A method and system for periodic optimization scheduling of water conservancy hub

Abstract

The present invention provides a method and system for periodic optimization scheduling of a water conservancy hub, which relates to the technical field of data processing. The method includes: obtaining historical discharge data of the water conservancy hub; determining the data period of the historical discharge data; mapping the historical discharge data within the data period to a linear space; constructing a discharge prediction function with a noise adjustment term and a deviation adjustment term according to the mapped historical discharge data; determining the state transfer matrix, the noise adjustment term and the deviation adjustment term in the discharge prediction function by a maximum likelihood estimation algorithm, updating the discharge prediction function to replicate the volatility of the historical discharge data; inputting the historical discharge data into the updated discharge prediction function, and outputting the discharge prediction value under a preset time step; reversely mapping the discharge prediction value to a nonlinear space to obtain a real discharge prediction value; and optimizing the scheduling of the water conservancy hub according to the real discharge prediction value. Improving scheduling accuracy and timeliness.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to a method and system for periodic optimization scheduling of a water conservancy hub. Background Art

[0002] A water conservancy hub refers to a comprehensive water conservancy project facility consisting of a series of hydraulic structures, usually including dams, gates, diversion channels, hydropower stations, etc., which are used to regulate river water volume to meet various needs such as flood control, power generation, irrigation, shipping, and water resource supply. Discharge refers to the amount of water discharged to the downstream through a water conservancy project (such as a reservoir, dam, or gate), usually expressed in terms of flow per unit time (such as cubic meters per second, m³ / s). It reflects the process of water conservancy hubs transporting water resources to the downstream and is an important indicator for water conservancy project scheduling.

[0003] Water conservancy hubs need to meet multiple functional objectives at the same time, and the temporal and spatial distribution and demand of water resources are highly uncertain and periodic. Through optimized scheduling, a balance can be found between flood control and water supply, the efficiency of water resource utilization can be improved, the losses caused by floods and droughts can be reduced, and the sustainable development of the ecological environment and social economy can be guaranteed.

[0004] However, the existing water conservancy hub scheduling system often schedules the discharge volume too much on demand or based on subjective rules customized by humans when scheduling water conservancy hubs. It is difficult to dynamically capture the cyclical fluctuations and nonlinear relationships of the discharge volume, and the accuracy is insufficient, which may cause waste of resources or ecological damage. Summary of the invention

[0005] In order to solve the technical problems existing in the prior art that water conservancy hub scheduling often schedules the discharge volume too much on demand or based on subjectively customized rules, making it difficult to dynamically capture the periodic fluctuations and nonlinear relationships of the discharge volume, and the accuracy is insufficient, which may cause waste of resources or ecological damage, the present invention provides a periodic optimization scheduling method and system for water conservancy hubs.

[0006] The technical solution provided by the embodiment of the present invention is as follows:

[0007] First aspect

[0008] An embodiment of the present invention provides a method for periodic optimization and scheduling of a water conservancy hub, comprising:

[0009] S1: Obtain the historical discharge data of the water conservancy hub;

[0010] S2: Determine the data period of historical discharge data by Fourier transform;

[0011] S3: Mapping the historical downflow data within the data period to the linear space through the kernel multidimensional scaling algorithm;

[0012] S4: constructing a discharge prediction function with a noise adjustment term and a deviation adjustment term according to the mapped historical discharge data;

[0013] S5: determining the state transfer matrix, noise adjustment term and bias adjustment term in the discharge amount prediction function by using the maximum likelihood estimation algorithm, and updating the discharge amount prediction function to replicate the volatility of the historical discharge amount data;

[0014] S6: inputting the historical discharge volume data into the updated discharge volume prediction function, and outputting the discharge volume prediction value under the preset time step;

[0015] S7: reversely map the predicted value of the discharge volume to the nonlinear space to obtain the predicted value of the actual discharge volume;

[0016] S8: Optimize the dispatch of water conservancy hubs based on the actual discharge volume prediction value.

[0017] Second aspect

[0018] An embodiment of the present invention provides a periodic optimization scheduling system for a water conservancy hub, comprising:

[0019] processor;

[0020] A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method for periodic optimization scheduling of a water conservancy hub as described in the first aspect is implemented.

[0021] The third aspect

[0022] An embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, the method for periodic optimization scheduling of a water conservancy hub as described in the first aspect is implemented.

