CN118211722A - Water level prediction method and device, storage medium and computer program product - Google Patents

Abstract

The application discloses a water level prediction method and device, a storage medium and a computer program product, and relates to the field of water level prediction, wherein the water level prediction method comprises the following steps: preprocessing monitoring data of a target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data; inputting a first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with the first group of historical data for calculating the historical runoff of a target hydropower station as an input sample and the historical runoff as an output sample; inputting the runoff predicted value and the second group of calculation data into a water level predicted model to obtain a water level predicted value. By adopting the technical scheme, the problem of how to improve the accuracy of the water level prediction result is solved.

Description

Water level prediction method and device, storage medium and computer program product

Technical Field

The application belongs to the field of water level prediction, and particularly relates to a water level prediction method and device, a storage medium and a computer program product.

Background

Hydropower stations are important components in an electric power system, and reservoir water level data of the hydropower stations can guide management and scheduling of water resources, so that scientific decision basis is provided for reservoir operation, hydroelectric power generation and other works, and water level prediction of the hydropower stations is very important. The current water level prediction research method has two methods, one is a traditional time sequence prediction method, is suitable for short-term water level prediction, has poor effect when predicting water level data with complex nonlinear changes, and the other is a deep neural network time sequence prediction method, has advantages in processing nonlinear time sequence data, is suitable for long-term water level prediction, but needs enough calculation data and time to train, so that the short-term prediction effect is poor. The existing water level prediction model generally guides prediction by a single water level prediction algorithm, and the problem of poor accuracy of a prediction result generally exists.

Aiming at the problem of how to improve the accuracy of the water level prediction result in the related art, no effective solution is proposed at present.

Accordingly, there is a need for improvements in the related art to overcome the drawbacks of the related art.

Disclosure of Invention

The embodiment of the application provides a water level prediction method and device, a storage medium and a computer program product, which at least solve the problem of how to improve the accuracy of a water level prediction result in the related technology.

According to an aspect of the embodiment of the present application, there is provided a water level prediction method including: preprocessing monitoring data of a target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data; inputting the first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with the first group of historical data for calculating the historical runoff of the target hydropower station as an input sample and the historical runoff as an output sample; and inputting the runout predicted value and the second group of calculation data into a water level predicted model to obtain a water level predicted value.

In an exemplary embodiment, inputting the first set of calculation data into a traffic prediction model to obtain a traffic prediction value includes: inputting the first group of calculation data into a first sub-model in the runout prediction model to obtain a runout initial prediction value, wherein the model type of the first sub-model is a seasonal autoregressive moving average model; inputting the initial runoff predicted value and the first group of calculation data into a second sub-model in the runoff prediction model to obtain a runoff prediction residual value, wherein the model type of the second sub-model is a long-term and short-term memory model; and determining the sum value of the runoff initial predicted value and the runoff predicted residual value as the runoff predicted value.

In an exemplary embodiment, the second sub-model is trained by: calculating the difference value between the historical runoff value and the initial runoff predicted value to obtain a runoff residual value; and determining the first group of historical data and the historical runoff value as input samples of a first initial model, determining the runoff residual value as output samples of the first initial model, and performing model training on the first initial model to obtain the second sub model.

In an exemplary embodiment, inputting the runoff amount predicted value and the second set of calculation data into a water level prediction model to obtain a water level predicted value includes: inputting the runout predicted value and the second group of calculation data into a third sub-model in the water level predicted model to obtain a water level initial predicted value, wherein the model type of the third sub-model is a seasonal autoregressive moving average model; inputting the water level initial predicted value and the second group of calculated data into a fourth sub-model in the water level predicted model to obtain a water level predicted residual value, wherein the model type of the fourth sub-model is a long-period and short-period memory model; and determining the sum value of the water level initial predicted value and the water level predicted residual value as the water level predicted value.

In an exemplary embodiment, the fourth sub-model is trained by: acquiring a historical water level value of the target hydropower station; calculating the difference value between the historical water level value and the initial water level predicted value to obtain a water level residual value; and determining a second set of historical data for calculating the historical water level value and the historical water level value as input samples of a second initial model, determining the water level residual value as output samples of the second initial model, and performing model training on the second initial model to obtain the fourth sub-model.

