A method for coupling meteorological forecast products with medium and long term runoff forecast and related device

Abstract

The application discloses a method for coupling meteorological forecast product medium and long term runoff forecast and related device, relates to runoff forecast technical field, and the method comprises the following steps: firstly, runoff data, meteorological data and climate index data of several time periods before the to-be-predicted time period and precipitation correction forecast data of the to-be-predicted time period are acquired as input data sets; then, after pretreatment and redundant variable screening, the obtained input data sets are input into the optimal runoff forecast model determined through training and comparison, and runoff forecast data of the to-be-predicted time period is obtained; compared with the existing physical hydrological model, the application couples precipitation correction forecast data to carry out runoff prediction, avoids complex hydrological physical calculation, and improves the adaptability and generalization ability of the model; compared with the existing data-driven method, the application couples precipitation correction forecast data to carry out medium and long term runoff forecast, and overcomes the fact that the influence of future meteorology on medium and long term runoff is not fully considered in the existing method.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of runoff prediction, in particular to a long-term runoff prediction method coupled with meteorological prediction products and a related device. BACKGROUND

[0002] Long-term runoff prediction is a key tool for water resources management, and plays a crucial role in solving water supply and demand contradictions, efficient development and optimal allocation of water resources. First of all, it provides valuable long-term reference information for the operation and management of water conservancy projects, which helps to alleviate the contradiction between water storage and water abandonment, and improves the comprehensive utilization efficiency of water resources. Secondly, accurate long-term prediction provides a scientific basis for the optimal operation of reservoirs, helps water conservancy departments to reasonably allocate water, achieve supply and demand balance, and maximize the economic and social benefits of reservoirs. In the process of formulating reservoir water supply plans, river basin water resources scheduling schemes and promoting modern water conservancy construction, long-term runoff prediction is an indispensable foundation support. Therefore, strengthening long-term runoff prediction and improving its accuracy is a key way to realize the optimal allocation of water resources and improve the level of water resources management.

[0003] At present, there are two methods for long-term runoff prediction: process-driven method and data-driven method. The process-driven method describes the hydrological cycle of the basin through physical equations, which has strong scientific explanation, but the modeling process is complex, the calculation cost is high, and there are limitations in dealing with the non-stationarity of future climate change. The data-driven method can effectively mine the nonlinear relationship between data and improve the prediction accuracy, but it generally relies on historical data for training and fails to fully utilize future climate prediction information, resulting in insufficient adaptability to long-term runoff trends. Therefore, there is an urgent need for a prediction method that can combine meteorological prediction products to quickly and accurately predict reservoir monthly inflow runoff data. SUMMARY

[0004] The purpose of the present application is to provide a long-term runoff prediction method coupled with meteorological prediction products and related devices, which can combine meteorological prediction products to quickly and accurately predict reservoir monthly inflow runoff data.

[0005] To achieve the above purpose, the present application provides the following solutions:

[0006] In a first aspect, the present application provides a long-term runoff prediction method coupled with meteorological prediction products, comprising:

[0007] Obtaining runoff data, meteorological data and climate index data of a plurality of time periods before a to-be-predicted time period and precipitation correction prediction data of the to-be-predicted time period to obtain an input data set; the runoff data, meteorological data and climate index data each include a plurality of input variables.

[0008] The meteorological data and the precipitation corrected forecast data are subjected to downscaling processing, and the runoff data and the climate index data are subjected to interpolation completion, so as to obtain a pretreated input data set.

[0009] The LASSO algorithm is used to screen the input variables in the pretreated input data set, so as to remove the redundant input variables, and a screened input data set is obtained.

[0010] The screened input data set is input into an optimal runoff forecast model, so as to obtain the medium and long term runoff forecast data of the to-be-predicted period; the optimal runoff forecast model is a runoff forecast model with the highest prediction accuracy among a plurality of runoff forecast models obtained by training an initial runoff forecast model by using different training sample sets; the input variables included in the samples in the different training sample sets are not completely the same; the input variables included in the samples in the training sample set used for training the optimal runoff forecast model are completely the same as the input variables in the screened input data set.