[0023] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0024] In an embodiment of the present invention, Fourier transform is used to obtain the data period of the change in the discharge volume from the historical discharge volume data, and then the historical discharge volume data is mapped to the linear space according to the data period, which can effectively reduce the nonlinear complexity, make the data structure clearer, and facilitate model fitting and analysis. A discharge volume prediction function with noise adjustment terms and deviation adjustment terms is established to fit the data in the linear space to accurately capture the data volatility, improve the prediction accuracy, and enhance the robustness and adaptability of the model. After that, the state transfer matrix, noise adjustment terms and deviation adjustment terms in the discharge volume prediction function are determined by the maximum likelihood estimation algorithm, which can accurately reproduce the fluctuation characteristics of the historical discharge volume, and dynamically adjust the model parameters to ensure the sensitivity and adaptability of the prediction function to the changing environment, improve the accuracy and stability of future discharge volume predictions, and optimize the scheduling effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0026] Figure 1 A schematic diagram of a process flow of a periodic optimization scheduling method for a water conservancy hub provided by an embodiment of the present invention;

[0027] Figure 2 A schematic structural diagram of a periodic optimization scheduling system for a water conservancy hub provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0028] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0029] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations or explanations. Any embodiment or design described as "example" in the present invention should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of the word "example" is intended to present the concept in a specific way. In addition, in the embodiments of the present invention, the meaning expressed by "and / or" can be both, or it can be either of the two.

[0030] In order to make the technical problems, technical solutions and advantages to be solved by the present invention more clear, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0031] Reference Manual Attached Figure 1 , showing a flow chart of a method for periodic optimization scheduling of a water conservancy hub provided in an embodiment of the present invention.

[0032] The embodiment of the present invention provides a method for periodic optimization and scheduling of a water conservancy hub, which can be implemented by a periodic optimization and scheduling device for a water conservancy hub, and the periodic optimization and scheduling device for a water conservancy hub can be a terminal or a server. The processing flow of the method for periodic optimization and scheduling of a water conservancy hub can include the following steps:

[0033] S1: Obtain the historical discharge data of the water conservancy hub.

[0034] It should be noted that obtaining historical discharge data of water conservancy hubs provides a basis for subsequent analysis. Historical data can reflect the periodic changes, trends and fluctuation characteristics of the discharge volume, and provide a key reference for building prediction models and optimizing scheduling.

[0035] S2: Determine the data period of historical discharge data through Fourier transform.

[0036] Among them, Fourier transform is a mathematical method used to decompose time domain signals (such as historical discharge data) into sinusoidal wave components of different frequencies, so as to analyze the spectral characteristics of the signal, that is, the energy distribution at different frequencies. The data period refers to the length of time for the repetition of regular changes in the data. In Fourier transform, periodic information can be obtained by the inverse of the main frequency in the spectrum. By extracting the periodic characteristics of historical discharge data through Fourier transform, the law and period of data change can be accurately identified, providing a reliable periodic basis for mapping to linear space and model construction in subsequent steps, thereby improving the accuracy of prediction.

[0037] In a possible implementation, S2 specifically includes:

[0038] S201: Perform Fourier transform on the historical discharge data to obtain the frequency domain characteristics of the historical discharge data:

[0039] ;

[0040] Among them, FFT stands for Fast Fourier Transform, express t Historical discharge data at all times At different frequencies f The energy distribution is The frequency domain signal obtained after FFT transformation.

[0041] S202: Determine the data period in the frequency domain features in combination with the spectrum entropy value of the frequency domain signal:

[0042] ;

[0043] in, Indicates the data cycle, represents the dominant frequency of the frequency domain feature, and Represent the real part of the frequency domain signal and the imaginary part of the frequency domain signal respectively, Indicates that the amplitude of the frequency domain signal is taken Maximum f , log represents the logarithmic function, express f The probability distribution of Indicates about f The spectrum entropy value of .

[0044] Specifically, the process extracts the frequency domain features of the historical discharge data by performing a fast Fourier transform, including the amplitude, main frequency and spectrum entropy value of the frequency domain signal, from which the main period of the data is determined. The period information corresponding to the main frequency can accurately identify the regular changes and repetitive characteristics of the data. The advantage is that the Fourier transform can transform complex time series into a clear spectrum distribution, which is convenient for discovering hidden periodic patterns, and combined with the calculation of the spectrum entropy value, it can evaluate the distribution characteristics of the frequency domain signal, thereby enhancing the accuracy and robustness of data period extraction, and providing a reliable basis for subsequent model construction and optimization.

[0045] S3: Map the historical downflow data within the data period to the linear space through the kernel multidimensional scaling algorithm.