In an exemplary embodiment, preprocessing monitoring data of a target hydropower station to obtain standard monitoring data includes: acquiring monitoring data of different indexes acquired by the target hydropower station in a preset time period; acquiring an index monitoring value corresponding to each index from the monitoring data, calculating a standard deviation corresponding to the index monitoring value, and calculating an average value of a plurality of index monitoring values; traversing the index monitoring values, and calculating an offset difference of each index monitoring value in the index monitoring values, wherein the offset difference is obtained by dividing a difference between each index monitoring value and the average value by the standard deviation; and generating the standard monitoring data according to a plurality of offset differences corresponding to the different indexes.

In one exemplary embodiment, determining a first set of calculation data for calculating the runoff amount of the target hydropower station and a second set of calculation data for calculating the water level of the target hydropower station from the standard monitoring data includes: determining first data corresponding to a first index from the standard monitoring data; calculating a first correlation between the first data and the historical runoff value, marking the first data as first candidate data under the condition that the first correlation is larger than a first preset value, and determining a plurality of first candidate data as the first group of calculation data; determining second data corresponding to a second index from the standard monitoring data; and calculating a second correlation between the second data and the historical water level value, marking the second data as second candidate data under the condition that the second correlation is larger than a second preset value, and determining a plurality of second candidate data as the second group of calculation data.

According to still another aspect of the embodiment of the present application, there is also provided a water level prediction apparatus including: the data determining module is used for preprocessing the monitoring data of the target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data; the first prediction module is used for inputting the first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with the first group of historical data for calculating the historical runoff of the target hydropower station as an input sample and the historical runoff as an output sample; and the second prediction module is used for inputting the runout predicted value and the second group of calculation data into a water level prediction model to obtain a water level predicted value.

According to a further aspect of embodiments of the present application, there is also provided a computer readable storage medium having a computer program stored therein, wherein the computer program is arranged to perform the above-described water level prediction method when run.

According to still another aspect of the embodiment of the present application, there is also provided an electronic device including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein the processor executes the water level prediction method through the computer program.

According to a further aspect of embodiments of the present application, there is also provided a computer program product comprising a computer program which, when executed by a processor, implements the steps of the method described in the various embodiments of the application.

According to the application, the monitoring data of the hydropower station can be preprocessed, then a first group of calculation data related to calculating the runoff and a second group of calculation data for calculating the water level are respectively determined, the first group of calculation data is input into the runoff prediction model to obtain a runoff predicted value, and then the obtained runoff predicted value and the second group of calculation data are input into the water level prediction model together to obtain the water level predicted value. The advantages of the short-term prediction model and the long-term prediction model are combined, the runoff of the hydropower station is predicted firstly, and then the precise prediction of the reservoir water level is realized by combining the runoff of the hydropower station, so that the problem of how to improve the accuracy of the water level prediction result in the related technology is solved.

Drawings

The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the application and together with the description, serve to explain the principles of the application.

In order to more clearly illustrate the embodiments of the application or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.

FIG. 1 is a block diagram of a hardware configuration of a computer terminal of a water level prediction method according to an embodiment of the present application;

FIG. 2 is a flowchart of a water level prediction method according to an embodiment of the present application;

FIG. 3 is a schematic flow chart of a water level prediction method according to an embodiment of the present application;

fig. 4 is a block diagram illustrating a water level predicting apparatus according to an embodiment of the present application.

Detailed Description

In order that those skilled in the art will better understand the present application, a technical solution in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings in which it is apparent that the described embodiments are only some embodiments of the present application, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present application without making any inventive effort, shall fall within the scope of the present application.

It should be noted that the terms "first," "second," and the like in the description and the claims of the present application and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the application described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

The method embodiments provided in the embodiments of the present application may be executed in a computer terminal or similar computing device. Taking the example of running on a computer terminal, fig. 1 is a block diagram of a hardware structure of a computer terminal of a water level prediction method according to an embodiment of the present application. As shown in fig. 1, a computer terminal may include one or more (only one is shown in fig. 1) processors 102 (the processor 102 may include, but is not limited to, a microprocessor (Central Processing Unit, MCU), a programmable logic device (Field Programmable GATE ARRAY, FPGA), etc.) and a memory 104 for storing data, where the computer terminal may also include a transmission device 106 for communication functions and an input-output device 108. It will be appreciated by those skilled in the art that the configuration shown in fig. 1 is merely illustrative and is not intended to limit the configuration of the computer terminal described above. For example, the computer terminal may also include more or fewer components than shown in FIG. 1, or have a different configuration than shown in FIG. 1.

