Coordinated scheduling method and system for cross-voltage-class load prediction, medium and processor

Abstract

The invention relates to the technical field of bus load prediction, and discloses a coordinated scheduling method and system for cross-voltage-level load prediction, a medium and a processor, and the method comprises the steps: obtaining historical load data and topological structure data of each voltage-level bus in a power system; correlation analysis is carried out on the historical load data of the buses of all the voltage grades to obtain a cross-voltage-grade load transfer rule; constructing a cross-voltage-level load transfer mathematical model; inputting the historical load data of each voltage level bus into a deep learning LSTM model for training and learning to obtain a trained bus load prediction model; performing bus load prediction of each voltage level by using the trained bus load prediction model; and in combination with the prediction load curve and the cross-voltage-class load transfer mathematical model, obtaining a cross-voltage-class load scheduling strategy. According to the invention, through integration of bus load information of different voltage levels, cross-voltage-level coordinated prediction and optimal scheduling are realized.

Description

Technical Field

[0001] The present invention relates to the technical field of bus load prediction, and in particular to a coordinated scheduling method, system, medium and processor for cross-voltage level load prediction. Background Art

[0002] Short-term bus load forecasting plays a key role in maintaining grid frequency stability, preventing overload, and ensuring power supply reliability. Accurate bus load forecasting is important for optimizing energy structure and promoting clean energy consumption.

[0003] Existing methods often predict bus loads at each voltage level independently, ignoring the inherent correlation and mutual influence between different voltage levels, such as flow transfer, voltage control, load transfer, etc. As the scale of power grids expands and the complexity of their structures increases, the interconnection and interaction of regional multi-voltage level power grids increases, and the limitations of single voltage level prediction become increasingly obvious. Therefore, it is necessary to consider the synergy effect across voltage levels to achieve global optimization.

[0004] In view of this, a coordinated scheduling method, system, medium and processor for cross-voltage level load forecasting are needed. Summary of the invention

[0005] In view of the problem that the single voltage level prediction in the prior art cannot meet the existing complex power grid scale, the present invention provides a coordinated scheduling method, system, medium and processor for cross-voltage level load prediction, which can effectively integrate bus load information of different voltage levels and realize cross-voltage level coordinated prediction and optimized scheduling. The specific technical solution is as follows:

[0006] A coordinated dispatching method for load forecasting across voltage levels, comprising:

[0007] Obtain historical load data and topological structure data of buses of various voltage levels in the power system;

[0008] Conduct correlation analysis on the historical load data of buses at each voltage level to obtain the load transfer rules across voltage levels;

[0009] According to the load transfer law across voltage levels and topological structure data, a mathematical model of load transfer across voltage levels is constructed;

[0010] The historical load data of buses of each voltage level is input into the deep learning LSTM model for training and learning, and a bus load prediction model that has been trained to predict each voltage level is obtained;

[0011] Use the trained bus load prediction model to predict the bus load of each voltage level and obtain the predicted load curve of each voltage level bus;

[0012] Combining the predicted load curves of buses at each voltage level and the mathematical model of load transfer across voltage levels, the load dispatch strategy across voltage levels is obtained.

[0013] Preferably, a coordinated scheduling method for cross-voltage level load forecasting also includes:

[0014] The historical load data of buses at each voltage level are cleaned and normalized.

[0015] Preferably, the correlation analysis of the historical load data of buses of each voltage level to obtain the cross-voltage level load transfer rule includes:

[0016] By analyzing the historical load data of buses of different voltage levels in time series, the time-varying patterns of the loads of buses of different voltage levels are obtained.

[0017] By performing cross-spectral analysis on the historical load data of buses at different voltage levels, the resonance or synchronization of loads at different voltage levels at different frequencies can be obtained;

[0018] By conducting causal relationship analysis on the historical load data of buses of different voltage levels, the causal relationship between the loads of buses of different voltage levels is obtained;

[0019] According to the variation law of bus loads of each voltage level over time, the resonance or synchronization of loads of different voltage levels at different frequencies and the causal relationship between bus loads of different voltage levels, the law of load transfer across voltage levels is obtained.