[0046] Among them, the kernel multidimensional scaling algorithm is a dimensionality reduction method that uses kernel functions to calculate the similarity between high-dimensional data points and embeds the data into a low-dimensional (usually linear) space while maintaining the structural characteristics of the original data as much as possible. Compared with traditional multidimensional scaling, it can handle nonlinear data distribution. Linear space is a mathematical structure in which the relationship between data points can be described by linear algebra, such as vector addition and scalar multiplication operations, which facilitates simple and efficient modeling and analysis of data. Mapping historical discharge data to linear space through the kernel multidimensional scaling algorithm can effectively reduce the nonlinear complexity of the data, retain key features, reduce noise interference, provide a concise and clear representation for predictive model construction, and improve modeling and computing efficiency.

[0047] In a possible implementation, S3 specifically includes:

[0048] S301: Calculate the kernel matrix of historical discharge data:

[0049] ;

[0050] in, Indicates the number of historical discharge data i Data points and j Data points The Euclidean distance between represents the bandwidth of the Gaussian kernel that controls the similarity range, K Description and The kernel matrix of the similarity between them, exp represents the natural exponential function.

[0051] In one possible implementation, the Gaussian kernel bandwidth is determined in conjunction with the Fisher information matrix:

[0052] ;

[0053] in, represents partial derivative, Indicates about Fisher information, Indicates that When the maximum value is taken As the Gaussian kernel bandwidth , Indicates about The Gaussian kernel function, It means to find the mathematical expectation.

[0054] The mathematical expectation is specifically used to perform a weighted average operation on the second-order derivative of the Gaussian kernel function. Specifically, it represents the overall weighted mean of the second-order derivative for all possible values ​​of different data points. This operation can comprehensively consider the impact of all data points on the kernel bandwidth, thereby finding an optimal bandwidth that maximizes the Fisher information.

[0055] It should be noted that the essence of Fisher information is to evaluate bandwidth through the distribution of data. Kernel function similarity The Fisher information maximization method can directly capture the discriminability information in the data and is applicable to high-dimensional data. Specifically, the expansion of the Fisher information matrix is:

[0056] ;

[0057] It should be noted that the determination of Gaussian kernel bandwidth through the Fisher information matrix has the following advantages when mapping historical discharge data to linear space: this method can automatically select the optimal bandwidth based on the distribution characteristics of the data, so that the Gaussian kernel function can more accurately capture the similarity and nonlinear structure between data points. This bandwidth optimization avoids the problem of over-smoothing or over-fitting caused by improper bandwidth selection, so that the mapped linear space retains the main features of the historical discharge data and effectively reduces noise and irrelevant information, thereby improving the accuracy and efficiency of subsequent analysis and modeling.

[0058] S302: Centralize the kernel matrix:

[0059] ;

[0060] in, represents the centralized kernel matrix, H represents the centralization matrix, I express The identity matrix, Indicates all 1s matrix, n Indicates the number of samples of historical discharge data.

[0061] S303: Perform eigenvalue decomposition on the centralized kernel matrix:

[0062] ;

[0063] in, D represents a diagonal matrix of eigenvalues ​​in descending order, express D The corresponding eigenvector matrix, the subscript T indicates the transpose.

[0064] S304: retain target eigenvalues ​​greater than the preset eigenvalues, and retain the eigenvectors corresponding to the target eigenvalues ​​as target eigenvectors:

[0065] ;

[0066] in, represents the target eigenvector matrix composed of target eigenvectors, represents the target eigenvalue diagonal matrix composed of target eigenvalues, Indicates historical discharge data x Linear space representation of .

[0067] It should be noted that taking the square root of the target eigenvalue diagonal matrix adjusts the scale of the eigenvector so that the data maintains the relative proportion of the original distribution in the linear space after dimensionality reduction, thereby more accurately reflecting the data characteristics.

[0068] It should be noted that those skilled in the art can set the size of the preset characteristic value according to actual needs, and the present invention is not limited here.

[0069] Specifically, the process describes the similarity between data points by calculating the kernel matrix, eliminates data offsets by centralization, performs eigenvalue decomposition, extracts important features, and reduces the dimension to a linear space. By retaining the target eigenvalues ​​and their eigenvectors, a linear representation of the data is obtained, while the square root adjustment ensures that the proportional characteristics of the original data are maintained. This method can effectively reduce the nonlinear complexity of the data, retain key features and filter noise, provide a simplified and accurate representation for subsequent modeling and analysis, and improve the efficiency and accuracy of the model.