The memory 104 may be used to store a computer program, for example, a software program of application software and a module, such as a computer program corresponding to a water level prediction method in an embodiment of the present application, and the processor 102 executes the computer program stored in the memory 104 to perform various functional applications and data processing, that is, implement the above-mentioned method. Memory 104 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include memory remotely located relative to the processor 102, which may be connected to the computer terminal via a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

A wireless network provided by a communication provider of the computer terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, simply referred to as a NIC) that can connect to other network devices through a base station to communicate with the internet. In one example, the transmission device 106 may be a Radio Frequency (RF) module, which is configured to communicate with the internet wirelessly.

In this embodiment, a water level prediction method is provided, fig. 2 is a flowchart of a water level prediction method according to an embodiment of the present application, and as shown in fig. 2, the flowchart includes the following steps:

Step S202, preprocessing monitoring data of a target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data;

optionally, in step S202, the monitored data of the target hydropower station includes water condition data and environmental data affecting the water condition data, including, but not limited to, time, precipitation, temperature, humidity, upstream runoff, hydropower station power consumption, reservoir water level, etc.

Step S204, inputting the first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with a first group of historical data for calculating the historical runoff of the target hydropower station as an input sample and the historical runoff as an output sample;

And S206, inputting the runoff predicted value and the second set of calculation data into a water level predicted model to obtain a water level predicted value.

Through the steps, the monitoring data of the hydropower station can be standardized, then a first group of calculation data related to the calculated runoff and a second group of calculation data for calculating the water level are respectively determined, the first group of calculation data is input into a runoff prediction model to obtain a runoff predicted value, and then the obtained runoff predicted value and the second group of calculation data are input into the water level prediction model together to obtain the water level predicted value. Therefore, the problem of how to improve the accuracy of the water level prediction result in the related technology is solved.

In an exemplary embodiment, for the implementation process of inputting the first set of calculation data into the traffic prediction model in the step S204 to obtain the traffic prediction value, the implementation process specifically includes: inputting the first set of calculation data into a runoff prediction model to obtain a runoff prediction value, including: inputting the first group of calculation data into a first sub-model in the runout prediction model to obtain a runout initial prediction value, wherein the model type of the first sub-model is a seasonal autoregressive moving average model; inputting the initial runoff predicted value and the first group of calculation data into a second sub-model in the runoff prediction model to obtain a runoff prediction residual value, wherein the model type of the second sub-model is a long-term and short-term memory model; and determining the sum value of the runoff initial predicted value and the runoff predicted residual value as the runoff predicted value.

Optionally, in the above embodiment, the first sub-model is a model for predicting the runoff amount by using SARIMA, where SARIMA includes 7 parameters, and p, d, and q are 3 parameters that mainly reflect the trend of the sequence, and P, D, Q, S mainly reflects the periodicity of the sequence. The SARIMA (P, D, Q, P, D, Q, S) model is expressed as:

in the above formula, B represents input data of the model, P represents an autoregressive order of the trend, D represents a trend differential order, Q represents a moving average order of the trend, P represents a seasonal autoregressive order, D represents a seasonal differential order, Q represents a seasonal moving average order, and S is a period length. Representing the sequence after trend differentiation,/>Representing the sequence after seasonal differentiation. In the above formula, P represents the autoregressive order of the trend, D represents the trend differential order, Q represents the moving average order of the trend, P represents the seasonal autoregressive order, D represents the seasonal differential order, Q represents the seasonal moving average order, and S is the period length.Representing the sequence after trend differentiation,/>Representing the sequence after seasonal differentiation.

Optionally, in the above embodiment, the stability test of the runoff sequence data needs to be performed during the SARIMA modeling, and the stability of the sequence is determined by observing the time sequence diagram, the autocorrelation diagram, and the unit root test result, and the trend difference and the season difference are performed. And then judging the order of the sequence according to the autocorrelation diagrams and the partial autocorrelation diagrams. And estimating model parameters according to the runoff quantity sequence data, performing significance test of the model parameters and randomness test of residual errors, and judging the effectiveness of the model.

