Parallel reservoir water regimen data prediction method, device, equipment, medium and product

Abstract

The invention relates to the technical field of computers, and discloses a parallel reservoir water regimen data prediction method, device and equipment, a medium and a product, and the parallel reservoir water regimen data prediction method comprises the steps: obtaining a water level matrix and a water regimen matrix value time sequence corresponding to historical water regimen data; converting the water level matrix into a space weight matrix; generating a spatial autoregression model containing a first target parameter according to the water regimen matrix value time sequence and the spatial weight matrix; performing probability maximization analysis on a first target parameter in the spatial autoregression model to obtain a target spatial autoregression model containing a second target parameter; and predicting the water regimen data of the parallel reservoirs according to the target space autoregression model to obtain target water regimen data. According to the invention, the water regimen data is predicted by constructing the spatial autoregression model representing the time dependency relationship and the spatial dependency relationship among the variable data of the plurality of reservoirs, so that the accuracy of water regimen data prediction is improved.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to a method, device, equipment, medium and product for predicting water regime data of parallel reservoirs. Background Art

[0002] Parallel reservoirs refer to multiple reservoirs located on different rivers or on the main stream and tributaries of the same river and jointly undertaking water conservancy tasks. The joint operation of parallel reservoirs adjusts the inflow and outflow of each reservoir to make more reasonable and efficient use of water resources. Water regime data is data reflecting the relevant conditions and changes of water bodies. Therefore, accurately predicting the water regime data of multiple reservoirs in parallel reservoirs is of great significance for better playing the role of the joint operation of parallel reservoirs.

[0003] In related technologies, when predicting the water regime data of parallel reservoirs, usually from the perspective of a single reservoir, a certain water regime data of a certain reservoir is predicted. However, there are relatively close geographical locations and similar meteorological conditions among multiple reservoirs of parallel reservoirs, and the water regime data of multiple reservoirs shows a certain correlation, and the water regime data of multiple reservoirs may also change over time. Therefore, the correlation among multiple reservoirs has a certain impact on the water regime data. When using related technologies to predict the water regime data of parallel reservoirs, only predicting from the perspective of a single reservoir, the obtained water regime data is not accurate enough. Summary of the Invention

[0004] In view of this, the present invention provides a method, device, equipment, medium and product for predicting water regime data of parallel reservoirs to solve the problem that the prediction results obtained by using related technologies to predict the water regime data of parallel reservoirs are not accurate enough.

[0005] In a first aspect, the present invention provides a method for predicting water regime data of parallel reservoirs, including: obtaining a water level matrix of multiple reservoirs in parallel reservoirs and a time series of water regime matrix values corresponding to historical water regime data; the water level matrix is a matrix constructed based on preset rules for water level data affecting the operation of multiple reservoirs, and the time series of water regime matrix values is a sequence used to describe the change of historical water regime data over time; converting the water level matrix into a spatial weight matrix for describing the degree of mutual influence between the water level data of multiple reservoirs; generating a spatial autoregressive model containing a first target parameter according to the time series of water regime matrix values and the spatial weight matrix; the spatial autoregressive model is used to characterize the time dependence relationship and spatial dependence relationship between the historical water regime data of multiple reservoirs; performing probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter; predicting the water regime data of parallel reservoirs according to the target spatial autoregressive model to obtain target water regime data.

[0006] The present invention obtains a water level matrix constructed based on preset rules according to water level data affecting the operation of multiple reservoirs in a parallel reservoir and a time series of water regime matrix values corresponding to historical water regime data for describing the change of historical water regime data over time, and converts the water level matrix into a spatial weight matrix for describing the degree of mutual influence between the water level data of multiple reservoirs. By converting the water level matrix into a spatial weight matrix, the present invention quantifies the geographical location and hydrological connection between multiple reservoirs as a spatial weight matrix, taking into account the mutual influence between the water level data of multiple reservoirs and the change of the spatial weight matrix over time, and avoiding the defect of ignoring spatial dependence in the related art. According to the time series of water regime matrix values and the spatial weight matrix, the present invention generates a spatial autoregressive model containing a first target parameter, performs probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter, and predicts the water regime data of the parallel reservoir according to the target spatial autoregressive model to obtain target water regime data. The spatial autoregressive model of the present invention simultaneously introduces the time series of water regime matrix values and the spatial weight matrix, can capture the change trend of the water level of a single reservoir over time, reflect the mutual influence between multiple reservoirs, and simultaneously reflect the time dependence relationship and spatial dependence relationship between multiple reservoirs. Therefore, using the target spatial autoregressive model with parameter processing for water regime data prediction improves the accuracy of water regime data prediction for parallel reservoirs, making the obtained target water regime data more in line with the actual situation.

[0007] In an alternative embodiment, obtaining the water level matrix of multiple reservoirs in the parallel reservoir and the time series of water regime matrix values corresponding to the historical water regime data includes: obtaining the water level data of multiple reservoirs in the parallel reservoir, and constructing a water level matrix according to the water level data according to preset rules; constructing a time series of water regime matrix values according to the historical water regime data of each power station in chronological order.

[0008] By separately constructing the water level matrix and the time series of water regime matrix values, the present invention organizes the water level data of multiple reservoirs and their historical changes, providing a structured data basis for subsequent analysis and facilitating the exploration of potential relationships between data.

[0009] In an alternative embodiment, converting the water level matrix into a spatial weight matrix for describing the degree of mutual influence between the water level data of multiple reservoirs includes: performing a linear transformation of the historical water level matrix at a preset historical moment through a coefficient matrix to obtain a transformation result; determining the current water level matrix relationship at the current moment according to the sum of the transformation result and the error matrix; the preset historical moment is the previous moment of the current moment; and determining the spatial weight matrix according to the current water level matrix relationships of different reservoirs.

[0010] The present invention determines the spatial weight matrix by using linear transformation and related matrices, quantitatively describes the mutual influence degree among different reservoir water level data, and provides key support for accurately constructing a spatial autoregressive model subsequently.