[0070] S4: Construct a discharge prediction function with noise adjustment items and deviation adjustment items according to the mapped historical discharge data.

[0071] Among them, the noise adjustment term is a parameter that modulates the random fluctuations or external interference in the data, and is used to adjust the response of the prediction model to noise. When the noise adjustment term is a unit matrix, the noise amplitude remains unchanged. When it is a zero matrix, the noise is completely ignored. The deviation adjustment term is a parameter that compensates for the systematic deviation between the model prediction and the actual data, aiming to correct the error caused by the periodicity or trend of the data, so that the model prediction is closer to the true value. The discharge prediction function is a mathematical model used to simulate and predict the future discharge of water conservancy hubs. It is based on historical discharge data, combined with periodic characteristics, noise adjustment and deviation compensation parameters to construct a function expression that can reflect the dynamic change law of data. The main goal of the function is to generate a discharge prediction value within a certain time step in the future by calculating the current and historical data. By constructing a discharge prediction function with noise adjustment terms and deviation adjustment terms, the fluctuation characteristics and trend deviations of historical data can be effectively captured, and the robustness of the model to noise and errors can be enhanced, thereby improving the prediction accuracy and providing a reliable basis for subsequent scheduling optimization.

[0072] S5: Determine the state transfer matrix, noise adjustment term and bias adjustment term in the discharge prediction function through the maximum likelihood estimation algorithm, and update the discharge prediction function to replicate the volatility of historical discharge data.

[0073] Among them, the maximum likelihood estimation algorithm is a parameter estimation method that optimizes the model by finding the parameter value that maximizes the probability of the observed data. In probability statistics, this method can maximize the matching of the distribution characteristics of historical data and improve the prediction ability of the model. The state transfer matrix is ​​a relationship matrix that describes the transition of the system from one state to the next in a dynamic system. In the prediction model, it reflects the transmission law of data over time and is an important tool for capturing trends and fluctuations. By optimizing the state transfer matrix, noise adjustment terms, and deviation adjustment terms through the maximum likelihood estimation algorithm, the dynamic fluctuation characteristics of historical discharge data can be accurately replicated, ensuring that the parameter updates of the prediction function are more in line with the changing laws of actual data, thereby significantly improving the reliability and accuracy of future discharge forecasts.

[0074] In a possible implementation manner, the discharge amount prediction function is specifically:

[0075] ;

[0076] in, and Respectively represent the historical discharge volume data after mapping k The amount of discharge at the moment and k +1 moment of leakage, Represents the time step k The noise amplitude adjustment matrix, Represents the description based on the mapped historical discharge data right The state transition matrix of contribution, B express k White noise at all times right The contribution of the noise filter matrix, represents the noise adjustment term, express k The additional deviation at any time is the deviation adjustment item.

[0077] Specifically, the mapped historical discharge data is fitted by the least squares method to obtain the state transfer matrix, namely: ,in, Indicates the minimum A under. It can be understood that the noise amplitude adjustment matrix is ​​to modulate the noise. When is the identity matrix, the noise will follow the historical amplitude. When is a zero matrix, the noise will also be zero. k That is, the length of time between two moments k Time to k +1 The length of time between moments.

[0078] It should be noted that the discharge prediction function established by introducing the state transfer matrix, noise adjustment term and deviation adjustment term can accurately capture the dynamic fluctuation characteristics of historical discharge data, while effectively regulating noise and systematic deviations, enhancing the robustness and adaptability of the model, and improving the accuracy and stability of the prediction.

[0079] In a possible implementation, S5 specifically includes:

[0080] S501: Initialize the state transfer matrix, the noise adjustment term and the deviation adjustment term to initialize the discharge amount prediction function.

[0081] S502: Inputting the mapped historical discharge volume data into the initialized discharge volume prediction function according to the data period to perform separate predictions to obtain multiple discharge volumes at each time step.

[0082] S503: Determine the deviation adjustment item according to the predicted discharge volume at each time step:

[0083] ;

[0084] in, and Respectively represent the predicted k Moment and k The average discharge volume at time +1.

[0085] S504: Calculate the covariance matrix of each time step:

[0086] ;

[0087] in, B represents the white noise filter matrix, Describes the time step k The auxiliary matrix of the covariance information passed recursively, represents the initial covariance matrix, represents the prediction bias vector for the first time step, It means taking the average value, Indicates k + the covariance matrix of 1 time step, Represents the time step k- 1's noise amplitude adjustment matrix, Indicates k +1 time step forecast bias vector.