Optionally, in the above embodiment, the long short term memory model (LSTM) is a special recurrent neural network structure, and the structure mainly includes a memory unit and a gating mechanism. Specifically, LSTM consists of several key parts: and the memory unit is responsible for storing and transmitting the long-term information in the sequence. This is one of the main differences between LSTM and traditional RNN, which enables LSTM to handle longer sequences and avoid gradient vanishing problems. An input gate controls the amount of new information entering the memory cell. It decides which information should be remembered based on the current input and the previous state. This is typically achieved by a sigmoid function, the output of which is between 0 and 1, indicating the extent to which the information is retained. Forget gate to decide which information to discard from the memory unit. It also uses a sigmoid function and decides which information should be forgotten based on the current input and the previous state. And an output gate controlling information output from the memory unit to the current output of the LSTM. It decides what information to output based on the state of the memory cell and the current input. Through these gating mechanisms, LSTM can selectively retain and forget information, thereby enabling efficient modeling of long-term dependencies. This structure provides a significant advantage for LSTM when processing sequence data with long-term dependencies.

In an exemplary embodiment, the second sub-model may be trained by: calculating the difference value between the historical runoff value and the initial runoff predicted value to obtain a runoff residual value; and determining the first group of historical data and the historical runoff value as input samples of a first initial model, determining the runoff residual value as output samples of the first initial model, and performing model training on the first initial model to obtain the second sub model.

In an exemplary embodiment, for the implementation process of inputting the runoff amount predicted value and the second set of calculation data into the water level prediction model to obtain the water level predicted value in the step S206, the implementation process specifically includes: inputting the runout predicted value and the second group of calculation data into a third sub-model in the water level predicted model to obtain a water level initial predicted value, wherein the model type of the third sub-model is a seasonal autoregressive moving average model; inputting the water level initial predicted value and the second group of calculated data into a fourth sub-model in the water level predicted model to obtain a water level predicted residual value, wherein the model type of the fourth sub-model is a long-period and short-period memory model; and determining the sum value of the water level initial predicted value and the water level predicted residual value as the water level predicted value.

Optionally, in the above embodiment, the third sub-model is a model for predicting the reservoir water level by using SARIMA, and before starting to build the SARIMA water level model, the method needs to prepare the runoff time series data, perform stability test on the runoff time series data, observe the stability of the time series graph, the autocorrelation graph, and the unit root test result judgment sequence, and perform trend difference and season difference. And then judging the order of the sequence according to the autocorrelation diagrams and the partial autocorrelation diagrams. And estimating model parameters according to the runoff quantity sequence data, performing significance test of the model parameters and randomness test of residual errors, and judging the effectiveness of the model. Through this series of steps, a SARIMA water level model, suitable for specific time series data, can be trained for future water level predictions and analyses.

In an exemplary embodiment, the fourth sub-model may be trained by: acquiring a historical water level value of the target hydropower station; calculating the difference value between the historical water level value and the initial water level predicted value to obtain a water level residual value; and determining a second set of historical data for calculating the historical water level value and the historical water level value as input samples of a second initial model, determining the water level residual value as output samples of the second initial model, and performing model training on the second initial model to obtain the fourth sub-model.

In an exemplary embodiment, the monitoring data of the target hydropower station is preprocessed in step S202 to obtain standard monitoring data, where the following steps are performed: acquiring monitoring data of different indexes acquired by the target hydropower station in a preset time period; acquiring an index monitoring value corresponding to each index from the monitoring data, calculating a standard deviation corresponding to the index monitoring value, and calculating an average value of a plurality of index monitoring values; traversing the index monitoring values, and calculating an offset difference of each index monitoring value in the index monitoring values, wherein the offset difference is obtained by dividing a difference between each index monitoring value and the average value by the standard deviation; and generating the standard monitoring data according to a plurality of offset differences corresponding to the different indexes.

Optionally, in the foregoing embodiment, for example, the indexes monitored on the target hydropower station include precipitation, temperature, relative humidity, upstream runoff, hydropower station power consumption, and reservoir water level, and the specific index monitoring data are shown in table 1:

TABLE 1

Taking the precipitation amount in table 1 as an example, the average value and standard deviation of all the data of precipitation amount x were calculated first, and the average value mean (x) of precipitation amount was 65.38 and standard deviation std (x) was 38.66. For all index monitoring values of the precipitation x, offset difference calculation is performed by the following formula:

For example, for a precipitation index monitoring value of number 1 of 25.2, the offset difference is-1.04 by the above formula, and so on, the offset differences corresponding to all index monitoring values in table 1 can be obtained, and standard monitoring data is generated according to all offset differences.