[0020] Preferably, constructing a mathematical model of load transfer across voltage levels according to the load transfer law across voltage levels and topological structure data includes:

[0021] Determine variables and parameters according to the load transfer law across voltage levels, and establish a power balance equation based on the variables and parameters;

[0022] Set the constraints of the model according to the load transfer rules across voltage levels and topological structure data;

[0023] An objective function of the total system loss is constructed, and a mathematical model of load transfer across voltage levels is obtained by combining the power balance equation, constraints and objective function.

[0024] Preferably, the historical load data of buses of each voltage level is input into the deep learning LSTM model for training and learning, and the trained bus load prediction model capable of predicting each voltage level is obtained, which includes:

[0025] According to the load data characteristics of buses of each voltage level, characteristic data of load impact on buses of each voltage level are extracted;

[0026] Divide the influencing characteristic data of buses of different voltage levels into training sets and test sets corresponding to different levels;

[0027] The training sets and test sets corresponding to different levels are input into the deep learning LSTM model for training and learning, and a bus load prediction model that has been trained to predict each voltage level is obtained.

[0028] Preferably, the method of using the trained bus load prediction model to predict the bus loads of each voltage level to obtain the predicted load curves of the buses of each voltage level includes:

[0029] Input the currently acquired load impact characteristic data into the trained bus load prediction model, and output the load prediction values ​​corresponding to a series of time points of the buses of each voltage level;

[0030] Process the load forecast values ​​of buses at each voltage level to remove abnormal values ​​and interpolate missing values;

[0031] According to the processed load forecast value, the time axis and load axis are set to construct the forecast load curve of each voltage level bus.

[0032] Preferably, the cross-voltage level load dispatching strategy obtained by combining the predicted load curves of buses of each voltage level and the cross-voltage level load transfer mathematical model includes:

[0033] According to the predicted load curve of each voltage level bus, the load prediction value of each voltage level bus at the same time point is obtained;

[0034] Substitute the load forecast value of each voltage level bus at the same time point as the boundary condition into the cross-voltage level load transfer mathematical model;

[0035] By solving the mathematical model of load transfer across voltage levels, the load transfer path and transfer amount between different voltage levels are obtained;

[0036] According to the load transfer path and transfer amount, the load dispatching strategy across voltage levels is obtained.

[0037] A coordinated dispatching system for cross-voltage level load forecasting, applied to the above-mentioned coordinated dispatching method for cross-voltage level load forecasting, comprises:

[0038] A data acquisition unit, used to acquire historical load data and topological structure data of buses of various voltage levels in the power system;

[0039] The correlation analysis unit is used to perform correlation analysis on the historical load data of buses of each voltage level to obtain the load transfer rules across voltage levels;

[0040] A load transfer model building unit, used to build a mathematical model of load transfer across voltage levels according to the load transfer law across voltage levels and topological structure data;

[0041] The load forecasting model training unit is used to input the historical load data of buses of various voltage levels into the deep learning LSTM model for training and learning, so as to obtain a trained bus load forecasting model capable of predicting various voltage levels;

[0042] A load prediction unit is used to predict the bus loads of each voltage level by using the trained bus load prediction model to obtain a predicted load curve of each voltage level bus;

[0043] The dispatching strategy generating unit is used to combine the predicted load curves of buses of each voltage level and the mathematical model of load transfer across voltage levels to obtain the load dispatching strategy across voltage levels.

[0044] A computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the aforementioned coordinated scheduling method for load forecasting across voltage levels.