[0011] In an optional implementation manner, a spatial autoregressive model containing a first target parameter is generated according to the time series of water regime matrix values and the spatial weight matrix, including: fusing the first water regime data matrix at the current moment in the time series of water regime matrix values and the first spatial weight matrix at the current moment to obtain a first fusion result; adjusting the first fusion result by using a first diagonal matrix to obtain a contemporaneous prediction term in the spatial dimension; fusing the second water regime data matrix at a preset historical moment in the time series of water regime matrix values and the second spatial weight matrix at the preset historical moment to obtain a second fusion result; adjusting the second fusion result by using a second diagonal matrix to obtain a lag prediction term in the spatial dimension; adjusting the second water regime data matrix at a preset historical moment in the time series of water regime matrix values by using a regression coefficient matrix to obtain a prediction term in the time dimension; and constructing a spatial autoregressive model containing the first target parameter by using the contemporaneous prediction term, the lag prediction term, and the prediction term in the time dimension through a residual matrix and the spatial weight matrix.

[0012] The present invention constructs prediction terms from two dimensions of contemporaneous and lag periods respectively, and constructs a spatial autoregressive model in combination with a regression coefficient matrix, etc. It comprehensively considers the contemporaneous and lag influence relationships in the spatial dimension and the lag relationship in time, effectively captures the dependence characteristics of multiple reservoir variable data in space and time, improves the capturing ability of the spatial autoregressive model for the dynamic changes of water regime data, and makes the prediction of the spatial autoregressive model more reliable and accurate.

[0013] In an optional implementation manner, probability maximization analysis is performed on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter, including: performing vector transformation on the spatial autoregressive model to obtain a vector transformation result; performing logarithmic likelihood function transformation on the vector transformation result to obtain a target logarithmic likelihood function; performing probability maximization processing on the target logarithmic likelihood function to obtain the second target parameter; and substituting the second target parameter into the spatial autoregressive model to obtain the target spatial autoregressive model.

[0014] In an optional implementation manner, the water regime data of the parallel reservoirs are predicted according to the target spatial autoregressive model to obtain target water regime data, including: obtaining preset historical water regime data, and predicting the water regime data of the parallel reservoirs according to the preset historical water regime data and the target spatial autoregressive model to obtain the target water regime data.

[0015] Second aspect, the present invention provides a prediction device for the water regime data of parallel reservoirs, including: a data acquisition module, configured to acquire a water level matrix of multiple reservoirs in the parallel reservoirs and a time series of water regime matrix values corresponding to historical water regime data; the water level matrix is a matrix constructed based on preset rules from the water level data affecting the operation of multiple reservoirs, and the time series of water regime matrix values is a sequence used to describe the change of historical water regime data over time; a weight conversion module, configured to convert the water level matrix into a spatial weight matrix for describing the mutual influence degree between the water level data of multiple reservoirs; a model generation module, configured to generate a spatial autoregressive model containing a first target parameter according to the time series of water regime matrix values and the spatial weight matrix; the spatial autoregressive model is used to characterize the time dependence relationship and spatial dependence relationship between the historical water regime data of multiple reservoirs; a probability analysis module, configured to perform probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter; a water regime prediction module, configured to predict the water regime data of the parallel reservoirs according to the target spatial autoregressive model to obtain target water regime data.

[0016] Third aspect, the present invention provides a computer device, including: a memory and a processor, which are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to execute the prediction method for the water regime data of parallel reservoirs in the first aspect or any corresponding embodiment thereof.

[0017] Fourth aspect, the present invention provides a computer-readable storage medium, on which computer instructions are stored, and the computer instructions are used to cause a computer to execute the prediction method for the water regime data of parallel reservoirs in the first aspect or any corresponding embodiment thereof.

[0018] Fifth aspect, the present invention provides a computer program product, including computer instructions, and the computer instructions are used to cause a computer to execute the prediction method for the water regime data of parallel reservoirs in the first aspect or any corresponding embodiment thereof. Description of the Drawings

[0019] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the related art, the following will briefly introduce the drawings required to be used in the description of the specific embodiments or the related art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained according to these drawings.

[0020] Figure 1 It is a flowchart of the prediction method for the water regime data of parallel reservoirs according to an embodiment of the present invention.

[0021] Figure 2It is a schematic flowchart of another prediction method for the water regime data of parallel reservoirs according to an embodiment of the present invention.

[0022] Figure 3 It is a schematic flowchart of yet another prediction method for the water regime data of parallel reservoirs according to an embodiment of the present invention.

[0023] Figure 4 It is a structural block diagram of a prediction device for the water regime data of parallel reservoirs according to an embodiment of the present invention.

[0024] Figure 5 It is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Specific embodiments

[0025] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0026] The joint operation of parallel reservoirs refers to regulating the inflow and outflow of each reservoir to more reasonably and efficiently utilize water resources. Accurately predicting the water regime data of multiple reservoirs in parallel reservoirs is of great significance for better exerting the joint operation role of parallel reservoirs.

[0027] In the related art, for the method of predicting the water regime data of parallel reservoirs, it mainly starts from the perspective of a single reservoir and separately predicts a certain water regime data of a single reservoir. However, this method is limited to a single perspective and cannot simultaneously consider the lag impact relationship in the time dimension and the interaction between different water regime data.

[0028] In addition, since there are relatively close geographical locations and similar meteorological conditions among multiple reservoirs in parallel reservoirs, the water regime data of these reservoirs show a certain correlation, that is, there is a natural spatial effect among multiple reservoirs in parallel reservoirs. This spatial effect may also change over time. Therefore, it is necessary to consider the time-varying nature of the spatial effect and conduct research and analysis by constructing a time-varying spatial weight matrix. In this process, when constructing the spatial weight matrix based on relevant variable data, the spatial weight matrix may have a certain association with the water regime data to be predicted, resulting in an endogeneity problem of the spatial weight matrix. Therefore, it is particularly important to model and analyze the spatial effect among multiple reservoirs in parallel reservoirs and the time-varying endogeneity of the spatial weight matrix from the perspective of spatial econometrics.

[0029] In the process of complex time series modeling, an important structural form is the water regime matrix-valued time series, that is, the data at each time point presents a matrix form. For example, the water regime data of multiple reservoirs every day can be constructed into a matrix form, and the continuous daily water regime data forms a water regime matrix-valued time series.

