff prediction data acquisition module is used to input the watershed hydrological and meteorological data, watershed climate index data and multiple runoff return data of the target watershed into several runoff prediction models to obtain runoff data prediction data.

6. The runoff prediction device according to claim 1, characterized in that: Each of the initial runoff prediction models includes a long short-term memory network encoder, a first fully connected layer, a repeated input layer, a long short-term memory network decoder, a time distribution layer, and a second fully connected layer; The prediction model acquisition module includes: A time series vector extraction submodule, used for inputting historical hydrological and meteorological data and historical climate index data into the long short-term memory network encoder for encoding, and obtaining the historical hydrological and meteorological time series vector and the historical climate index time series vector output by the long short-term memory network encoder; A report data time series information acquisition submodule, used for inputting the monthly time series of the historical runoff report data into the first fully connected layer to obtain the report data time series information output by the first fully connected layer; A time series expansion submodule, used for inputting the historical hydrological and meteorological time series vector, the historical climate index time series vector, and the time series information of the feedback data into the repeated input layer for repeated expansion processing in the time dimension, and obtaining the time series expansion information output by the repeated input layer; A time series feature acquisition module, used for inputting the time series extension information into the long short-term memory network decoder for decoding, and obtaining the time series features output by the long short-term memory network decoder; A training prediction data acquisition submodule, used for inputting the time series features into the second fully connected layer through the time distribution layer, so as to apply the second fully connected layer to each time step of the time series features to obtain training prediction data; The prediction model acquisition submodule is used to adjust the model parameters of the corresponding initial runoff prediction model according to the data difference between each of the training prediction data and the historical runoff data, so as to obtain a plurality of the runoff prediction models.

7. The runoff prediction device according to claim 6, characterized in that: The prediction model acquisition submodule includes: A data difference acquisition unit, used for acquiring the data difference between the training prediction data and the historical runoff data according to the loss function; The model parameter updating unit is used to adjust the model parameters of the corresponding initial runoff prediction model if the data difference is greater than a preset difference threshold, so as to update the training prediction data until the data difference is less than or equal to the difference threshold, thereby obtaining the corresponding runoff prediction model.

8. The runoff prediction device according to claim 5, characterized in that: The prediction model acquisition module includes: A sample acquisition submodule, used to take the historical hydrological and meteorological data, the historical climate index data and the historical runoff return data as input samples, and take the historical runoff data as output samples; A sample division submodule, used for dividing the input sample and the output sample to obtain an input training sample, an input verification sample, an output training sample and an output verification sample; A model training submodule, used for training the several initial runoff prediction models according to the input training samples and the output training samples to obtain several candidate runoff prediction models; The model verification and optimization submodule is used to verify and optimize the several candidate runoff prediction models according to the input verification samples and the output verification samples to obtain the several runoff prediction models.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the runoff prediction method according to any one of claims 1 to 4 are implemented.

10. A computer device, characterized in that: The invention comprises a memory, a processor and a computer program stored in the memory and executable by the processor, wherein the processor implements the steps of the runoff prediction method according to any one of claims 1 to 4 when executing the computer program.

Images