[0045] A processor is used to run a program, wherein the program, when running, executes the aforementioned coordinated scheduling method for load forecasting across voltage levels.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] A coordinated dispatching method for cross-voltage level load prediction of the present invention obtains historical load data and topological structure data of buses of each voltage level in the power system; performs correlation analysis on the historical load data of buses of each voltage level to obtain the law of cross-voltage level load transfer; constructs a mathematical model of cross-voltage level load transfer according to the law of cross-voltage level load transfer and topological structure data; inputs the historical load data of buses of each voltage level into a deep learning LSTM model for training and learning to obtain a trained bus load prediction model capable of predicting each voltage level; uses the trained bus load prediction model to predict the bus load of each voltage level to obtain the predicted load curve of buses of each voltage level; combines the predicted load curve of buses of each voltage level and the mathematical model of cross-voltage level load transfer to obtain a load dispatching strategy across voltage levels. The present invention realizes coordinated prediction and optimized dispatch across voltage levels by integrating bus load information of different voltage levels, thereby enhancing the flexibility and adaptability of the overall operation of the power grid. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the description of the prior art. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0049] Figure 1 This is a flow chart of a coordinated scheduling method for cross-voltage level load forecasting according to the present invention.

[0050] Figure 2 The present invention is a flow chart of another embodiment of a coordinated scheduling method for load forecasting across voltage levels according to the present invention.

[0051] Figure 3 This is a schematic diagram of a coordinated dispatching system for load forecasting across voltage levels according to the present invention. DETAILED DESCRIPTION

[0052] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0053] It should be understood that when used in this specification and the appended claims, the terms "include" and "comprises" indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or combinations thereof.

[0054] It should also be understood that the terms used in the present specification are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the present specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0055] It should be further understood that the term "and / or" used in the present description and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0056] See the following examples Figures 1 to 3 .

[0057] The embodiment of the present application provides a coordinated scheduling method for load forecasting across voltage levels, including:

[0058] Step S01, obtaining historical load data and topological structure data of buses of various voltage levels in the power system;

[0059] The historical load data includes meteorological data that affect the load (including but not limited to temperature, humidity, light intensity, wind speed, etc.) and user electricity consumption behavior data. These data reflect the impact of the external environment and user habits on the power load. The topological structure data contains the physical connection information of the power grid, such as the node (bus) number used to uniquely identify each connection point in the power grid; the line connection relationship explains how the buses are connected through the line; the line impedance parameters affect the transmission characteristics of the power on the line; the transformer ratio determines the conversion relationship between different voltage levels; the switch status reflects the current connection configuration of the power grid.

[0060] The historical load data of buses of each voltage level are cleaned and normalized. The purpose of data processing is to remove outliers and erroneous data in the data. For example, there may be unreasonable load readings caused by sensor failures, and these interferences can be eliminated through data cleaning. Normalization is to convert data of different magnitudes into data of the same scale, usually mapping the data to a specific interval. It helps to improve the efficiency and accuracy of subsequent data analysis and model training.

[0061] Step S02: Perform correlation analysis on the historical load data of buses of each voltage level to obtain the load transfer rules across voltage levels;

[0062] Specifically, the correlation analysis of the historical load data of buses of each voltage level to obtain the load transfer rules across voltage levels includes:

[0063] By analyzing the historical load data of buses of different voltage levels in time series, the time-varying patterns of the loads of buses of different voltage levels are obtained.

[0064] By using time series models such as ARIMA, seasonal ARIMA, and state space models, the changing patterns of bus loads at various voltage levels over time, especially the periodicity and trend, can be analyzed. By fitting and predicting historical data, the changing characteristics of loads on different time scales can be quantitatively described.

[0065] By performing cross-spectral analysis on the historical load data of buses at different voltage levels, the resonance or synchronization of loads at different voltage levels at different frequencies can be obtained;

[0066] By using methods such as Fourier transform, wavelet analysis or empirical mode decomposition (EMD), the frequency domain characteristics of load data are extracted to reveal the resonance or synchronization of loads of different voltage levels at different frequencies, which indirectly reflects the load transfer situation. For example, if at a certain frequency, two bus loads of different voltage levels resonate, it means that there is a load transfer phenomenon on the time scale corresponding to the frequency.