[0011] Optionally, the method further comprises the following steps:

[0012] According to historical observation data, a training data set is constructed; the training data set includes the precipitation corrected forecast data and the runoff data of a target period, and the runoff data, the meteorological data and the climate index data of a plurality of periods before the target period; wherein, the runoff data, the meteorological data and the climate index data of the plurality of periods before the target period, and the precipitation corrected forecast data of the target period are input data of the model, and the runoff data of the target period is target output data of the model; the target period is any historical period with known runoff data.

[0013] The meteorological data and the precipitation corrected forecast data are subjected to downscaling processing, and the runoff data, the runoff data of a plurality of periods before the target period and the climate index data are subjected to interpolation completion, so as to obtain a pretreated training data set.

[0014] The LASSO algorithm is used to screen the input variables in the pretreated training data set, so as to remove the redundant input variables, and a screened training data set is obtained.

[0015] According to different combinations of the input data in the screened training data set, a plurality of input data combination modes are determined.

[0016] For any one input data combination mode, a training sample set and a verification sample set are constructed by using the screened training data set.

[0017] The training sample set is used to train an initial runoff forecast model, so as to obtain a trained runoff forecast model.

[0018] The trained runoff prediction model is verified by using the verification sample set to determine the prediction accuracy of the runoff prediction model.

[0019] The runoff prediction models are arranged in descending order of prediction accuracy, and the runoff prediction model with the highest prediction accuracy is taken as the optimal runoff prediction model.

[0020] Optionally, the runoff data include reservoir inflow runoff; the meteorological data include precipitation, minimum temperature and maximum temperature of the reservoir upstream catchment area; the climate index data include atmospheric circulation index and sea temperature index; and the precipitation correction prediction data are data obtained by fusing a plurality of precipitation prediction products through Bayesian model averaging.

[0021] Optionally, the initial runoff prediction model is an XGBoost machine learning model, and the hyperparameters of the initial runoff prediction model are automatically optimized by a Bayesian optimization algorithm when the initial runoff prediction model is trained by using the training sample set; the hyperparameters of the initial runoff prediction model include the number of trees, the learning rate, the maximum tree depth, the minimum child weight, the L1 regularization term weight and the L2 regularization term weight.

[0022] Optionally, after obtaining the screened training data set, the method further includes: performing normalization processing on the data in the screened training data set according to the following formula:

[0023]

[0024] wherein, x t is the value of the normalized data x at time t, x t is the value of the data x before normalization at time t, a is the lower limit of the data x, and b is the upper limit of the data x.

[0025] After obtaining the medium and long term runoff prediction data of the to-be-predicted period, the method further includes: performing inverse normalization processing on the medium and long term runoff prediction data of the to-be-predicted period according to the following formula:

[0026] x t = x t '(b-a)+a.

[0027] Optionally, the bilinear interpolation method is used to downscale the meteorological data and the precipitation correction prediction data, and the linear interpolation method is used to interpolate and complete the runoff data and the climate index data.

[0028] In a second aspect, the present application provides a medium and long term runoff prediction system coupled with meteorological prediction products, comprising:

[0029] The input data set acquisition module is configured to acquire runoff data, meteorological data, and climate index data of a plurality of time periods before a to-be-predicted time period and precipitation correction forecast data of the to-be-predicted time period, to obtain an input data set; the runoff data, the meteorological data, and the climate index data each include a plurality of input variables.

[0030] The input data preprocessing module is configured to perform downscaling processing on the meteorological data and the precipitation correction forecast data, and perform interpolation completion on the runoff data and the climate index data, to obtain preprocessed input data set.

[0031] The redundant variable screening module is configured to screen the input variables in the preprocessed input data set by using a LASSO algorithm, to remove redundant input variables, to obtain screened input data set.

[0032] The medium and long term runoff prediction module is configured to input the screened input data set into an optimal runoff prediction model, to obtain medium and long term runoff prediction data of the to-be-predicted time period; the optimal runoff prediction model is a runoff prediction model with the highest prediction accuracy among a plurality of runoff prediction models obtained by training an initial runoff prediction model using different training sample sets; the input variables included in the samples in the different training sample sets are not completely the same; the input variables included in the samples in the training sample set used to train the optimal runoff prediction model are completely the same as the input variables in the screened input data set.