[0030] Specifically, due to the natural upstream and downstream relationship between cascade hydropower stations, there will be a certain spatial influence relationship between the upstream and downstream power stations. Therefore, a spatial autoregressive model based on the water regime matrix-valued time series can be used to model the data structure and spatial characteristics between cascade hydropower stations. However, in the study of spatial econometrics, the selection of the spatial weight matrix is crucial. In the related technologies regarding the study of spatial econometrics, most are based on the inverse matrix of geographic information to construct the spatial weight matrix. However, when considering constructing the spatial weight matrix based on the hydrological information of different hydropower stations, the spatial weight matrix may be time-varying, and more importantly, the spatial weight matrix may be endogenous. This means that the spatial weight matrix will change over time and has an endogenous relationship with the water regime matrix-valued time series data. However, the related technologies have not considered water regime prediction from the perspective of spatial econometrics, let alone the time-varying and endogenous problems of the spatial weight matrix.

[0031] An embodiment of the present invention provides a method for predicting water regime data of parallel reservoirs. By constructing a spatial autoregressive model that characterizes the time dependence relationship and spatial dependence relationship between the variable data of multiple reservoirs, the water regime data is predicted to achieve the effect of improving the accuracy of water regime data prediction.

[0032] According to an embodiment of the present invention, there is provided an embodiment of a method for predicting water regime data of parallel reservoirs. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than here.

[0033] In this embodiment, a method for predicting water regime data of parallel reservoirs is provided, which can be used for upper computer devices. Figure 1 It is a flowchart of the method for predicting water regime data of parallel reservoirs according to an embodiment of the present invention, as Figure 1 shown, the process includes the following steps:

[0034] Step S101, obtain the water level matrix of multiple reservoirs in the parallel reservoirs and the water regime matrix-valued time series corresponding to the historical water regime data; the water level matrix is a matrix constructed based on the water level data affecting the operation of multiple reservoirs according to a preset rule, and the water regime matrix-valued time series is a series used to describe the change of historical water regime data over time.

[0035] Among them, the historical water regime data are the data reflecting the relevant states and changes of water bodies in multiple reservoirs of the parallel-connected reservoirs at historical moments. The historical water regime data include the inflow rate, outflow rate, upstream and downstream water levels, etc. of each reservoir at historical moments. In the embodiment of the present invention, the historical water regime data are collected and obtained through the reservoir hydrological monitoring system.

[0036] In some optional embodiments, obtaining the water level matrix of multiple reservoirs in the parallel-connected reservoirs and the time series of water regime matrix values corresponding to the historical water regime data includes: obtaining the water level data of multiple reservoirs in the parallel-connected reservoirs, and constructing a water level matrix according to the water level data according to a preset rule; constructing the time series of water regime matrix values according to the historical water regime data of each power station in chronological order.

[0037] Among them, the water level data are collected and obtained through the reservoir hydrological monitoring system. The preset rule may be to define the columns as the upstream water level and the downstream water level, and the rows as different reservoirs, that is, to arrange the water level data in the order of different reservoirs for rows and the upstream water level and the downstream water level for columns to construct the water level matrix.

[0038] Among them, the historical water regime data are collected and obtained through the reservoir hydrological monitoring system. The rows are defined as different reservoirs, and the columns are defined as the inflow rate, outflow rate, upstream water level and downstream water level. That is, the water regime data are arranged in the order of different reservoirs for rows and the inflow rate, outflow rate, upstream water level and downstream water level for columns respectively to construct the water regime matrix at a certain moment, and the water regime matrices at consecutive moments constitute the time series of water regime matrix values.

[0039] In some optional embodiments, perform a stationarity test on the historical water regime data, adjust the historical water regime data according to the test results, and construct the time series of water regime matrix values according to the adjusted historical water regime data in chronological order.

[0040] Among them, performing a stationarity test on the historical water regime data includes: identifying the outliers in the historical water regime data, removing the outliers, drawing a trend chart according to the historical water regime data after removing the outliers, and performing a stationarity test according to the trend chart; specifically, using a clustering algorithm or the 3-σ principle to identify the outliers in the historical water regime data, and for the data missing and abnormal, using the moving average technique and the exponential smoothing technique for filling and processing; performing a stationarity test according to the trend chart includes: determining whether there are long-term trends, seasonal variations or periodic fluctuations in the trend chart, and combining the unit root test to judge the stationarity of the historical water regime data.

[0041] If the test results show a long-term trend, the first-order or second-order difference method is applied; if there are seasonal changes in the test results, they are handled by deducting the quarterly average; and to remove deterministic trends, one or more regression models about constants and time can be constructed to eliminate linear or higher-order trend components in historical water data.

[0042] In some optional implementations, a water regime matrix value time series is constructed according to the adjusted historical water regime data in chronological order, including: defining rows as different hydropower stations, defining columns as historical water regime data at a certain moment, and constructing a water regime data matrix Y t , that is, each row in the water regime data matrix represents the historical water regime data of a hydropower station at a certain moment, and the water regime matrix value time series {Y t}.

[0043] Step S102: converting the water level matrix into a spatial weight matrix for describing the degree of mutual influence between water level data of multiple reservoirs.

[0044] In some optional embodiments, the water level matrix is converted into a spatial weight matrix for describing the degree of mutual influence between water level data of multiple reservoirs, including: subjecting the historical water level matrix of a preset historical moment to a linear transformation of the coefficient matrix to obtain a transformation result; determining the current water level matrix relationship at the current moment based on the sum of the transformation result and the error matrix; the preset historical moment is the moment before the current moment; determining the spatial weight matrix at the current moment based on the current water level matrix relationship of different reservoirs; wherein, each moment corresponds to a spatial weight matrix.

[0045] Among them, the historical water level matrix of the preset historical moment is subjected to a linear transformation of the coefficient matrix to obtain a transformation result, including: obtaining the transformation result according to the product of the first coefficient matrix and the second coefficient matrix with the historical water level matrix of the preset historical moment; determining the current water level matrix relationship at the current moment according to the sum of the transformation result and the error matrix.