In an exemplary embodiment, the step S202 determines a first set of calculation data for calculating the runoff amount of the target hydropower station and a second set of calculation data for calculating the water level of the target hydropower station from the standard monitoring data, and the following steps are performed: determining first data corresponding to a first index from the standard monitoring data; calculating a first correlation between the first data and the historical runoff value, marking the first data as first candidate data under the condition that the first correlation is larger than a first preset value, and determining a plurality of first candidate data as the first group of calculation data; determining second data corresponding to a second index from the standard monitoring data; and calculating a second correlation between the second data and the historical water level value, marking the second data as second candidate data under the condition that the second correlation is larger than a second preset value, and determining a plurality of second candidate data as the second group of calculation data.

Alternatively, in the above embodiment, the first index and the second index may be the same or different. Taking the index of precipitation as an example, if the calculated first correlation of precipitation is greater than a first preset value, the first set of calculated data includes precipitation data, and if the calculated second correlation of precipitation is greater than a second preset value, the second set of data also includes precipitation data.

Alternatively, in the above embodiment, the method for calculating the correlation of data may use pearson correlation coefficient method, where pearson correlation coefficient is the most commonly used correlation measurement method for measuring the strength and direction of the linear relationship between two variables. The value range of the pearson correlation coefficient is between-1 and 1, where-1 represents a complete negative correlation, 1 represents a complete positive correlation, and 0 represents no correlation.

It will be apparent that the embodiments described above are merely some, but not all, embodiments of the application. In order to better understand the above water level prediction method, the following description will explain the above process with reference to the embodiments, but is not intended to limit the technical solution of the embodiments of the present application, specifically:

in an alternative embodiment, the water level prediction method of the present application is further described with reference to fig. 3. The specific process is as follows:

And (3) data processing: firstly, monitoring data of a hydropower station are collected, wherein the monitoring data comprise time, precipitation, temperature, humidity, upstream runoff, hydropower station power consumption, reservoir water level and the like. And calculating the average value and standard deviation of each index, and carrying out standardized calculation to obtain a deviation value. As shown in the following formula, x represents an index monitor value of the index, mean (x) represents an average value of the index calculated, and std (x) represents a standard deviation of the index. x ^′ represents the deviation value corresponding to the index monitoring value x.

Constructing a runoff quantity prediction model: the runoff amount is predicted by SARIMA (corresponding to the first sub-model), and the runoff amount initial predicted value (predicted runoff amount in FIG. 3) of SARIMA is subtracted from the history data of the runoff amount to obtain a runoff amount residual value of the runoff amount initial predicted value. The runoff residual values were predicted using LSTM (corresponding to the second sub-model). And taking the precipitation amount, the temperature, the humidity, the upstream runoff amount, the predicted runoff amount of SARIMA (corresponding to the initial predicted value of the runoff amount), the residual value of the runoff amount and the historical data of the runoff amount as LSTM input, taking the future data of the runoff amount as LSTM output, constructing an LSTM network and performing model training. The LSTM network is used to learn the residual values of the fitting predicted runoff amount. And adding the initial runoff predicted value of SARIMA and the residual runoff value predicted by LSTM to obtain a final runoff predicted value.

Constructing a water level prediction model: the reservoir level is predicted by SARIMA (corresponding to a third sub-model), and the water level residual value data of the water level initial predicted value is obtained by subtracting the water level initial predicted value of SARIMA (the predicted reservoir water level of FIG. 3) from the history data of the reservoir water level. The water level residual value is predicted by using LSTM (corresponding to the fourth sub model) to obtain a water level predicted residual value (reservoir water level residual in figure 3). The historical data of runoff, the runoff predicted value output by the runoff prediction model, the power consumption of the hydropower station, the water level initial predicted value of SARIMA, the water level predicted residual value and the historical data of reservoir water level are used as inputs of LSTM, the future data of reservoir water level is used as outputs of LSTM, an LSTM network is constructed, and model training is carried out. The LSTM network is used to learn the fit water level prediction residual value. And adding the initial water level predicted value of SARIMA and the LSTM water level predicted residual value to obtain a final water level predicted value.