[0033] In a third aspect, the present application provides a computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the medium and long term runoff prediction method coupled with meteorological forecast products.

[0034] In a fourth aspect, the present application provides a computer readable storage medium, which stores a computer program, wherein the computer program is executable on a processor to implement the steps of the medium and long term runoff prediction method coupled with meteorological forecast products.

[0035] In a fifth aspect, the present application provides a computer program product, comprising a computer program, wherein the computer program is executable on a processor to implement the steps of the medium and long term runoff prediction method coupled with meteorological forecast products.

[0036] According to the embodiments provided in the present application, the following technical effects are disclosed:

[0037] The application provides a method for coupling meteorological prediction products to predict medium and long-term runoff and a related device. In the method, runoff data, meteorological data and climate index data of a plurality of time periods before a to-be-predicted time period and precipitation correction prediction data of the to-be-predicted time period are acquired as an input data set. After preprocessing and redundant variable screening, the input data set obtained after screening is input into an optimal runoff prediction model determined through training and comparison to obtain medium and long-term runoff prediction data of the to-be-predicted time period. Compared with existing physical hydrological models, the application couples meteorological prediction products to predict runoff, avoids complex hydrological physical calculation, and improves the adaptability and generalization ability of the model. Compared with existing data-driven methods, the application couples precipitation correction prediction data to predict medium and long-term runoff, and overcomes the problem that the influence of future meteorological conditions on medium and long-term runoff is not fully considered in existing methods. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0039] Figure 1 A flowchart of a method for coupling meteorological prediction products to predict medium and long-term runoff according to an embodiment of the present application.

[0040] Figure 2 A flowchart of training a model and selecting an optimal runoff prediction model in a method for coupling meteorological prediction products to predict medium and long-term runoff according to an embodiment of the present application.

[0041] Figure 3 A functional module schematic diagram of a system for coupling meteorological prediction products to predict medium and long-term runoff according to an embodiment of the present application.

[0042] Figure 4 A structural schematic diagram of a computer device according to an embodiment of the present application. DETAILED DESCRIPTION

[0043] The technical solutions in the embodiments of the present application will be described clearly and completely below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0044] In order to make the above objectives, characteristics and advantages of the present application more obvious and easy to understand, the present application is further described in detail below with reference to the drawings and specific embodiments.

[0045] The method for coupling meteorological forecast products and medium and long term runoff forecast provided by the embodiments of the present application includes the following steps in an exemplary embodiment, as shown in the figure: Figure 1

[0046] A1, obtaining runoff data, meteorological data and climate index data of several time periods before a to-be-predicted time period and precipitation correction forecast data of the to-be-predicted time period, to obtain an input data set; the runoff data, the meteorological data and the climate index data all include several input variables. Specifically, the runoff data includes reservoir inflow runoff; the meteorological data includes precipitation, minimum temperature and maximum temperature of a reservoir upstream catchment area; the climate index data includes atmospheric circulation index and sea temperature index; the precipitation correction forecast data is data obtained by fusing several precipitation forecast products through Bayesian model averaging.

[0047] A2, performing downscaling processing on the meteorological data and the precipitation correction forecast data, and performing interpolation completion on the runoff data and the climate index data, to obtain a preprocessed input data set. Specifically, the bilinear interpolation method is used to perform downscaling processing on the meteorological data and the precipitation correction forecast data, and the linear interpolation method is used to perform interpolation completion on the runoff data and the climate index data. It can be understood that when the runoff data does not have missing values, the interpolation completion processing on the runoff data can be omitted.

[0048] A3, using the LASSO algorithm to screen the input variables in the preprocessed input data set, to remove redundant input variables, to obtain a screened input data set. Before the next step, the Min-Max normalization method can be used to normalize all the input variables, to reduce the scale influence between different data and improve the training convergence speed. Specifically, after obtaining the screened input data set, it further includes: normalizing the data in the screened input data set according to the following formula:

[0049]

[0050] wherein, x t is the value of the data x at time t after normalization, x t is the value of the data x at time t before normalization, a is the lower limit of the data x, and b is the upper limit of the data x.