[0046] For example, the current water level matrix relationship at the current moment can be:

[0047] Z t =PQ t-1 Q T +U t

[0048] Among them, Z t is the water level matrix, P is the first coefficient matrix of m×m dimensions, Q is the second coefficient matrix of p×p dimensions, U t is an m×p dimensional error matrix.

[0049] In some alternative embodiments, determining the spatial weight matrix according to the current water level matrix relationship of different reservoirs includes: using a bounded function to combine the current water level matrix relationships of different reservoirs to obtain the spatial weight matrix. Exemplarily, the expression of the spatial weight matrix is:

[0050] W t,ij = h(Z t,i , Z t,j )

[0051] where W t,ij is the element corresponding to reservoir i and reservoir j in the spatial weight matrix at the current time t, which is used to measure the magnitude of the spatial interaction between the two reservoirs, Z t,i is the water level data of reservoir i at the current time t, and Z t,j is the water level data of reservoir j at the current time t.

[0052] Step S103: Generate a spatial autoregressive model containing the first target parameter according to the time series of water regime matrix values and the spatial weight matrix; the spatial autoregressive model is used to characterize the temporal dependence and spatial dependence relationships among the historical water regime data of multiple reservoirs.

[0053] In some alternative embodiments, generating a spatial autoregressive model containing the first target parameter according to the time series of water regime matrix values and the spatial weight matrix includes: fusing the time series of water regime matrix values and the spatial weight matrix to obtain a spatial autoregressive model containing the first target parameter; specifically, constructing a spatial autoregressive model containing the first target parameter according to the first water regime data matrix at the current time in the time series of water regime matrix values, the first spatial weight matrix at the current time, the second water regime data matrix at a preset historical time in the time series of water regime matrix values, the second spatial weight matrix at the preset historical time, and the residual matrix.

[0054] Specifically, according to the time series of water regime matrix values and the spatial weight matrix, a spatial autoregressive model containing a first target parameter is generated, including: fusing the first water regime data matrix at the current moment in the time series of water regime matrix values with the first spatial weight matrix at the current moment to obtain a first fusion result; adjusting the first fusion result using a first diagonal matrix to obtain a contemporaneous prediction term in the spatial dimension; fusing the second water regime data matrix at a preset historical moment in the time series of water regime matrix values with the second spatial weight matrix at the preset historical moment to obtain a second fusion result; adjusting the second fusion result using a second diagonal matrix to obtain a lag prediction term in the spatial dimension; adjusting the second water regime data matrix at a preset historical moment in the time series of water regime matrix values using a regression coefficient matrix to obtain a prediction term in the time dimension; constructing a spatial autoregressive model containing the first target parameter using the contemporaneous prediction term, the lag prediction term, and the prediction term in the time dimension through a residual matrix and the spatial weight matrix.

[0055] Among them, the contemporaneous prediction term, the lag prediction term, and the prediction term in the time dimension are superimposed through the residual matrix and the spatial weight matrix to obtain a spatial autoregressive model containing the first target parameter.

[0056] Among them, the first target parameter is an unknown parameter. Exemplarily, the first target parameter is a first diagonal matrix, a second diagonal matrix, a regression coefficient matrix, etc.

[0057] Step S104, perform a probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter.

[0058] In some optional implementation manners, the probability maximization analysis is performed on the first target parameter in the spatial autoregressive model using the quasi-maximum likelihood method to obtain the second target parameter, and the second target parameter is substituted into the spatial autoregressive model to obtain the target spatial autoregressive model; specifically, the spatial autoregressive model is vector-transformed to obtain a vector transformation result; the vector transformation result is logarithmically likelihood-function-transformed to obtain a target logarithmic likelihood function; the probability maximization process is performed on the target logarithmic likelihood function to obtain the second target parameter; the second target parameter is substituted into the spatial autoregressive model to obtain the target spatial autoregressive model.

[0059] Among them, the second target parameter is a known parameter, and parameter estimation is performed on the first target parameter to obtain a specific value, that is, the second target parameter.

[0060] Step S105, predict the water regime data of the parallel reservoirs according to the target spatial autoregressive model to obtain target water regime data.

[0061] In some alternative embodiments, the water regime data of the parallel reservoirs are predicted according to the target spatial autoregressive model to obtain the target water regime data, including: obtaining the preset historical water regime data, and predicting the water regime data of the parallel reservoirs according to the preset historical water regime data and the target spatial autoregressive model to obtain the target water regime data.

[0062] Among them, the preset historical water regime data is substituted into the corresponding relational expression of the target spatial autoregressive model to obtain the target water regime data.

[0063] The prediction method of the water regime data of the parallel reservoirs provided in this embodiment obtains the water level matrix constructed based on the water level data affecting the operation of multiple reservoirs in the parallel reservoirs according to a preset rule and the time series of the water regime matrix values corresponding to the historical water regime data, which is used to describe the change of the historical water regime data over time. The water level matrix is converted into a spatial weight matrix for describing the degree of mutual influence between the water level data of multiple reservoirs. In the embodiment of the present invention, by converting the water level matrix into a spatial weight matrix, the geographical location and hydrological connection between multiple reservoirs are quantified as the spatial weight matrix, considering the mutual influence between the water level data of multiple reservoirs and the change of the spatial weight matrix over time, avoiding the defect of ignoring spatial dependence in the related art. According to the time series of the water regime matrix values and the spatial weight matrix, the embodiment of the present invention generates a spatial autoregressive model containing the first target parameter, performs probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain the target spatial autoregressive model containing the second target parameter, and predicts the water regime data of the parallel reservoirs according to the target spatial autoregressive model to obtain the target water regime data. The spatial autoregressive model of the embodiment of the present invention simultaneously introduces the time series of the water regime matrix values and the spatial weight matrix, which can not only capture the change trend of the water level of a single reservoir over time, but also reflect the mutual influence between multiple reservoirs, and at the same time reflect the time dependence relationship and spatial dependence relationship between multiple reservoirs. Therefore, using the target spatial autoregressive model with parameter processing for water regime data prediction improves the accuracy of the water regime data prediction of the parallel reservoirs.