[0067] By conducting causal relationship analysis on the historical load data of buses of different voltage levels, the causal relationship between the loads of buses of different voltage levels is obtained;

[0068] Using Granger causality test, transfer function model or structural vector autoregression (SVAR) model, the causal relationship between bus loads of different voltage levels is explored to clarify the direction and intensity of load transfer. For example, Granger causality test can determine whether the load change of a bus of one voltage level will cause the load change of another bus of another voltage level, and the lag time of such influence.

[0069] According to the variation law of bus loads of each voltage level over time, the resonance or synchronization of loads of different voltage levels at different frequencies and the causal relationship between bus loads of different voltage levels, the law of load transfer across voltage levels is obtained.

[0070] For example, if the high-voltage bus load peaks in the summer afternoon, and causal analysis finds that the peak will cause the low-voltage bus load to rise in the following hours, and cross-spectrum analysis shows that there is resonance between the two at the daily cycle frequency, it can be concluded that there is a specific load transfer pattern between the two voltage levels in the summer afternoon.

[0071] Step S03: constructing a mathematical model of load transfer across voltage levels according to the load transfer law across voltage levels and topological structure data;

[0072] In the model, the load transfer law is expressed by variables and equations. For example, the load transfer amount is set as a variable, and the relationship equation between the variables is established according to the direction and intensity of the load transfer. Parameters such as line impedance and transformer ratio in the topology data are used to construct the constraints in the model to ensure that the load transfer is carried out within the range allowed by the physical structure of the power grid. For example, according to the transmission capacity limit of the line, the upper limit constraint of the line transmission power is set in the model.

[0073] Step S04: input the historical load data of buses of each voltage level into the deep learning LSTM model for training and learning, so as to obtain a bus load prediction model that is trained to predict each voltage level;

[0074] Specifically, the historical load data of buses of each voltage level is input into the deep learning LSTM model for training and learning, and the trained bus load prediction model capable of predicting each voltage level is obtained, including:

[0075] According to the load data characteristics of buses of each voltage level, characteristic data of load impact on buses of each voltage level are extracted;

[0076] Divide the influencing characteristic data of buses of different voltage levels into training sets and test sets corresponding to different levels;

[0077] For example, the load influence characteristic data of each voltage level bus in the first historical time period is obtained as a training set; the load influence characteristic data of each voltage level bus in the second historical time period is obtained as a test set; the second historical time period is located after the first historical time period;

[0078] The training sets and test sets corresponding to different levels are input into the deep learning LSTM model for training and learning, and a bus load prediction model that has been trained to predict each voltage level is obtained.

[0079] Through the deep learning LSTM model, load forecasting is performed based on the load impact characteristic data of the first historical time period, and the predicted load results in the second historical time period are obtained; according to the difference between the historical predicted load results and the load impact characteristic data of the second historical time period, the deep learning LSTM model is trained to obtain a bus load forecasting model that has been trained and can predict various voltage levels.

[0080] Step S05: Use the trained bus load prediction model to predict the bus loads of each voltage level, and obtain the predicted load curves of the buses of each voltage level;

[0081] Specifically, the method of using the trained bus load prediction model to predict the bus load of each voltage level to obtain the predicted load curve of each voltage level bus includes:

[0082] Input the currently acquired load impact characteristic data into the trained bus load prediction model, and output the load prediction values ​​corresponding to a series of time points of the buses of each voltage level;

[0083] Process the load forecast values ​​of buses at each voltage level to remove abnormal values ​​and interpolate missing values;

[0084] According to the processed load forecast value, the time axis and load axis are set to construct the forecast load curve of each voltage level bus.

[0085] Step S06: Combine the predicted load curves of the buses at each voltage level and the mathematical model of load transfer across voltage levels to obtain a load dispatching strategy across voltage levels.

[0086] Specifically, the load dispatching strategy across voltage levels obtained by combining the predicted load curves of buses of each voltage level and the mathematical model of load transfer across voltage levels includes:

[0087] According to the predicted load curve of each voltage level bus, the load prediction value of each voltage level bus at the same time point is obtained;

[0088] Substitute the load forecast value of each voltage level bus at the same time point as the boundary condition into the cross-voltage level load transfer mathematical model;

[0089] By solving the mathematical model of load transfer across voltage levels, the load transfer path and transfer amount between different voltage levels are obtained.