[0051] ​A4, inputting the screened input data set into the optimal runoff prediction model to obtain the medium and long term runoff prediction data of the to-be-predicted period; the optimal runoff prediction model is a runoff prediction model with the highest prediction accuracy among a plurality of runoff prediction models obtained by training an initial runoff prediction model using different training sample sets; the samples in the different training sample sets include different input variables; the samples in the training sample set used to train the optimal runoff prediction model include the same input variables as those in the screened input data set.

[0052] Correspondingly, after obtaining the medium and long term runoff prediction data of the to-be-predicted period, the method further comprises: performing inverse normalization processing on the medium and long term runoff prediction data of the to-be-predicted period according to the following formula:

[0053] x t = x t '(b-a)+a.

[0054] wherein, x t is the value of the data x at time t after inverse normalization processing, x t ' is the value of the data x at time t before inverse normalization processing, a is the lower limit of the data x, and b is the upper limit of the data x.

[0055] Specifically, before step A4, the process of training the model and selecting the optimal runoff prediction model is also needed, as shown in FIG. 1, the process comprises the following steps: Figure 2

[0056] B1, constructing a training data set according to historical observation data; the training data set includes precipitation correction prediction data and runoff data of a target period, and runoff data, meteorological data and climate index data of a plurality of periods before the target period; wherein, the runoff data, meteorological data and climate index data of a plurality of periods before the target period and the precipitation correction prediction data of the target period are input data of the model, and the runoff data of the target period is the target output data of the model; the target period is any historical period with known runoff data.

[0057] Specifically, in this embodiment, the time lag effect of past runoff data, meteorological data and climate index data on future monthly runoff prediction is considered, in order to ensure that the model can fully utilize historical information and improve prediction accuracy, the plurality of periods before the target period are limited to the past 18 months before the target period.

[0058] ​The runoff data of the target period and the previous runoff data are both reservoir inflow, and the data is derived from the hydrological monitoring station record. The meteorological data includes precipitation, minimum temperature and maximum temperature in the upstream catchment area of the reservoir, and the data is derived from the CN05.1 grid observation data set, and the data resolution is 0.25°*0.25°. Then, the bilinear interpolation is used for regridding to match the spatial scale of the reservoir catchment area. The climate index data includes atmospheric circulation index, sea temperature index and the like, and a total of 111 index data is selected, and the data is derived from the National Climate Center of China Meteorological Administration. The precipitation correction forecast data GCM is derived from the NMME data set, and the precipitation forecast products of five global climate models CanSIPS-IC3, CCSM4, CFSv2, CM2p1-aer04 and CM2p5-FLOR-B01 are selected, and the data resolution is 1°, and the Bayesian model average (BMA) method is used for fusion to generate the GCM precipitation correction forecast data.

[0059] B2, the meteorological data and the precipitation correction forecast data are down-scaled, and the runoff data of the target period, the runoff data of several periods before the target period and the climate index data are interpolated and completed to obtain the pretreated training data set. Specifically, the bilinear interpolation method is used to regriddle the CN05.1 meteorological data and the GCM data to match the spatial resolution of the research basin, so as to match the spatial scale of the research area; the linear interpolation method is used to fill the missing values in the runoff data and the climate index data. It can be understood that when there is no missing value in the runoff data, the interpolation and completion of the runoff data can be omitted.

[0060] B3, the LASSO algorithm is used to screen the input variables in the pretreated training data set, and the redundant input variables are removed to obtain the screened training data set. The LASSO (Least Absolute Shrinkage and Selection Operator) algorithm is used to screen the input variables in the data set, remove redundant variables, and improve the calculation efficiency and prediction accuracy of the model.