[0088] Among them, time step k refers to the time interval between two adjacent moments in the prediction model. It is a discrete time span used to describe the dynamic change process of time, such as the distance from moment k to moment k+1. Moment k refers to a specific time point, which represents the state of the prediction model at a specific moment, such as the data state at t=k. Time step k is the span between two moments on the time axis. The prediction model recursively transfers the state within time step k and gradually calculates the result at moment k+1. The white noise filter matrix is ​​a matrix used to describe the influence characteristics of random white noise in the prediction model. White noise is essentially a random signal with zero mean and fixed variance, characterized by uniform energy distribution of all frequencies. The role of the filter matrix is ​​to regulate the influence of white noise on the system state transmission, ensuring that the prediction model can effectively filter the interference of noise within a certain range, while retaining the random information that may be contained in the white noise.

[0089] Specifically, ,in, represents the sample value predicted at the first time step, represents the sample mean of the first time step obtained from multiple predictions, ,in, It means that the discharge prediction function is k +1 sample value predicted at time step, It means that the discharge prediction function is k +1 time step forecast sample mean. , so based on The calculation method of is known, .

[0090] S505: Determine the accuracy matrix describing the prediction fit of the discharge amount prediction function based on the covariance matrix:

[0091] ;

[0092] in, Describing different moments k The precision matrix of the fit, represents the inverse of the covariance matrix, represents the inverse of the noise amplitude adjustment matrix, C represents the white noise covariance matrix, Represents the inverse transpose of the noise amplitude adjustment matrix.

[0093] Among them, the white noise covariance matrix is ​​a matrix used to describe the characteristics of white noise and the degree of correlation between different time steps or state variables. It contains the variance information of white noise (indicating the intensity of noise) and the covariance information between noises (indicating the correlation of noise). This matrix can quantify the interference degree and propagation mode of white noise on the prediction model, and help adjust the robustness of the model to make it closer to the volatility and randomness of actual data. The white noise covariance matrix describes the essential characteristics of noise. Combining the system's background noise and dynamic adjustment characteristics, the prediction model is more in line with the volatility and discreteness of actual data.

[0094] S506: Taking maximizing the prediction fit as the goal, that is, maximizing the prediction likelihood, construct an objective function for the noise amplitude adjustment matrix:

[0095] ;

[0096] Among them, min means taking the minimum value, ln means logarithmic function, Indicates j The deviation vector of the prediction, the subscript T indicates the transposition, J Represents the total number of predictions.

[0097] S507: Solve the objective function, and substitute the solved noise amplitude adjustment matrix into the noise adjustment item to update the noise adjustment item.

[0098] Specifically, the objective function optimizes the noise amplitude adjustment matrix by maximizing the prediction likelihood. The calculation involves the covariance matrix, bias vector, and precision matrix. Common methods include optimization algorithms such as gradient descent and Newton's method, which iteratively calculate the objective function until it converges to the optimal value. For example, Python, MATLAB, Gurobi / CPLEX, etc.

[0099] It should be noted that by dynamically optimizing the noise amplitude adjustment matrix, this method can balance the noise and bias characteristics in the data while maximizing the model fit, so that the prediction model can gradually improve its accuracy in multiple iterations. Especially in the case of large noise interference, it can significantly enhance the stability and robustness of the model and provide more reliable prediction results for practical applications.

[0100] S6: Input the historical discharge volume data into the updated discharge volume prediction function, and output the discharge volume prediction value under the preset time step.

[0101] Among them, the time step refers to the time interval between two consecutive prediction moments in the prediction model. For example, the time step can be hours, days, weeks or other time units, which is determined by actual needs. In the prediction, the time step affects the time accuracy and range of the prediction. By inputting historical discharge data into the updated prediction function and using the preset time step to generate future discharge forecast values, accurate water volume prediction for multiple moments in the future can be achieved to meet the needs of scheduling optimization at different time scales.

[0102] It should be noted that those skilled in the art can set the size of the preset time step according to actual needs, and the present invention is not limited here.

[0103] S7: Reversely map the predicted value of the discharge volume to the nonlinear space to obtain the predicted value of the actual discharge volume.

[0104] It should be noted that by reverse mapping the discharge forecast value from linear space to nonlinear space, the original complex structure and nonlinear characteristics of the data can be restored, making the forecast value more consistent with the actual situation. This process effectively retains the real volatility and nonlinear laws in the historical discharge volume, improves the reliability and practicality of the forecast results, and provides an accurate basis for subsequent scheduling.