[0090] According to the load transfer path and transfer amount, the load dispatching strategy across voltage levels is obtained.

[0091] In this embodiment, a load dispatching strategy across voltage levels is formulated based on the load transfer path and transfer amount obtained by the solution, combined with the actual operation and dispatching requirements of the power system. The strategy should include specific measures, time nodes, priorities, etc. for load transfer, ensuring the stable operation and supply and demand balance of the power system.

[0092] In this embodiment, when using the Long Short-Term Memory (LSTM) network for bus load forecasting, the main parameter adjustment strategies are as follows:

[0093] (1) Network architecture parameters:

[0094] Number of hidden layers and units: LSTM networks usually contain one or more hidden layers. Try different numbers of hidden layers (such as 1, 2 or more layers) and the number of LSTM units per layer (such as 64, 128, 256, etc.) to find the best network depth and width. Increasing the number of layers or units can capture more complex patterns.

[0095] Regression output layer: If the target is a continuous bus load value, a fully connected layer with an activation function (such as linear activation) is usually used as the output layer. Make sure the number of neurons in the output layer matches the number of load variables to be predicted.

[0096] (2) Time series related parameters:

[0097] Sequence length (window size): determines the number of historical data points that the model processes. Choose an appropriate sequence length to capture the periodicity and trend of load changes.

[0098] Step size (time interval): defines the time interval at which the model makes future load forecasts. For example, you can forecast hourly, daily, or weekly loads.

[0099] (3) Training parameters:

[0100] Learning rate: adjusts the speed at which the model updates weights. Too high a learning rate may cause training to not converge, while too low a learning rate may cause training to be slow and easy to fall into local optimality. Use strategies to dynamically adjust the learning rate, such as learning rate decay, exponential decay, cosine annealing, etc., or use adaptive learning rate methods (such as Adam, Adagrad, etc.).

[0101] Batch size: determines the number of samples used in each forward propagation and backpropagation calculation. Larger batches can improve computational efficiency.

[0102] Iterations / epochs: Set the number of times the model goes through the entire training set. Properly increasing the number of epochs can help the model learn the data better.

[0103] (4) Regularization and Dropout:

[0104] L2 regularization (weight decay): Prevent overfitting by adding the L2 norm penalty term of the weight matrix to the loss function.

[0105] Dropout: Randomly discard some neurons (and their connections) during training to reduce the dependency between neurons and prevent overfitting. Usually a certain Dropout ratio (such as 0.2, 0.5, etc.) is set in the hidden layer.

[0106] (5)Optimizer:

[0107] Choose a suitable optimizer, such as RMSprop, Adam, Adadelta, etc.

[0108] (6) Hyperparameter tuning:

[0109] Methods such as grid search, random search, and Bayesian optimization are used to systematically traverse the parameter space to find the best combination.

[0110] A coordinated dispatching method for cross-voltage level load prediction of the present invention obtains historical load data and topological structure data of buses of each voltage level in the power system; performs correlation analysis on the historical load data of buses of each voltage level to obtain the law of cross-voltage level load transfer; constructs a mathematical model of cross-voltage level load transfer according to the law of cross-voltage level load transfer and topological structure data; inputs the historical load data of buses of each voltage level into a deep learning LSTM model for training and learning to obtain a trained bus load prediction model capable of predicting each voltage level; uses the trained bus load prediction model to predict the bus load of each voltage level to obtain the predicted load curve of buses of each voltage level; combines the predicted load curve of buses of each voltage level and the mathematical model of cross-voltage level load transfer to obtain a load dispatching strategy across voltage levels. The present invention realizes coordinated prediction and optimized dispatch across voltage levels by integrating bus load information of different voltage levels, thereby enhancing the flexibility and adaptability of the overall operation of the power grid.