[0061] B4, according to different combinations of input data in the screened training data set, a plurality of input data combination modes are determined. Specifically, the runoff data, the meteorological data, the climate index data and the precipitation correction forecast data GCM are arranged and combined to construct a plurality of sample data sets, and correspond to the monthly reservoir inflow runoff data of the target prediction period respectively, forming four kinds of input data combinations, which are used to analyze the influence of the precipitation correction forecast data GCM data on the prediction accuracy, as shown in Table 1:

[0062] Table 1 Influence of precipitation correction forecast data GCM data on prediction accuracy

[0063]

[0064] B5. For any combination of input data, construct a training sample set and a validation sample set using the filtered training dataset. The ratio of the training sample set to the validation sample set is 3:1 to ensure the model's generalization ability.

[0065] B6. The initial runoff forecast model is trained using the training sample set to obtain a trained runoff forecast model. In this embodiment, the initial runoff forecast model is an XGBoost machine learning model. When training the initial runoff forecast model using the training sample set, the hyperparameters of the initial runoff forecast model are automatically optimized using a Bayesian optimization algorithm. The hyperparameters of the initial runoff forecast model include the number of trees, learning rate, maximum tree depth, minimum child node weight, L1 regularization term weight, and L2 regularization term weight.

[0066] B7. Validate the trained runoff forecasting model using the validation sample set to determine the prediction accuracy of the runoff forecasting model.

[0067] B8. Arrange several runoff forecasting models in descending order of prediction accuracy, and select the runoff forecasting model with the highest prediction accuracy as the optimal runoff forecasting model.

[0068] The following section uses the Bilu River Reservoir as a case study to introduce a medium- to long-term runoff forecasting method coupled with meteorological forecasting products provided in this embodiment. The Bilu River main stream is 156 kilometers long, with 100 kilometers located within Dalian. Its total drainage area is 2814 square kilometers, and the reservoir's catchment area is 2085 square kilometers, accounting for 74.1% of the total drainage area. The medium- to long-term runoff forecasting method for the Bilu River Reservoir provided in this example specifically includes the following steps:

[0069] Corresponding to step B1 of the aforementioned scheme, the historical observation data collected in this example covers the period from January 1, 1982 to December 31, 2020. The reservoir's initial inflow runoff data are obtained from hydrological monitoring station records. Monthly precipitation, maximum temperature, and minimum temperature data for the upstream catchment area of ​​the reservoir can be obtained from the CN05.1 gridded observation dataset. Data for 111 indices, including atmospheric circulation index and sea surface temperature index, are sourced from the National Climate Center of the China Meteorological Administration. GCM precipitation correction forecast data are sourced from the NMME dataset.

[0070] Corresponding to step B2 of the aforementioned scheme, this example uses bilinear interpolation to re-grid the CN05.1 meteorological data and GCM data to match the spatial resolution of the study watershed (the Biluhe Reservoir was adjusted to 0.1°), thus matching the spatial scale of the study area. In this example, linear interpolation is used to fill in missing values ​​in the meteorological data, runoff data, and climate index data.

[0071] Corresponding to step B5 of the foregoing scheme, the distribution date of the data in the constructed training sample set is January 1, 1982-December 31, 2010, and the distribution date of the data in the validation sample set is January 1, 2011-December 31, 2020.

[0072] In this embodiment, the forecast period is set to 1 month and 3 months, respectively, to predict the reservoir monthly runoff data of the target number of months, i.e., the medium and long-term runoff forecast data of the to-be-measured time period, by the above scheme.

[0073] Corresponding to step B7 of the foregoing scheme, the period from January 1982 to December 2010 is divided as the model calibration period, and the period from January 2011 to December 2020 is divided as the model validation period. The evaluation indexes include the Nash efficiency coefficient (NSE), the Kling-Gupta efficiency (KGE), the root mean square error (RMSE), and the Pearson correlation coefficient (R), and the evaluation indexes are specifically shown in the following formula:

[0074]

[0075] In the formula, is the observed value at the tth time step; is the predicted value at the tth time step; is the mean value of the observed value; n is the total number of observation data; r is the Pearson correlation coefficient between the model simulation value and the observed value, indicating the linear correlation between the predicted value and the observed value; and a is the bias coefficient between the predicted value and the observed value, used to measure the proportional relationship between the model output and the observation data. σ sim is the standard deviation of the predicted data, σ obs is the standard deviation of the observed data; and β is the mean ratio between the predicted data and the observed data. μ sim is the mean value of the predicted data, μ obs is the mean value of the observed data.