Alternatively, in the above embodiment, the set of prediction data obtained by the above water level prediction method is shown in table 2:

TABLE 2

By the water level prediction method provided by the embodiment, the comprehensive prediction model can be constructed based on the SARIMA model and the LSTM model, and the reservoir water level value of the hydropower station can be accurately calculated, so that the water resource utilization efficiency of the hydropower station is improved.

From the description of the above embodiments, it will be clear to a person skilled in the art that the method according to the above embodiments may be implemented by means of software plus the necessary general hardware platform, but of course also by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising several instructions for causing a terminal device (which may be a mobile phone, a computer, a server, or a network device, etc.) to perform the method of the various embodiments of the present application.

In this embodiment, a water level prediction apparatus is further provided, and the apparatus is used to implement the foregoing embodiments and preferred embodiments, and will not be described in detail. As used below, the term "module" may be a combination of software and/or hardware that implements a predetermined function. While the devices described in the following embodiments are preferably implemented in software, implementations in hardware, or a combination of software and hardware, are also possible and contemplated.

Fig. 4 is a block diagram illustrating a water level prediction apparatus according to an embodiment of the present application, the apparatus including:

The data determining module 42 is configured to pre-process the monitoring data of the target hydropower station to obtain standard monitoring data, and determine a first set of calculation data for calculating the runoff amount of the target hydropower station and a second set of calculation data for calculating the water level of the target hydropower station from the standard monitoring data;

A first prediction module 44, configured to input the first set of calculation data to a runoff prediction model to obtain a runoff prediction value, where the runoff prediction model is obtained by training with a first set of historical data for calculating a historical runoff of the target hydropower station as an input sample and the historical runoff as an output sample;

The second prediction module 46 is configured to input the runout predicted value and the second set of calculation data into a water level prediction model to obtain a water level predicted value.

Through the device, the monitoring data of the hydropower station can be standardized, then a first group of calculation data related to the calculated runoff and a second group of calculation data for calculating the water level are respectively determined, the first group of calculation data is input into a runoff prediction model to obtain a runoff predicted value, and then the obtained runoff predicted value and the second group of calculation data are input into the water level prediction model together to obtain the water level predicted value. Therefore, the problem of how to improve the accuracy of the water level prediction result in the related technology is solved.

In an exemplary embodiment, the first prediction module 44 is further configured to input the first set of calculation data into a first sub-model in the runout prediction model to obtain a runout initial prediction value, where a model type of the first sub-model is a seasonal autoregressive moving average model; inputting the initial runoff predicted value and the first group of calculation data into a second sub-model in the runoff prediction model to obtain a runoff prediction residual value, wherein the model type of the second sub-model is a long-term and short-term memory model; and determining the sum value of the runoff initial predicted value and the runoff predicted residual value as the runoff predicted value.

In an exemplary embodiment, the first prediction module 44 is further configured to train to obtain the second sub-model by: calculating the difference value between the historical runoff value and the initial runoff predicted value to obtain a runoff residual value; and determining the first group of historical data and the historical runoff value as input samples of a first initial model, determining the runoff residual value as output samples of the first initial model, and performing model training on the first initial model to obtain the second sub model.

In an exemplary embodiment, the second prediction module 46 is further configured to input the runout prediction value and the second set of calculation data into a third sub-model in the water level prediction model, to obtain an initial water level prediction value, where a model type of the third sub-model is a seasonal autoregressive moving average model; inputting the water level initial predicted value and the second group of calculated data into a fourth sub-model in the water level predicted model to obtain a water level predicted residual value, wherein the model type of the fourth sub-model is a long-period and short-period memory model; and determining the sum value of the water level initial predicted value and the water level predicted residual value as the water level predicted value.

In an exemplary embodiment, the second prediction module 46 is further configured to obtain a historical water level value of the target hydropower station; calculating the difference value between the historical water level value and the initial water level predicted value to obtain a water level residual value; and determining a second set of historical data for calculating the historical water level value and the historical water level value as input samples of a second initial model, determining the water level residual value as output samples of the second initial model, and performing model training on the second initial model to obtain the fourth sub-model.