[0111] Specifically, constructing a mathematical model of load transfer across voltage levels according to the load transfer law across voltage levels and topological structure data includes:

[0112] Step S31, determining variables and parameters according to the load transfer law across voltage levels, and establishing a power balance equation according to the variables and parameters;

[0113] Select the load power of the relevant nodes under each voltage level as the variable, and set represents the active power of the ith node at the kth voltage level, represents the reactive power of node i. In addition, variables such as load transfer ratio are set. It represents the proportion of active load transfer from node i at the kth voltage level to node i at the lth voltage level.

[0114] Parameters: These are parameters related to the topology of the network, such as the impedance of each line (in Represents the line impedance connecting node m and node n at the kth voltage level, including resistance Hedian two parts); the initial load value of each node (assuming are the initial active and reactive loads of the ith node at the kth voltage level, respectively).

[0115] Active power balance:

[0116] For the entire power grid system, when considering load transfer across voltage levels, the sum of the injected active power of all nodes at each voltage level should be equal to the sum of the load active power of all nodes (including the transferred load). At the kth voltage level, its active power balance equation can be expressed as:

[0117]

[0118] Among them, Ω k represents the set of all nodes under the kth voltage level; l represents the voltage level, l≠k. represents the initial reactive power of node i at the kth voltage level; Represents the initial reactive power of node j at the lth voltage level; the left side of the above equation represents the total active power injected into the nodes at the lth voltage level after the transfer, the first item on the right side is the total initial active load of the nodes at this voltage level, and the second item is the total active load transferred from other voltage levels.

[0119] Reactive power balance:

[0120] Similarly, the reactive power balance equation at the kth voltage level can be written as:

[0121]

[0122] In the formula, Indicates the reactive load transfer ratio.

[0123] Step S32: setting the constraints of the model according to the load transfer rules across voltage levels and topological structure data;

[0124] (1) Node voltage constraint: The voltage amplitude of each node is within the preset reasonable operating range. represents the voltage amplitude of the ith node at the kth voltage level, then:

[0125]

[0126] In the formula, and are the lower and upper limits of the voltage amplitude of the i-th node under the k-th voltage level.

[0127] (2) Line power flow constraint: The power (active and reactive) passing through each line cannot exceed its thermal stability limit. For the line connecting node m and node n at the kth voltage level, the active power flow constraint can be expressed as:

[0128]

[0129] The reactive power flow constraint is similar to:

[0130]

[0131] and They are respectively the maximum active and reactive power limits allowed to pass through the line, which can be determined by factors such as the current carrying capacity of the line and the rated capacity of the equipment.

[0132] (3) Load transfer ratio constraint: The load transfer ratio is within the preset range.

[0133]

[0134] Step S33: construct an objective function of the total system loss, and obtain a mathematical model of load transfer across voltage levels by combining the power balance equation, constraint conditions and objective function.

[0135] The total system loss mainly considers the active power loss on the line. For the line connecting node m and node n at the kth voltage level, its active power loss is It can be calculated according to the formula:

[0136]

[0137] In the formula, and are respectively the active and reactive power passing through the line; is the voltage amplitude at the line head end node m; is the line resistance.

[0138] The total loss of the entire system is the sum of the active power losses of all lines, which can be expressed as:

[0139]

[0140] Wherein, Λk represents the set of all lines at the kth voltage level.

[0141] The goal is to minimize the sum of losses, and the objective function is expressed as:

[0142]

[0143] Combining the above power balance equation, constraints and objective function, a mathematical model of load transfer across voltage levels can be obtained.

[0144] In this embodiment, by constructing a mathematical model of load transfer, the load transfer between different voltage levels can be accurately simulated and analyzed, so that load scheduling and allocation can be performed more flexibly in actual operation. The mathematical model can comprehensively consider various factors (such as load demand, voltage level, grid topology, etc.), so as to formulate a more reasonable power resource allocation plan, improve energy utilization efficiency, and reduce unnecessary energy waste.