[0064] In this embodiment, a prediction method for the water regime data of the parallel reservoirs is provided, which can be used for the upper computer device. Figure 2 It is a flowchart of another prediction method for the water regime data of the parallel reservoirs according to the embodiment of the present invention, as Figure 2 shown, and the process includes the following steps:

[0065] Step S201, obtaining the water level matrix of multiple reservoirs in the parallel reservoirs and the time series of the water regime matrix values corresponding to the historical water regime data; the water level matrix is a matrix constructed based on the water level data affecting the operation of multiple reservoirs according to a preset rule, and the time series of the water regime matrix values is a sequence used to describe the change of the historical water regime data over time. For details, please refer to Figure 1Step S101 of the illustrated embodiment will not be elaborated herein.

[0066] Step S202: Convert the water level matrix into a spatial weight matrix for describing the mutual influence degree among the water level data of multiple reservoirs. For details, please refer to Figure 1 Step S102 of the illustrated embodiment will not be elaborated herein.

[0067] Step S203: Generate a spatial autoregressive model containing a first target parameter according to the time series of water regime matrix values and the spatial weight matrix; the spatial autoregressive model is used to characterize the temporal dependence relationship and spatial dependence relationship among the historical water regime data of multiple reservoirs.

[0068] Specifically, the above step S203 includes:

[0069] Step S2031: Fuse the first water regime data matrix at the current moment and the first spatial weight matrix at the current moment in the time series of water regime matrix values to obtain a first fusion result.

[0070] Step S2032: Adjust the first fusion result by using a first diagonal matrix to obtain a contemporaneous prediction term in the spatial dimension.

[0071] Step S2033: Fuse the second water regime data matrix at a preset historical moment and the second spatial weight matrix at the preset historical moment in the time series of water regime matrix values to obtain a second fusion result.

[0072] Step S2034: Adjust the second fusion result by using a second diagonal matrix to obtain a lagged prediction term in the spatial dimension.

[0073] Step S2035: Adjust the second water regime data matrix at a preset historical moment in the time series of water regime matrix values by using a regression coefficient matrix to obtain a prediction term in the temporal dimension.

[0074] Step S2036: Construct a spatial autoregressive model containing a first target parameter by using the contemporaneous prediction term, the lagged prediction term, and the prediction term in the temporal dimension through a residual matrix and the spatial weight matrix.

[0075] Exemplarily, the relational expression corresponding to the spatial autoregressive model containing the first target parameter is:

[0076] Y t =D(λ1)W t Y t D(λ2)+AY t-1 B T +D(λ3)W t-1 Y t-1 D(λ4)+E t

[0077] Among them, Y t is the first water regime data matrix at the current moment in the water regime matrix value time series, t is the current moment, and W t is the first spatial weight matrix at the current moment, t - 1 is a preset historical moment, and W t-1 is the second spatial weight matrix at the preset historical moment, and Y t-1 is the second water regime data matrix at the preset historical moment in the water regime matrix value time series, A is the first regression coefficient matrix, B is the second regression coefficient matrix, and D(λ r ) = diag(λ r1 ,..., λ rm ) is an m×m - dimensional first diagonal matrix, where r = 1 or 3, D(λ c ) = diag(λ c1 ,..., λ cn ) is an n×n - dimensional second diagonal matrix, where c = 2 or 4, and E t is the residual matrix.

[0078] Among them, is the m×m - dimensional first regression coefficient matrix, and a i (i = 1,..., m) is the i - th row of A, is the n×n - dimensional second regression coefficient matrix, and b i (i = 1,..., n) is the i - th row of B. The first regression coefficient matrix and the second regression coefficient matrix are the coefficient matrices corresponding to the lag term Y t when predicting Y t-1 . D(λ1)W t Y t D(λ2) is the contemporaneous prediction term in the spatial dimension, reflecting the role of the pure spatial effect in prediction. D(λ3)W t-1 Y t-1 D(λ2) is the lag prediction term in the spatial dimension, reflecting the role of the spatial lag autoregressive term in prediction. D(λ3) and D(λ4) are the proportionality parameters reflecting the row dimension and the column dimension respectively.

[0079] In some alternative embodiments, the endogeneity of the spatial weight matrix comes from the residual matrix E t and the error matrix U t . The spatial weight matrix is a function related to U t . If E t ]>is related to U t , the spatial weight matrix will generate endogeneity. Defining then the relational expression corresponding to the spatial autoregressive model containing the first target parameter is:

[0080]

[0081] Among them, Y t is the first water regime data matrix at the current moment in the water regime matrix value time series, t is the current moment, W t is the first spatial weight matrix at the current moment, t - 1 is the preset historical moment, W t-1 is the second spatial weight matrix at the preset historical moment, Y t-1 is the second water regime data matrix at the preset historical moment in the water regime matrix value time series, A is the first regression coefficient matrix, B is the second regression coefficient matrix, D(λ r ) = diag(λ r1 ,..., λ rm ) is an m×m - dimensional first diagonal matrix, where r = 1 or 3, D(λ c ) = diag(λ c1 ,..., λ cn ) is an n×n - dimensional second diagonal matrix, where c = 2 or 4, Z t is the water level matrix at the current moment, Z t-1 is the water level matrix at the preset historical moment, P is an m×m - dimensional first coefficient matrix, Q is a p×p - dimensional second coefficient matrix, is the conversion parameter.

[0082] Step S204, perform probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing the second target parameter.

[0083] Specifically, the above - mentioned step S204 includes:

[0084] Step S2041, perform vector transformation on the spatial autoregressive model to obtain a vector transformation result.