[0076] When the forecast period is 1 month, the evaluation coefficients are shown in Table 2:

[0077] Table 2 Evaluation coefficients of different input data when the forecast period is 1 month

[0078]

[0079] When the forecast period is 3 months, the evaluation coefficients are shown in Table 3:

[0080] Table 3 Evaluation coefficients of different input data when the forecast period is 3 months

[0081]

[0082] In summary, the four evaluation indexes of combination 1 in this example have consistent trends. After adding GCM forecast data, the prediction accuracy of 1-month and 3-month prediction periods is improved, and the improvement of 3-month prediction period is particularly significant, with the KGE index rising by 19%. Through comprehensive analysis of the four combinations, it is found that in the prediction of the monthly inflow runoff of Bilu River Reservoir, the addition of GCM forecast information has a significant improvement effect on the prediction accuracy of 1-month and 3-month prediction periods. In addition, the prediction accuracy of each combination under different prediction periods is quite different: in the 1-month prediction period, combination 4 (with GCM) performs best, with an NSE of 0.646; in the 3-month prediction period, combination 1 (with GCM) has the highest NSE of 0.721. It is proved that the monthly inflow runoff data of the reservoir can be quickly and accurately predicted by coupling meteorological forecast products, and the adaptability to long-term runoff trends is improved.

[0083] Based on the same inventive concept, the embodiments of the present application also provide a system for implementing the above-mentioned method for coupling meteorological forecast products to predict medium and long-term runoff. The implementation scheme of the system for solving the problem is similar to the implementation scheme described in the above method. In one exemplary embodiment, as shown in Figure 3 Fig. 1, a system for coupling meteorological forecast products to predict medium and long-term runoff is provided, which includes the following functional modules:

[0084] An input data set acquisition module is configured to acquire runoff data, meteorological data, and climate index data of a plurality of time periods before a to-be-predicted time period and precipitation correction forecast data of the to-be-predicted time period, and obtain an input data set. The runoff data, the meteorological data, and the climate index data each include a plurality of input variables.

[0085] An input data preprocessing module is configured to perform downscaling processing on the meteorological data and the precipitation correction forecast data, and perform interpolation completion on the runoff data and the climate index data, and obtain a preprocessed input data set.

[0086] A redundant variable screening module is configured to screen the input variables in the preprocessed input data set by using a LASSO algorithm, remove redundant input variables, and obtain a screened input data set.

[0087] A medium and long-term runoff prediction module is configured to input the screened input data set into an optimal runoff prediction model, and obtain medium and long-term runoff prediction data of the to-be-predicted time period. The optimal runoff prediction model is a runoff prediction model with the highest prediction accuracy among a plurality of runoff prediction models obtained by training an initial runoff prediction model using different training sample sets. The samples in the different training sample sets include different input variables. The samples in the training sample set used to train the optimal runoff prediction model include the same input variables as those in the screened input data set.

[0088] Of course, Figure 3 The illustrated architecture is only exemplary, and in actual implementation, some components can be omitted Figure 3 One or at least two components in the illustrated system.

[0089] In an exemplary embodiment, a computer device is provided, which can be a server or a terminal, and its internal structure diagram can be as shown. Figure 4 The computer device includes a processor, a memory, an input / output interface (I / O) and a communication interface. Among them, the processor, the memory and the input / output interface are connected through a system bus, and the communication interface is connected to the system bus through the input / output interface. Among them, the processor of the computer device is used to provide computing and control capability. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operating system and the computer program in the non-volatile storage medium to run. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals through network connection. The computer program is executed by the processor, which can realize the long-term runoff prediction method of coupling meteorological prediction products mentioned in the foregoing embodiments.