[0145] The embodiment of the present application further provides a coordinated dispatching system for cross-voltage level load forecasting, which is applied to the aforementioned coordinated dispatching method for cross-voltage level load forecasting, including:

[0146] A data acquisition unit, used to acquire historical load data and topological structure data of buses of various voltage levels in the power system;

[0147] The correlation analysis unit is used to perform correlation analysis on the historical load data of buses of each voltage level to obtain the load transfer rules across voltage levels;

[0148] A load transfer model building unit, used to build a mathematical model of load transfer across voltage levels according to the load transfer law across voltage levels and topological structure data;

[0149] The load forecasting model training unit is used to input the historical load data of buses of various voltage levels into the deep learning LSTM model for training and learning, so as to obtain a trained bus load forecasting model capable of predicting various voltage levels;

[0150] A load prediction unit is used to predict the bus loads of each voltage level by using the trained bus load prediction model to obtain a predicted load curve of each voltage level bus;

[0151] The dispatching strategy generating unit is used to combine the predicted load curves of buses of each voltage level and the mathematical model of load transfer across voltage levels to obtain the load dispatching strategy across voltage levels.

[0152] The functional explanation of each unit of a coordinated dispatching system for cross-voltage level load forecasting in this embodiment is the same as that of an embodiment of a coordinated dispatching method for cross-voltage level load forecasting, and the technical effects are the same, so they will not be repeated here.

[0153] An embodiment of the present application also provides a computer-readable storage medium, which includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the aforementioned coordinated scheduling method for cross-voltage level load forecasting.

[0154] The technical effect of this embodiment is the same as the technical effect of a coordinated scheduling method for load forecasting across voltage levels in an embodiment, and will not be repeated here.

[0155] The present invention can be used in many general or special computer system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and the like.

[0156] An embodiment of the present application also provides a processor, which is used to run a program, wherein the program executes the aforementioned coordinated scheduling method for load prediction across voltage levels when running.

[0157] The technical effect of this embodiment is the same as the technical effect of a coordinated scheduling method for load forecasting across voltage levels in embodiment 1, and will not be repeated here.

[0158] The processor in this embodiment may be a central processing unit (CPU), a controller, a microcontroller, or other data processing chips.

[0159] Those of ordinary skill in the art will appreciate that the units of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition of each example has been generally described in terms of function in the above description. 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.

[0160] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0161] 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. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0162] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-0nlyMemory), random access memory (RAM, RandomAccessMemory), mobile hard disk, magnetic disk or optical disk, etc., which can store program code.

[0163] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A coordinated dispatching method for load forecasting across voltage levels, characterized in that: include: Obtain historical load data and topological structure data of buses of various voltage levels in the power system; Conduct correlation analysis on the historical load data of buses at each voltage level to obtain the load transfer rules across voltage levels; According to the load transfer law across voltage levels and topological structure data, a mathematical model of load transfer across voltage levels is constructed; The historical load data of buses of each voltage level is input into the deep learning LSTM model for training and learning, and a bus load prediction model that has been trained to predict each voltage level is obtained; Use the trained bus load prediction model to predict the bus load of each voltage level and obtain the predicted load curve of each voltage level bus; Combining the predicted load curves of buses at each voltage level and the mathematical model of load transfer across voltage levels, the load dispatch strategy across voltage levels is obtained.

2. A coordinated dispatching method for load forecasting across voltage levels according to claim 1, characterized in that: Also includes: The historical load data of buses at each voltage level are cleaned and normalized.

3. A coordinated dispatching method for load forecasting across voltage levels according to claim 2, characterized in that: The above-mentioned correlation analysis of the historical load data of buses of each voltage level to obtain the load transfer rules across voltage levels includes: By analyzing the historical load data of buses of different voltage levels in time series, the time-varying patterns of the loads of buses of different voltage levels are obtained. By performing cross-spectral analysis on the historical load data of buses at different voltage levels, the resonance or synchronization of loads at different voltage levels at different frequencies can be obtained; By conducting causal relationship analysis on the historical load data of buses of different voltage levels, the causal relationship between the loads of buses of different voltage levels is obtained; According to the variation law of bus loads of each voltage level over time, the resonance or synchronization of loads of different voltage levels at different frequencies and the causal relationship between bus loads of different voltage levels, the law of load transfer across voltage levels is obtained.