[0085] Among them, perform vector transformation on the relational expression corresponding to the spatial autoregressive model containing the first target parameter to obtain:

[0086]

[0087] Among them, z t = vec(Z t ), I m is the identity matrix, vec is the vectorization operator, W t is the first spatial weight matrix at the current moment, Y t is the first water regime data matrix at the current moment in the water regime matrix value time series, D(λ r ) = diag(λ r1 ,..., λ rm) is the first diagonal matrix of dimension m×m, where r = 1 or 3, D(λ c ) = diag(λ c1 ,..., λ cn ) is the second diagonal matrix of dimension n×n, W t-1 is the second spatial weight matrix at a preset historical moment, Y t-1 is the second water regime data matrix at a preset historical moment in the time series of water regime matrix values, A is the first regression coefficient matrix, B is the second regression coefficient matrix, P is the first coefficient matrix of dimension m×m, Q is the second coefficient matrix of dimension p×p, Z t is the water level matrix at the current moment, Z t-1 is the water level matrix at a preset historical moment, and δ is the conversion coefficient.

[0088] Step S2042: Perform a log-likelihood function transformation on the vector conversion result to obtain the target log-likelihood function.

[0089] The target log-likelihood function under the normal distribution norm can be written as:

[0090]

[0091] where L is the target log-likelihood function, z t = vec(Z t ), T is the time length of the time series of water regime matrix values, represents the Kronecker product. I mn represents the identity matrix of dimension mn, that is, an mn×mn matrix with only diagonal elements being 1 and the rest being 0, m, n, p are matrix dimensions, U is the error matrix of dimension m×p, Z t is the water level matrix at the current moment, Z t-1 is the water level matrix at a preset historical moment, P is the first coefficient matrix of dimension m×m, Q is the second coefficient matrix of dimension p×p, is the conversion parameter.

[0092] Step S2043: Perform probability maximization on the target log-likelihood function to obtain the second target parameter.

[0093] Among them, by maximizing the probability of the target log-likelihood function, a second target parameter is obtained, including: Initializing parameters: An initial value needs to be selected for the first target parameter, and the first-order derivative and second-order derivative of the target log-likelihood function with respect to the first target parameter matrix are calculated. Given other first target parameters, the Newton-Raphson algorithm is used to solve the estimated value of a certain first target parameter matrix; then all the first target parameter matrices are updated in turn, and iteration is continuously performed until the convergence condition is met. The convergence condition is that the number of iterations reaches 100, or the sum of the absolute differences of all the first target parameter matrices in two consecutive iterations is less than 0.001. The parameter that meets the convergence condition is the second target parameter.

[0094] Step S2044: Substitute the second target parameter into the spatial autoregressive model to obtain the target spatial autoregressive model.

[0095] Step S205: Predict the water regime data of the parallel reservoirs according to the target spatial autoregressive model to obtain the target water regime data. For details, please refer to Figure 1 Step S105 of the embodiment shown, which will not be elaborated here.

[0096] The prediction method for the water regime data of the parallel reservoirs provided in this embodiment constructs prediction items from two dimensions of the same period and the lag period respectively, and combines a regression coefficient matrix, etc. to construct a spatial autoregressive model, comprehensively considering the synchronous and lag influence relationships in the spatial dimension, as well as the lag relationship in time, effectively capturing the dependence characteristics of multiple reservoir variable data in space and time, improving the ability of the spatial autoregressive model to capture the dynamic changes of the water regime data, and making the prediction of the spatial autoregressive model more reliable and accurate.

[0097] In this embodiment, a prediction method for the water regime data of the parallel reservoirs is provided, which can be used in an upper computer device. Figure 3 It is a flowchart of another prediction method for the water regime data of the parallel reservoirs according to the embodiment of the present invention. As Figure 3 shown, this process includes:

[0098] Data collection: Collect historical water regime data such as the inflow, outflow, and upstream and downstream water levels of multiple reservoirs in the parallel reservoirs to obtain a water regime prediction variable data set. For details, please refer to Figure 1 Step S101 of the embodiment shown, which will not be elaborated here.

[0099] Data preprocessing: Perform a stationarity test on the historical water regime data in the water regime prediction variable data set, and perform periodic adjustment on the non-stationary time series. For details, please refer to Figure 1 Step S101 of the embodiment shown, which will not be elaborated here.

[0100] Matrix data structure: Arrange the historical water regime data of each reservoir in a parallel reservoir at each moment to obtain a time series of water regime matrix values. For details, please refer to Figure 1 Step S101 of the embodiment shown, which will not be elaborated here.

[0101] Spatial weight matrix construction: Construct a spatial weight matrix based on water level data. For details, please refer to Figure 1 Step S102 of the embodiment shown, which will not be elaborated here.

[0102] Model construction: Construct a spatial autoregressive model of the time series of water regime matrix values with a time-varying endogenous spatial weight matrix. For details, please refer to Figure 1 Step S103 of the embodiment shown, which will not be elaborated here.

[0103] Parameter estimation and prediction: Obtain the model parameter estimation results based on the quasi-maximum likelihood estimation method, and predict the water regime data based on the model parameter estimation results and the model structure. For details, please refer to Figure 1 Steps S104 and S105 of the embodiment shown, which will not be elaborated here.

[0104] In this embodiment, a prediction device for the water regime data of a parallel reservoir is also provided. This device is used to implement the above-mentioned embodiments and preferred implementation manners, and those that have been described will not be elaborated here. As used below, the term "module" can be a combination of software and / or hardware that can achieve a predetermined function. Although the devices described in the following embodiments are preferably implemented in software, implementation in hardware, or a combination of software and hardware is also possible and contemplated.

[0105] This embodiment provides a prediction device for the water regime data of a parallel reservoir, as Figure 4 shown, including:

[0106] A data acquisition module 401, configured to acquire a water level matrix of multiple reservoirs in a parallel reservoir and a time series of water regime matrix values corresponding to historical water regime data; the water level matrix is a matrix constructed based on water level data affecting the operation of multiple reservoirs according to a preset rule, and the time series of water regime matrix values is a sequence used to describe the change of historical water regime data over time.

[0107] A weight conversion module 402, configured to convert the water level matrix into a spatial weight matrix used to describe the mutual influence degree between the water level data of multiple reservoirs.

[0108] A model generation module 403, configured to generate a spatial autoregressive model containing a first target parameter according to the time series of water regime matrix values and the spatial weight matrix; the spatial autoregressive model is used to characterize the time-dependent relationship and spatial-dependent relationship between the historical water regime data of multiple reservoirs.

