A power probability distribution prediction method for key section of power grid based on deep learning

Abstract

The application discloses a kind of power grid key section power probability distribution prediction methods based on deep learning, comprising: first, new energy output and load power probability distribution is expressed as the effective input feature required by deep learning model, and the power grid key section power probability distribution is expressed as the output feature of deep learning model;Second, using the new energy output and load power probability distribution, determine the power grid key section power non-parametric probability distribution;Establish the sample set containing input and output feature vectors;Then, use sample set to train deep learning probability distribution prediction model;Finally, the new energy output and load power probability distribution feature vector corresponding to the power grid key section power probability distribution to be predicted are input to the trained deep learning probability distribution prediction model, and the power grid key section power probability distribution to be predicted is obtained by model output. The application guarantees the accuracy of power grid key section power probability distribution prediction while improving the efficiency of prediction.

Description

Technical Field

[0001] This application relates to the field of power grid safe operation and control technology, and in particular to a deep learning-based method for predicting the power probability distribution of key power sections in a power grid. Background Technology

[0002] Understanding the transmission power limits of each transmission section, and ensuring that they can meet load requirements without exceeding these limits, is highly beneficial for the stability and economy of power system operation. In traditional power grids, the impact of uncertainties is relatively small, and deterministic power flow calculations are sufficient to estimate the overall operating state of the system. However, in modern power systems, due to the significant increase in the proportion of renewable energy sources, the increased uncertainties have drastically changed the power flow patterns, greatly increasing the uncertainty of power flow at each transmission section. Traditional methods are insufficient to assess whether the system can operate safely and stably. Therefore, obtaining the probability distribution of power at each key transmission section while considering the uncertainties of renewable energy sources and load fluctuations has become an urgent problem to be solved.

[0003] The foundation for solving the power probability distribution at critical sections lies in uncertainty analysis methods, represented by probabilistic power flow. Existing technologies include three probabilistic power flow algorithms: simulation, analytical, and approximation. However, these methods still face challenges in practical applications within high-proportion renewable energy power systems. The main reasons are: approximation and analytical methods are conceptually clear and computationally fast, but often exhibit large errors in systems with high renewable energy penetration; simulation methods offer high computational accuracy, but the high computational burden resulting from large-scale sampling and deterministic power flow calculations hinders their online application.

[0004] Therefore, there is an urgent need in related technologies for a way to improve the accuracy and speed of predicting the power probability distribution at key sections of the power grid. Summary of the Invention

[0005] Therefore, it is necessary to provide a deep learning-based method for predicting the power probability distribution of critical sections of a power grid, which can improve the accuracy and speed of power probability distribution prediction at the aforementioned technical issues.

[0006] Firstly, this application provides a deep learning-based method for predicting the power probability distribution of critical sections of a power grid. The method includes the following steps:

[0007] S1. Represent the probability distribution of new energy output and load power as the effective input features required by the deep learning model, and represent the probability distribution of power at key sections of the power grid as the output features of the deep learning model.

[0008] S2. Using the power output and load power probability distribution of the new energy sources, determine the nonparametric probability distribution of power at key sections of the power grid, and establish a sample set containing input and output feature vectors;

[0009] S3. Use the sample set to train a deep learning probability distribution prediction model;

[0010] S4. Input the feature vectors of the new energy output and load power probability distribution corresponding to the power probability distribution of the key section of the power grid to be predicted into the trained deep learning probability distribution prediction model, and the model outputs the power probability distribution of the key section of the power grid to be predicted.

[0011] Optionally, in one embodiment of this application, if the operating mode of the grid synchronous generator is different from the operating mode used when training the deep learning probability distribution prediction model, the power probability distribution of the key section of the grid obtained in S4 is corrected.