[0090] Those skilled in the art can understand, Figure 4 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device to which the scheme of the present application is applied. The specific computer device can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0091] In an exemplary embodiment, a computer device is also provided, which includes a memory and a processor, the memory stores a computer program, and the processor executes the computer program to realize the steps in the above method embodiments.

[0092] In an exemplary embodiment, a computer readable storage medium is provided, which stores a computer program, and the computer program is executed by a processor to realize the steps in the above method embodiments.

[0093] In an exemplary embodiment, a computer program product is provided, which includes a computer program, and the computer program is executed by a processor to realize the steps in the above method embodiments.

[0094] It should be noted that the user information (including but not limited to user equipment information, user personal information, etc.) and data (including but not limited to data for analysis, stored data, displayed data, etc.) involved in the present application are all information and data authorized by the user or authorized by all parties, and the collection, use and processing of related data need to comply with relevant regulations.

[0095] It can be understood by those skilled in the art that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing related hardware through a computer program, and the computer program can be stored in a non-volatile computer readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments of each method. In the embodiments provided in the present application, any reference to memory, database or other medium can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (Read-Only Memory, ROM), magnetic tape, floppy disk, flash memory, optical storage, high-density embedded non-volatile memory, resistive memory (ReRAM), magnetoresistive random access memory (Magnetoresistive Random Access Memory, MRAM), ferroelectric memory (Ferroelectric Random Access Memory, FRAM), phase change memory (Phase Change Memory, PCM), graphene memory, etc. Volatile memory can include random access memory (Random Access Memory, RAM) or external cache memory, etc. As an illustration but not limitation, RAM can be in various forms, such as static random access memory (Static Random Access Memory, SRAM) or dynamic random access memory (Dynamic Random Access Memory, DRAM), etc.

[0096] The database involved in the embodiments provided in the present application can include at least one of a relational database and a non-relational database. The non-relational database can include a distributed database based on a blockchain, etc., without being limited thereto. The processor involved in the embodiments provided in the present application can be a general-purpose processor, a central processing unit, a graphics processing unit, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., without being limited thereto.

[0097] The technical features of the above embodiments can be combined arbitrarily. In order to make the description simple, not all possible combinations of the technical features in the above embodiments are described, but as long as the combination of the technical features does not exist contradictory, it should be considered as the scope of the present application.

[0098] The principles and implementations of the present application are described in detail with specific examples in this paper, and the above examples are only used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the idea of the present application, the specific implementation and application range will be changed. Therefore, the content of the specification should not be understood as a limitation of the present application.

Claims

1. A method of coupling a mid-long term runoff forecast with a weather forecast product, characterized in that, The method comprises the following steps: obtain runoff data, meteorological data and climate index data of several periods before the to-be-predicted period and precipitation correction forecast data of the to-be-predicted period to obtain an input data set; the runoff data, the meteorological data and the climate index data each comprise several input variables; perform downscaling processing on the meteorological data and the precipitation correction forecast data, and perform interpolation completion on the runoff data and the climate index data to obtain a preprocessed input data set; perform screening on the input variables in the preprocessed input data set by using a LASSO algorithm to remove redundant input variables and obtain a screened input data set; input the screened input data set into an optimal runoff prediction model to obtain medium and long term runoff prediction data of the to-be-predicted period; the optimal runoff prediction model is a runoff prediction model with the highest prediction accuracy among several runoff prediction models obtained by training an initial runoff prediction model using different training sample sets; the input variables included in the samples in different training sample sets are not completely the same; the input variables included in the samples in the training sample set used to train the optimal runoff prediction model are completely the same as the input variables in the screened input data set.