[0105] In a possible implementation manner, the actual leakage amount prediction value is specifically:

[0106] ;

[0107] in, and Respectively k +1 time point predicted value of discharge and actual predicted value of discharge, Represents the inverse mapping of the kernel multidimensional scaling algorithm.

[0108] Specifically, the kernel multidimensional scaling algorithm is used to reversely map the predicted value of the discharge to a nonlinear space to obtain the actual predicted value of the discharge.

[0109] It should be noted that, through the inverse mapping of the kernel multidimensional scaling algorithm, the discharge prediction value in the linear space is converted back to the nonlinear space to restore the original complex structure and characteristics of the data, and obtain a more realistic discharge prediction value. This method can retain the nonlinear laws in the historical data, improve the reliability and practicality of the prediction results, and provide a more accurate basis for the dispatch of water conservancy hubs.

[0110] S8: Optimize the dispatch of water conservancy hubs based on the actual discharge volume prediction value.

[0111] In a possible implementation manner, S8 specifically includes:

[0112] The discharge volume of the water conservancy hub is set according to the actual discharge volume prediction value to complete the optimal scheduling of the water conservancy hub.

[0113] Specifically, the process obtains the historical discharge data of the water conservancy hub, extracts the periodic characteristics by Fourier transform, maps the data to linear space, constructs a discharge prediction function containing noise and deviation adjustment terms, and then optimizes the prediction function parameters by maximum likelihood estimation to predict the discharge values ​​of multiple time steps in the future. After that, the real discharge characteristics are restored through reverse mapping and optimized scheduling is performed. This method can accurately capture the periodic and nonlinear characteristics of the data, reduce noise interference, and improve prediction accuracy, thereby achieving more scientific and efficient water conservancy hub scheduling management.

[0114] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0115] In an embodiment of the present invention, Fourier transform is used to obtain the data period of the change in the discharge volume from the historical discharge volume data, and then the historical discharge volume data is mapped to the linear space according to the data period, which can effectively reduce the nonlinear complexity, make the data structure clearer, and facilitate model fitting and analysis. A discharge volume prediction function with noise adjustment terms and deviation adjustment terms is established to fit the data in the linear space to accurately capture the data volatility, improve the prediction accuracy, and enhance the robustness and adaptability of the model. After that, the state transfer matrix, noise adjustment terms and deviation adjustment terms in the discharge volume prediction function are determined by the maximum likelihood estimation algorithm, which can accurately reproduce the fluctuation characteristics of the historical discharge volume, and dynamically adjust the model parameters to ensure the sensitivity and adaptability of the prediction function to the changing environment, improve the accuracy and stability of future discharge volume predictions, and optimize the scheduling effect.

[0116] Reference Manual Attached Figure 2 , showing a structural schematic diagram of a periodic optimization scheduling system for a water conservancy hub provided by the present invention.

[0117] The present invention further provides a water conservancy hub periodic optimization scheduling system 20, which is applied to the above-mentioned water conservancy hub periodic optimization scheduling method, comprising:

[0118] Processor 201.

[0119] The memory 202 stores computer-readable instructions, and when the computer-readable instructions are executed by the processor 201, the periodic optimization scheduling method of the water conservancy hub in the method embodiment is implemented.

[0120] The water conservancy hub periodic optimization scheduling system 20 provided by the present invention can execute the above-mentioned water conservancy hub periodic optimization scheduling method and achieve the same or similar technical effects. To avoid repetition, the present invention will not go into details.

[0121] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0122] In an embodiment of the present invention, Fourier transform is used to obtain the data period of the change in the discharge volume from the historical discharge volume data, and then the historical discharge volume data is mapped to the linear space according to the data period, which can effectively reduce the nonlinear complexity, make the data structure clearer, and facilitate model fitting and analysis. A discharge volume prediction function with noise adjustment terms and deviation adjustment terms is established to fit the data in the linear space to accurately capture the data volatility, improve the prediction accuracy, and enhance the robustness and adaptability of the model. After that, the state transfer matrix, noise adjustment terms and deviation adjustment terms in the discharge volume prediction function are determined by the maximum likelihood estimation algorithm, which can accurately reproduce the fluctuation characteristics of the historical discharge volume, and dynamically adjust the model parameters to ensure the sensitivity and adaptability of the prediction function to the changing environment, improve the accuracy and stability of future discharge volume predictions, and optimize the scheduling effect.