In an exemplary embodiment, the data determining module 42 is further configured to obtain monitoring data of different indexes acquired by the target hydropower station in a preset period of time; acquiring an index monitoring value corresponding to each index from the monitoring data, calculating a standard deviation corresponding to the index monitoring value, and calculating an average value of a plurality of index monitoring values; traversing the index monitoring values, and calculating an offset difference of each index monitoring value in the index monitoring values, wherein the offset difference is obtained by dividing a difference between each index monitoring value and the average value by the standard deviation; and generating the standard monitoring data according to a plurality of offset differences corresponding to the different indexes.

In an exemplary embodiment, the data determining module 42 is further configured to determine, from the standard monitoring data, first data corresponding to a first index; calculating a first correlation between the first data and the historical runoff value, marking the first data as first candidate data under the condition that the first correlation is larger than a first preset value, and determining a plurality of first candidate data as the first group of calculation data; determining second data corresponding to a second index from the standard monitoring data; and calculating a second correlation between the second data and the historical water level value, marking the second data as second candidate data under the condition that the second correlation is larger than a second preset value, and determining a plurality of second candidate data as the second group of calculation data.

Embodiments of the present application also provide a computer readable storage medium having a computer program stored therein, wherein the computer program is arranged to perform the steps of any of the method embodiments described above when run.

Alternatively, in the present embodiment, the above-described storage medium may be configured to store a computer program for performing the steps of:

S1, preprocessing monitoring data of a target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data;

S2, inputting the first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with the first group of history data for calculating the history runoff of the target hydropower station as an input sample and the history runoff as an output sample;

S3, inputting the runoff predicted value and the second group of calculation data into a water level predicted model to obtain a water level predicted value.

In one exemplary embodiment, the computer readable storage medium may include, but is not limited to: a usb disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory RAM), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing a computer program.

Specific examples in this embodiment may refer to the examples described in the foregoing embodiments and the exemplary implementation, and this embodiment is not described herein.

An embodiment of the application also provides an electronic device comprising a memory having stored therein a computer program and a processor arranged to run the computer program to perform the steps of any of the method embodiments described above.

Alternatively, in the present embodiment, the above-described processor may be configured to execute the following steps by a computer program:

S1, preprocessing monitoring data of a target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data;

S2, inputting the first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with the first group of history data for calculating the history runoff of the target hydropower station as an input sample and the history runoff as an output sample;

S3, inputting the runoff predicted value and the second group of calculation data into a water level predicted model to obtain a water level predicted value.

In an exemplary embodiment, the electronic apparatus may further include a transmission device connected to the processor, and an input/output device connected to the processor.

Embodiments of the present application also provide a computer program product comprising a non-transitory computer readable storage medium storing a computer program product which, when executed by a processor, implements the steps of the method described in the various embodiments of the application.

Alternatively, in this embodiment, the above computer program may be configured to, when executed by a processor, implement the steps of:

S1, preprocessing monitoring data of a target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data;

S2, inputting the first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with the first group of history data for calculating the history runoff of the target hydropower station as an input sample and the history runoff as an output sample;

S3, inputting the runoff predicted value and the second group of calculation data into a water level predicted model to obtain a water level predicted value.

Specific examples in this embodiment may refer to the examples described in the foregoing embodiments and the exemplary implementation, and this embodiment is not described herein.

It will be appreciated by those skilled in the art that the modules or steps of the application described above may be implemented in a general purpose computing device, they may be concentrated on a single computing device, or distributed across a network of computing devices, they may be implemented in program code executable by computing devices, so that they may be stored in a storage device for execution by computing devices, and in some cases, the steps shown or described may be performed in a different order than that shown or described herein, or they may be separately fabricated into individual integrated circuit modules, or multiple modules or steps of them may be fabricated into a single integrated circuit module. Thus, the present application is not limited to any specific combination of hardware and software.

The foregoing is merely a preferred embodiment of the present application and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present application, which are intended to be comprehended within the scope of the present application.

Claims (10)

1. A water level prediction method, comprising:

preprocessing monitoring data of a target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data;

Inputting the first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with the first group of historical data for calculating the historical runoff of the target hydropower station as an input sample and the historical runoff as an output sample;

And inputting the runout predicted value and the second group of calculation data into a water level predicted model to obtain a water level predicted value.