[0109] A probability analysis module 404 is configured to perform probability maximization analysis on a first target parameter in a spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter.

[0110] A water regime prediction module 405 is configured to predict water regime data of a parallel reservoir according to the target spatial autoregressive model to obtain target water regime data.

[0111] In some alternative embodiments, the data acquisition module 401 includes:

[0112] A water level matrix construction unit is configured to obtain water level data of multiple reservoirs in the parallel reservoir and construct a water level matrix according to the water level data according to a preset rule.

[0113] A time series construction unit is configured to construct a time series of water regime matrix values in chronological order according to historical water regime data of each power station.

[0114] In some alternative embodiments, the weight conversion module 402 includes:

[0115] A linear transformation unit is configured to perform a linear transformation of a historical water level matrix at a preset historical moment through a coefficient matrix to obtain a transformation result.

[0116] An addition unit is configured to determine a current water level matrix relationship at the current moment according to the sum of the transformation result and an error matrix; the preset historical moment is the previous moment of the current moment.

[0117] A weight matrix determination unit is configured to determine a spatial weight matrix according to the current water level matrix relationships of different reservoirs.

[0118] In some alternative embodiments, the model generation module 403 includes:

[0119] A first fusion unit is configured to fuse a first water regime data matrix at the current moment and a first spatial weight matrix at the current moment in the time series of water regime matrix values to obtain a first fusion result.

[0120] A first adjustment unit is configured to adjust the first fusion result by using a first diagonal matrix to obtain a contemporaneous prediction term in the spatial dimension.

[0121] A second fusion unit is configured to fuse a second water regime data matrix at a preset historical moment and a second spatial weight matrix at the preset historical moment to obtain a second fusion result.

[0122] A second adjustment unit is configured to adjust the second fusion result by using a second diagonal matrix to obtain a lag prediction term in the spatial dimension.

[0123] A third adjustment unit, configured to adjust a second water condition data matrix at a preset historical moment in a time series of water condition matrix values by using a regression coefficient matrix, so as to obtain a prediction item in the time dimension.

[0124] A model construction unit, configured to construct a spatial autoregressive model containing a first target parameter by using a contemporaneous prediction item, a lag prediction item, and a prediction item in the time dimension through a residual matrix and a spatial weight matrix.

[0125] In some alternative embodiments, the probability analysis module 404 includes:

[0126] A first conversion unit, configured to perform vector conversion on the spatial autoregressive model to obtain a vector conversion result.

[0127] A second conversion unit, configured to perform logarithmic likelihood function conversion on the vector conversion result to obtain a target logarithmic likelihood function.

[0128] A function processing unit, configured to perform probability maximization processing on the target logarithmic likelihood function to obtain a second target parameter.

[0129] A model determination unit, configured to substitute the second target parameter into the spatial autoregressive model to obtain a target spatial autoregressive model.

[0130] In some alternative embodiments, the water condition prediction module 405 includes:

[0131] A water condition prediction unit, configured to obtain preset historical water condition data, and predict the water condition data of the parallel reservoirs according to the preset historical water condition data and the target spatial autoregressive model to obtain target water condition data.

[0132] The further function descriptions of the above-mentioned modules and units are the same as those in the corresponding foregoing embodiments, and will not be elaborated herein.

[0133] The prediction device for the water condition data of the parallel reservoirs in this embodiment is presented in the form of functional units. Here, the unit refers to an ASIC (Application Specific Integrated Circuit) circuit, a processor and a memory that execute one or more software or fixed programs, and / or other devices that can provide the above functions.

[0134] An embodiment of the present invention further provides a computer device having the above-mentioned Figure 4 prediction device for the water condition data of the parallel reservoirs as shown.

[0135] Please refer to Figure 5 , Figure 5 which is a schematic structural diagram of a computer device provided by an alternative embodiment of the present invention. As shown in Figure 5As shown, the computer device includes: one or more processors 10, a memory 20, and interfaces for connecting the components, including a high-speed interface and a low-speed interface. Each component communicates with each other using different buses and can be installed on a common motherboard or installed in other ways as needed. The processor can process instructions executed within the computer device, including instructions stored in the memory or on the memory to display graphical information of the GUI on an external input / output device (such as a display device coupled to the interface). In some alternative embodiments, if necessary, multiple processors and / or multiple buses can be used together with multiple memories and multiple memories. Similarly, multiple computer devices can be connected, and each device provides some necessary operations (such as an array of servers, a set of blade servers, or a multi-processor system). Figure 5 In the figure, a processor 10 is taken as an example.

[0136] The processor 10 can be a central processing unit, a network processor, or a combination thereof. Among them, the processor 10 can further include a hardware chip. The above hardware chip can be an application-specific integrated circuit, a programmable logic device, or a combination thereof. The above programmable logic device can be a complex programmable logic device, a field programmable gate array, a generic array logic, or any combination thereof.

[0137] Among them, the memory 20 stores instructions executable by at least one processor 10, so that the at least one processor 10 executes the method shown in the above embodiments.

[0138] The memory 20 can include a program storage area and a data storage area. Among them, the program storage area can store an operating system and application programs required for at least one function; the data storage area can store data created according to the use of the computer device, etc. In addition, the memory 20 can include a high-speed random access memory, and can also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage devices. In some alternative embodiments, the memory 20 can optionally include a memory remotely set relative to the processor 10, and these remote memories can be connected to the computer device through a network. Examples of the above network include but are not limited to the Internet, an enterprise intranet, a local area network, a mobile communication network, and combinations thereof.

[0139] The memory 20 can include a volatile memory, such as a random access memory; the memory can also include a non-volatile memory, such as a flash memory, a hard disk, or a solid-state drive; the memory 20 can also include a combination of the above types of memories.

[0140] The computer device further includes a communication interface 30 for the computer device to communicate with other devices or communication networks.