[0123] It should be understood that the processor in the embodiment of the present invention may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0124] It should also be understood that the memory in the embodiments of the present invention may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0125] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0126] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0127] In the present invention, "at least one" means one or more, and "plurality" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can mean: a, b, c, ab, ac, bc, or abc, where a, b, and c can be single or multiple.

[0128] It should be understood that in various embodiments of the present invention, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0129] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0130] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described equipment, devices and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0131] In the several embodiments provided by the present invention, it should be understood that the disclosed devices, apparatuses and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0132] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0133] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0134] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0135] An embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon. When the program is executed by a processor, the method for periodic optimization scheduling of a water conservancy hub as described in the method embodiment is implemented.

[0136] A computer-readable storage medium provided by the present invention can implement the steps and effects of the periodic optimization scheduling method of the water conservancy hub in the above method embodiment. To avoid repetition, the present invention will not go into details.

[0137] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0138] In an embodiment of the present invention, Fourier transform is used to obtain the data period of the change in the discharge volume from the historical discharge volume data, and then the historical discharge volume data is mapped to the linear space according to the data period, which can effectively reduce the nonlinear complexity, make the data structure clearer, and facilitate model fitting and analysis. A discharge volume prediction function with noise adjustment terms and deviation adjustment terms is established to fit the data in the linear space to accurately capture the data volatility, improve the prediction accuracy, and enhance the robustness and adaptability of the model. After that, the state transfer matrix, noise adjustment terms and deviation adjustment terms in the discharge volume prediction function are determined by the maximum likelihood estimation algorithm, which can accurately reproduce the fluctuation characteristics of the historical discharge volume, and dynamically adjust the model parameters to ensure the sensitivity and adaptability of the prediction function to the changing environment, improve the accuracy and stability of future discharge volume predictions, and optimize the scheduling effect.

[0139] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0140] There are a few points to note:

[0141] (1) The drawings of the embodiments of the present invention only involve structures related to the embodiments of the present invention. Other structures may refer to conventional designs.

[0142] (2) For the sake of clarity, in the drawings used to describe the embodiments of the present invention, the thickness of layers or regions is exaggerated or reduced, that is, these drawings are not drawn according to the actual scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being "on" or "under" another element, the element may be "directly" "on" or "under" the other element or there may be intermediate elements.

[0143] (3) In the absence of conflict, the embodiments of the present invention and the features therein may be combined with each other to obtain new embodiments.

[0144] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. The protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A method for periodic optimization and dispatching of a water conservancy hub, characterized in that: include: S1: Obtaining historical discharge data of the water conservancy hub; S2: Determine the data period of the historical discharge volume data by Fourier transform; S3: Mapping the historical leakage data in the data period to a linear space by a kernel multidimensional scaling algorithm; S4: constructing a discharge prediction function with a noise adjustment term and a deviation adjustment term according to the mapped historical discharge data; S5: determining a state transfer matrix, a noise adjustment term and a deviation adjustment term in the discharge amount prediction function by a maximum likelihood estimation algorithm, and updating the discharge amount prediction function to replicate the volatility of the historical discharge amount data; S6: inputting the historical discharge volume data into the updated discharge volume prediction function, and outputting the discharge volume prediction value under the preset time step; S7: Reversely mapping the predicted value of the discharge volume to a nonlinear space to obtain a true predicted value of the discharge volume; S8: Optimizing the dispatching of the water conservancy hub according to the predicted value of the actual discharge volume; The discharge volume prediction function is specifically: ； in, and Respectively represent the historical discharge volume data after mapping k The amount of discharge at the moment and k +1 moment of leakage, Represents the time step k The noise amplitude adjustment matrix, Represents the description based on the mapped historical discharge data right The state transition matrix of contribution, B express k White noise at all times right The contribution of the noise filter matrix, represents the noise adjustment term, express k The additional deviation at any time is the deviation adjustment item.

2. The method for periodic optimization and dispatching of a water conservancy hub according to claim 1 is characterized in that: The S2 specifically includes: S201: Perform Fourier transform on the historical discharge data to obtain frequency domain characteristics of the historical discharge data: ； Among them, FFT stands for Fast Fourier Transform, express t Historical discharge data at all times At different frequencies f The energy distribution is The frequency domain signal obtained after FFT transformation; S202: Determine the data period in the frequency domain features in combination with the spectrum entropy value of the frequency domain signal: ； in, Indicates the data cycle, represents the dominant frequency of the frequency domain feature, and Represent the real part of the frequency domain signal and the imaginary part of the frequency domain signal respectively, Indicates that the amplitude of the frequency domain signal is taken Maximum f , log represents the logarithmic function, express f The probability distribution of Indicates about f The spectrum entropy value of .