4. A coordinated dispatching method for load forecasting across voltage levels according to claim 3, characterized in that: The construction of a mathematical model of load transfer across voltage levels according to the load transfer law across voltage levels and topological structure data includes: Determine variables and parameters according to the load transfer law across voltage levels, and establish a power balance equation based on the variables and parameters; Set the constraints of the model according to the load transfer rules across voltage levels and topological structure data; An objective function of the total system loss is constructed, and a mathematical model of load transfer across voltage levels is obtained by combining the power balance equation, constraints and objective function.

5. The coordinated dispatching method for load forecasting across voltage levels according to claim 2 is characterized in that: The historical load data of buses of each voltage level are input into the deep learning LSTM model for training and learning, and the trained bus load prediction model capable of predicting each voltage level is obtained, including: According to the load data characteristics of buses of each voltage level, characteristic data of load impact on buses of each voltage level are extracted; Divide the influencing characteristic data of buses of different voltage levels into training sets and test sets corresponding to different levels; The training sets and test sets corresponding to different levels are input into the deep learning LSTM model for training and learning, and a bus load prediction model that has been trained to predict each voltage level is obtained.

6. A coordinated dispatching method for load forecasting across voltage levels according to claim 5, characterized in that: The method of using the trained bus load prediction model to predict the bus load of each voltage level to obtain the predicted load curve of each voltage level bus includes: Input the currently acquired load impact characteristic data into the trained bus load prediction model, and output the load prediction values ​​corresponding to a series of time points of the buses of each voltage level; Process the load forecast values ​​of buses at each voltage level to remove abnormal values ​​and interpolate missing values; According to the processed load forecast value, the time axis and load axis are set to construct the forecast load curve of each voltage level bus.

7. A coordinated dispatching method for load forecasting across voltage levels according to claim 4, characterized in that: The load dispatching strategy across voltage levels obtained by combining the predicted load curves of buses of each voltage level and the mathematical model of load transfer across voltage levels includes: According to the predicted load curve of each voltage level bus, the load prediction value of each voltage level bus at the same time point is obtained; the load prediction value of each voltage level bus at the same time point is substituted into the cross-voltage level load transfer mathematical model as the boundary condition; By solving the mathematical model of load transfer across voltage levels, the load transfer path and transfer amount between different voltage levels are obtained; based on the load transfer path and transfer amount, the load scheduling strategy across voltage levels is obtained.

8. A coordinated dispatching system for load forecasting across voltage levels, characterized in that: The method applied to any one of claims 1 to 7, comprising: A data acquisition unit, used to acquire historical load data and topological structure data of buses of various voltage levels in the power system; The correlation analysis unit is used to perform correlation analysis on the historical load data of buses of each voltage level to obtain the load transfer rules across voltage levels; A load transfer model building unit, used to build a mathematical model of load transfer across voltage levels according to the load transfer law across voltage levels and topological structure data; The load forecasting model training unit is used to input the historical load data of buses of various voltage levels into the deep learning LSTM model for training and learning, so as to obtain a trained bus load forecasting model capable of predicting various voltage levels; A load prediction unit is used to predict the bus loads of each voltage level by using the trained bus load prediction model to obtain a predicted load curve of each voltage level bus; The dispatching strategy generating unit is used to combine the predicted load curves of buses of each voltage level and the mathematical model of load transfer across voltage levels to obtain the load dispatching strategy across voltage levels.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute a coordinated scheduling method for cross-voltage level load forecasting as described in any one of claims 1 to 7.

10. A processor, characterized in that: The processor is used to run a program, wherein the program, when running, executes a coordinated scheduling method for cross-voltage level load forecasting as described in any one of claims 1 to 7.

Images