[0141] Embodiments of the present invention also provide a computer-readable storage medium. The method according to the embodiments of the present invention can be implemented in hardware, firmware, or be implemented as computer code that can be recorded on a storage medium, or be implemented as computer code that is originally stored in a remote storage medium or a non-transitory machine-readable storage medium and downloaded through a network and will be stored in a local storage medium, so that the method described herein can be stored as such software processing on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory, a random access memory, a flash memory, a hard disk, or a solid-state drive, etc.; further, the storage medium can also include a combination of the above-mentioned types of memories. It can be understood that a computer, a processor, a microprocessor controller, or programmable hardware includes a storage component that can store or receive software or computer code, and when the software or computer code is accessed and executed by the computer, the processor, or the hardware, the method shown in the above embodiments is implemented.

[0142] A part of the present invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can call or provide the method and / or technical solution according to the present invention through the operation of the computer. Those skilled in the art should be able to understand that the forms of existence of computer program instructions in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executes the instructions, or the computer compiles the instructions and then executes the corresponding compiled program, or the computer reads and executes the instructions, or the computer reads and installs the instructions and then executes the corresponding installed program. Herein, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to the computer.

[0143] Although the embodiments of the present invention are described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the present invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A prediction method for water regime data of parallel reservoirs, characterized in that The method includes: Obtaining a water level matrix of multiple reservoirs in a parallel reservoir and a time series of water regime matrix values corresponding to historical water regime data; the water level matrix is a matrix constructed based on preset rules from water level data affecting the operation of the multiple reservoirs, and the time series of water regime matrix values is a series used to describe the variation of the historical water regime data over time; Converting the water level matrix into a spatial weight matrix for describing the degree of mutual influence between the water level data of the multiple reservoirs; Generating a spatial autoregressive model containing a first target parameter based on the time series of water regime matrix values and the spatial weight matrix; the spatial autoregressive model is used to characterize the temporal dependence relationship and spatial dependence relationship between the historical water regime data of the multiple reservoirs; Performing probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter; Predicting the water regime data of the parallel reservoir according to the target spatial autoregressive model to obtain target water regime data.

2. The method according to claim 1, characterized in that, The obtaining of the water level matrix of multiple reservoirs in a parallel reservoir and the time series of water regime matrix values corresponding to historical water regime data includes: Obtaining the water level data of multiple reservoirs in the parallel reservoir, and constructing the water level matrix according to the water level data according to the preset rules; Constructing the time series of water regime matrix values according to the historical water regime data of each power station in chronological order.

3. The method according to claim 1 or 2, characterized in that, The converting of the water level matrix into a spatial weight matrix for describing the degree of mutual influence between the water level data of the multiple reservoirs includes: Performing a linear transformation of a historical water level matrix at a preset historical moment through a coefficient matrix to obtain a transformation result; Determining the current water level matrix relationship at the current moment according to the sum of the transformation result and an error matrix; the preset historical moment is the previous moment of the current moment; Determining the spatial weight matrix according to the current water level matrix relationships of different reservoirs.

4. The method according to claim 3, wherein The generating of a spatial autoregressive model containing a first target parameter based on the time series of water regime matrix values and the spatial weight matrix includes: Fusing the first water regime data matrix at the current moment and the first spatial weight matrix at the current moment in the time series of water regime matrix values to obtain a first fusion result; Adjusting the first fusion result using a first diagonal matrix to obtain a contemporaneous prediction term in the spatial dimension; Fusing the second water regime data matrix at a preset historical moment and the second spatial weight matrix at the preset historical moment in the time series of water regime matrix values to obtain a second fusion result; Adjusting the second fusion result using a second diagonal matrix to obtain a lagged prediction term in the spatial dimension; Adjusting the second water regime data matrix at a preset historical moment in the time series of water regime matrix values using a regression coefficient matrix to obtain a prediction term in the time dimension; Constructing the spatial autoregressive model containing the first target parameter using the contemporaneous prediction term, the lagged prediction term, and the prediction term in the time dimension through a residual matrix and the spatial weight matrix.

5. The method according to claim 1 or 2, characterized in that, Performing probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter includes: Performing vector transformation on the spatial autoregressive model to obtain a vector transformation result; Performing logarithmic likelihood function transformation on the vector transformation result to obtain a target logarithmic likelihood function; Performing probability maximization processing on the target logarithmic likelihood function to obtain the second target parameter; Substituting the second target parameter into the spatial autoregressive model to obtain the target spatial autoregressive model.

6. The method according to claim 1 or 2, characterized in that Predicting the water regime data of the parallel reservoirs according to the target spatial autoregressive model to obtain target water regime data includes: Obtaining preset historical water regime data, and predicting the water regime data of the parallel reservoirs according to the preset historical water regime data and the target spatial autoregressive model to obtain the target water regime data.

7. A prediction device for the water regime data of a parallel reservoir, characterized in that, The device includes: A data acquisition module, configured to acquire a water level matrix of multiple reservoirs in the parallel reservoirs and a time series of water regime matrix values corresponding to historical water regime data; the water level matrix is a matrix constructed based on preset rules for water level data affecting the operation of the multiple reservoirs, and the time series of water regime matrix values is a series used to describe the change of the historical water regime data over time; A weight transformation module, configured to transform the water level matrix into a spatial weight matrix for describing the mutual influence degree between the water level data of the multiple reservoirs; A model generation module, configured to generate a spatial autoregressive model containing a first target parameter according to the time series of water regime matrix values and the spatial weight matrix; the spatial autoregressive model is used to characterize the temporal dependence and spatial dependence between the historical water regime data of the multiple reservoirs; A probability analysis module, configured to perform probability maximization analysis on the first target parameter in the spatial autoregressive model to obtain a target spatial autoregressive model containing a second target parameter; A water regime prediction module, configured to predict the water regime data of the parallel reservoirs according to the target spatial autoregressive model to obtain target water regime data.

8. A computer device, characterized in that, Including: A memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to execute the prediction method for the water regime data of the parallel reservoirs according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, Computer instructions are stored on the computer-readable storage medium, and the computer instructions are used to cause a computer to execute the prediction method for the water regime data of the parallel reservoirs according to any one of claims 1 to 6.

10. A computer program product, characterized in that, Including computer instructions, and the computer instructions are used to cause a computer to execute the prediction method for the water regime data of the parallel reservoirs according to any one of claims 1 to 6.

Images