3. The method for periodic optimization and dispatching of a water conservancy hub according to claim 1 is characterized in that: The S3 specifically includes: S301: Calculate the kernel matrix of the historical discharge data: ； in, Indicates the number of historical discharge data i Data points and j Data points The Euclidean distance between represents the bandwidth of the Gaussian kernel that controls the similarity range, K Description and The kernel matrix of similarity between them, exp represents the natural exponential function; S302: Centralize the core matrix: ； in, represents the centralized kernel matrix, H represents the centralization matrix, I express The identity matrix, Indicates all 1s matrix, n Indicates the number of samples of historical discharge data; S303: Perform eigenvalue decomposition on the centralized kernel matrix: ； in, D represents a diagonal matrix of eigenvalues ​​in descending order, express D The corresponding eigenvector matrix, the subscript T indicates transpose; S304: retain target eigenvalues ​​greater than the preset eigenvalues, and retain the eigenvectors corresponding to the target eigenvalues ​​as target eigenvectors: ； in, represents the target eigenvector matrix composed of target eigenvectors, represents the target eigenvalue diagonal matrix composed of target eigenvalues, Indicates historical discharge data x Linear space representation of .

4. The method for periodic optimization and dispatching of a water conservancy hub according to claim 3 is characterized in that: The Gaussian kernel bandwidth is determined in combination with the Fisher information matrix: ； in, represents partial derivative, Indicates about Fisher information, Indicates that When the maximum value is taken As the Gaussian kernel bandwidth , Indicates about The Gaussian kernel function, It means to find the mathematical expectation.

5. The method for periodic optimization and dispatching of a water conservancy hub according to claim 1, characterized in that: The S5 specifically includes: S501: Initializing the state transfer matrix, the noise adjustment item, and the deviation adjustment item to initialize the discharge amount prediction function; S502: inputting the mapped historical discharge volume data into the initialized discharge volume prediction function according to the data period to perform predictions respectively, and obtaining multiple discharge volumes at each time step; S503: Determine the deviation adjustment item according to the predicted discharge amount at each time step: ； in, and Respectively represent the predicted k Moment and k The mean discharge volume at time +1; S504: Calculate the covariance matrix of each time step: ； in, B represents the white noise filter matrix, Describes the time step k The auxiliary matrix of the covariance information passed recursively, represents the initial covariance matrix, represents the prediction bias vector for the first time step, It means taking the average value, Indicates k + the covariance matrix of 1 time step, Represents the time step k- 1's noise amplitude adjustment matrix, Indicates k +1 time step forecast bias vector; S505: Determine the accuracy matrix describing the prediction fit of the discharge amount prediction function based on the covariance matrix: ； in, Describing different moments k The precision matrix of the fit, represents the inverse of the covariance matrix, represents the inverse of the noise amplitude adjustment matrix, C represents the white noise covariance matrix, represents the inverse transpose of the noise amplitude adjustment matrix; S506: Taking maximizing the prediction fit as the goal, that is, maximizing the prediction likelihood, constructing an objective function for the noise amplitude adjustment matrix: ； Among them, min means taking the minimum value, ln means logarithmic function, Indicates j The deviation vector of the prediction, the subscript T indicates the transposition, J Indicates the total number of predictions; S507: Solve the objective function, and substitute the noise amplitude adjustment matrix obtained by the solution into the noise adjustment item to update the noise adjustment item.

6. The method for periodic optimization and dispatching of a water conservancy hub according to claim 1, characterized in that: The actual discharge volume prediction value is specifically: ； in, and Respectively k +1 time point predicted value of discharge and actual predicted value of discharge, Represents the inverse mapping of the kernel multidimensional scaling algorithm.

7. The method for periodic optimization and dispatching of a water conservancy hub according to claim 1, characterized in that: The S8 is specifically: The discharge volume of the water conservancy hub is set according to the actual discharge volume prediction value to complete the optimal scheduling of the water conservancy hub.

8. A periodic optimization dispatching system for a water conservancy hub, characterized in that: include: processor; A memory having computer-readable instructions stored thereon, wherein when the computer-readable instructions are executed by the processor, the method for periodic optimization scheduling of a water conservancy hub as claimed in any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the periodic optimization scheduling method for a water conservancy hub as described in any one of claims 1 to 7 is implemented.

Images