2. The coupled weather forecast product mid-long term runoff forecast method according to claim 1, characterized in that, The method further comprises the following steps: construct a training data set according to historical observation data; the training data set comprises precipitation correction forecast data and runoff data of a target period and runoff data, meteorological data and climate index data of several periods before the target period; wherein the runoff data, the meteorological data and the climate index data of several periods before the target period and the precipitation correction forecast data of the target period are input data of the model, and the runoff data of the target period is target output data of the model; the target period is any historical period with known runoff data; perform downscaling processing on the meteorological data and the precipitation correction forecast data, and perform interpolation completion on the runoff data of the target period, the runoff data of several periods before the target period and the climate index data to obtain a preprocessed training data set; perform screening on the input variables in the preprocessed training data set by using a LASSO algorithm to remove redundant input variables and obtain a screened training data set; determine several input data combination modes according to different combinations of input data in the screened training data set; for any input data combination mode, construct a training sample set and a verification sample set by using the screened training data set; train an initial runoff prediction model by using the training sample set to obtain a trained runoff prediction model; verify the trained runoff prediction model by using the verification sample set to determine the prediction accuracy of the runoff prediction model; arrange several runoff prediction models in descending order of prediction accuracy, and take the runoff prediction model with the highest prediction accuracy as an optimal runoff prediction model.

3. The coupled weather forecast product mid-long term runoff forecast method according to claim 2, characterized in that, The runoff data includes the inflow of water into the reservoir; the meteorological data includes precipitation, minimum temperature, and maximum temperature in the upstream catchment area of ​​the reservoir; the climate index data includes the atmospheric circulation index and the sea surface temperature index; and the precipitation correction forecast data is obtained by merging several precipitation forecast products through a Bayesian model.

4. The coupled weather forecast product mid-long term runoff forecast method according to claim 2, characterized in that, The initial runoff forecast model is an XGBoost machine learning model. When training the initial runoff forecast model using the training sample set, the hyperparameters of the initial runoff forecast model are automatically optimized by the Bayesian optimization algorithm. The hyperparameters of the initial runoff forecast model include the number of trees, learning rate, maximum tree depth, minimum child node weight, L1 regularization term weight, and L2 regularization term weight.

5. The coupled weather forecast product mid-long term runoff forecast method according to claim 2, characterized in that, After obtaining the filtered training dataset, the process further includes: normalizing the data in the filtered training dataset according to the following formula: wherein x t is the value of the normalized data x at time t, x t is the value of the data x before normalization at time t, a is the lower limit of the data x, and b is the upper limit of the data x; After obtaining the medium- and long-term runoff forecast data for the period to be predicted, the process also includes: performing inverse normalization on the medium- and long-term runoff forecast data for the period to be predicted according to the following formula: x t = x t '(b-a) + a.

6. The coupled weather forecast product mid-long term runoff forecast method according to claim 1, characterized in that, The meteorological data and the precipitation correction forecast data are downscaled using bilinear interpolation, and the runoff data and the climate index data are interpolated and completed using linear interpolation.

7. A medium-long term runoff forecasting system coupled with weather forecast products, characterized in that, include: The input dataset acquisition module is used to acquire runoff data, meteorological data, and climate index data for several periods before the period to be predicted, as well as precipitation correction forecast data for the period to be predicted, to obtain the input dataset; the runoff data, the meteorological data, and the climate index data all include several input variables; The input data preprocessing module is used to downscale the meteorological data and the precipitation correction forecast data, and to interpolate and complete the runoff data and the climate index data to obtain the preprocessed input dataset. The redundant variable filtering module is used to filter the input variables in the preprocessed input dataset using the LASSO algorithm, remove redundant input variables, and obtain the filtered input dataset. The medium- and long-term runoff forecasting module is used to input the filtered input dataset into the optimal runoff forecasting model to obtain medium- and long-term runoff forecasting data for the period to be predicted. The optimal runoff forecasting model is the runoff forecasting model with the highest prediction accuracy among several runoff forecasting models trained on the initial runoff forecasting model using different training sample sets. The input variables included in the samples in different training sample sets are not completely the same. The input variables included in the samples in the training sample set of the optimal runoff forecasting model are completely the same as the input variables in the filtered input dataset.

8. A computer device comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement a medium- to long-term runoff forecasting method for coupled meteorological forecast products according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, When executed by a processor, the computer program implements the medium- and long-term runoff forecasting method for coupled meteorological forecast products as described in any one of claims 1-6.

10. A computer program product comprising a computer program, characterized in that, The computer program, which is executed by a processor, implements the method of medium- and long-term runoff forecasting coupled with weather forecast products according to any one of claims 1-6.

Images