2. The method of claim 1, wherein inputting the first set of calculated data into a traffic prediction model to obtain a traffic prediction value comprises:

Inputting the first group of calculation data into a first sub-model in the runout prediction model to obtain a runout initial prediction value, wherein the model type of the first sub-model is a seasonal autoregressive moving average model;

inputting the initial runoff predicted value and the first group of calculation data into a second sub-model in the runoff prediction model to obtain a runoff prediction residual value, wherein the model type of the second sub-model is a long-term and short-term memory model;

and determining the sum value of the runoff initial predicted value and the runoff predicted residual value as the runoff predicted value.

3. The method according to claim 2, characterized in that the second sub-model is trained by:

calculating the difference value between the historical runoff value and the initial runoff predicted value to obtain a runoff residual value;

and determining the first group of historical data and the historical runoff value as input samples of a first initial model, determining the runoff residual value as output samples of the first initial model, and performing model training on the first initial model to obtain the second sub model.

4. The method of claim 1, wherein inputting the runoff amount predicted value and the second set of calculation data into a water level prediction model to obtain a water level predicted value comprises:

inputting the runout predicted value and the second group of calculation data into a third sub-model in the water level predicted model to obtain a water level initial predicted value, wherein the model type of the third sub-model is a seasonal autoregressive moving average model;

Inputting the water level initial predicted value and the second group of calculated data into a fourth sub-model in the water level predicted model to obtain a water level predicted residual value, wherein the model type of the fourth sub-model is a long-period and short-period memory model;

and determining the sum value of the water level initial predicted value and the water level predicted residual value as the water level predicted value.

5. The method of claim 4, wherein the fourth sub-model is trained by:

Acquiring a historical water level value of the target hydropower station;

Calculating the difference value between the historical water level value and the initial water level predicted value to obtain a water level residual value;

And determining a second set of historical data for calculating the historical water level value and the historical water level value as input samples of a second initial model, determining the water level residual value as output samples of the second initial model, and performing model training on the second initial model to obtain the fourth sub-model.

6. The method of claim 1, wherein preprocessing the monitoring data of the target hydropower station to obtain standard monitoring data comprises:

acquiring monitoring data of different indexes acquired by the target hydropower station in a preset time period;

Acquiring an index monitoring value corresponding to each index from the monitoring data, calculating a standard deviation corresponding to the index monitoring value, and calculating an average value of a plurality of index monitoring values;

Traversing the index monitoring values, and calculating an offset difference of each index monitoring value in the index monitoring values, wherein the offset difference is obtained by dividing a difference between each index monitoring value and the average value by the standard deviation;

and generating the standard monitoring data according to a plurality of offset differences corresponding to the different indexes.

7. The method of claim 1, wherein determining a first set of calculation data for calculating an amount of runoff of the target hydropower station and a second set of calculation data for calculating a water level of the target hydropower station from the standard monitoring data comprises:

determining first data corresponding to a first index from the standard monitoring data;

Calculating a first correlation between the first data and the historical runoff value, marking the first data as first candidate data under the condition that the first correlation is larger than a first preset value, and determining a plurality of first candidate data as the first group of calculation data;

Determining second data corresponding to a second index from the standard monitoring data;

and calculating a second correlation between the second data and the historical water level value, marking the second data as second candidate data under the condition that the second correlation is larger than a second preset value, and determining a plurality of second candidate data as the second group of calculation data.

8. A water level prediction apparatus, comprising:

The data determining module is used for preprocessing the monitoring data of the target hydropower station to obtain standard monitoring data, and determining a first group of calculation data for calculating the runoff amount of the target hydropower station and a second group of calculation data for calculating the water level of the target hydropower station from the standard monitoring data;

The first prediction module is used for inputting the first group of calculation data into a runoff prediction model to obtain a runoff prediction value, wherein the runoff prediction model is obtained by training with the first group of historical data for calculating the historical runoff of the target hydropower station as an input sample and the historical runoff as an output sample;

and the second prediction module is used for inputting the runout predicted value and the second group of calculation data into a water level prediction model to obtain a water level predicted value.

9. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program when run performs the method of any one of claims 1 to 7.

10. A computer program product comprising a computer program, characterized in that the computer program, when being executed by a processor, implements the steps of the method according to any one of claims 1 to 7.