[0012] Optionally, in one embodiment of this application, obtaining the effective input features and output features in S1 includes the following methods:

[0013] Take M points in the range (0, 1] for the cumulative distribution function of each new energy source output, load power, and power at key sections of the power grid. Based on the domain values ​​corresponding to these M points, obtain the power quantiles, thus constructing a system with dimension (K). r +K l The eigenvectors of +L)×M;

[0014] Among them, K r K l L represents the number of uncertain new energy sources, loads, and key grid sections in the power grid, respectively; K r ×M, K l ×M represents the dimension of the input feature vector formed by the probability distributions of new energy output and load power, respectively; L×M represents the dimension of the output feature vector formed by the probability distributions of power at key sections of the power grid.

[0015] Optionally, in one embodiment of this application, in S2, the method for determining the nonparametric probability distribution of power at key sections of the power grid includes:

[0016] Using the power output and load probability distribution of the new energy sources, the power of key sections is obtained by sampling using the Monte Carlo method, and then the nonparametric probability distribution of the power of key sections of the power grid is obtained by nonparametric estimation.

[0017] Optionally, in one embodiment of this application, the power probability distribution of the key section of the power grid obtained from the model output in S4 is corrected based on the sensitivity perturbation value of the synchronous motor output to the active power of the key section of the power grid.

[0018] Optionally, in one embodiment of this application, the sensitivity perturbation value is expressed as:

[0019]

[0020] Among them, P L,l Let P' be the active power at the l-th critical section, where l = 1, 2, ..., L. L,l ε represents the power at the l-th critical section after the change in the active power of the synchronous generator, and ε is the change in the active power of the synchronous generator.

[0021] Secondly, this application also provides a deep learning-based device for predicting the power probability distribution of critical sections of a power grid. The device includes:

[0022] The data acquisition and processing module is used to acquire the probability distribution of new energy output and load power.

[0023] The input and output feature vector determination module is used to represent the probability distribution of new energy output and load power as the effective input features required by the deep learning model, and to represent the probability distribution of power at key sections of the power grid as the output features of the deep learning model.

[0024] The power nonparametric probability distribution establishment module for key power sections of the power grid is used to determine the power nonparametric probability distribution of key power sections of the power grid by utilizing the power output and load power probability distribution of the new energy sources, and to establish a sample set containing input and output feature vectors.

[0025] The deep learning probability distribution prediction model training module is used to train a deep learning probability distribution prediction model using a sample set containing input and output feature vectors.

[0026] The power probability distribution determination module for key power sections of the power grid is used to input the feature vectors of the new energy output and load power probability distribution corresponding to the power probability distribution of the key power sections of the power grid to be predicted into the trained deep learning probability distribution prediction model, and the model outputs the power probability distribution of the key power sections of the power grid to be predicted.

[0027] Thirdly, this application also provides a computer device. The computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the steps of the methods described in the various embodiments above.

[0028] Fourthly, this application also provides a computer-readable storage medium. The computer-readable storage medium stores a computer program thereon, which, when executed by a processor, implements the steps of the methods described in the various embodiments above.

[0029] Compared with the prior art, the present invention has the following beneficial effects;

[0030] 1. This invention addresses the probability distribution of power output and load power for each new energy source, and uses Monte Carlo simulation to determine the nonparametric probability distribution of power at key sections of the power grid. It then constructs a deep learning feature vector in the form of a probability distribution, and establishes a sample set containing input and output features accordingly.

[0031] 2. This invention can realize the nonlinear mapping from the probability distribution of renewable energy output and load power to the probability distribution of power at key sections of the power grid. The deep learning method used takes into account a large number of renewable energy output and load power probability distributions, and has higher calculation accuracy than the probabilistic power flow convolution method. The sensitivity calculated by the perturbation method solves the problem that the operating mode of the synchronous generator in the power grid is different from the operating mode used in the training of the deep learning model, thus improving the accuracy of power probability distribution prediction at key sections of the power grid. Attached Figure Description

[0032] Figure 1 This is an application environment diagram of a deep learning-based power probability distribution prediction method for key power sections of a power grid, as shown in one embodiment.

[0033] Figure 2 This is a flowchart illustrating a deep learning-based method for predicting the power probability distribution of critical sections of a power grid in one embodiment.

[0034] Figure 3 This is a schematic diagram of a sample generation method that uses a probability distribution as input and output in one embodiment.

[0035] Figure 4 This is a schematic diagram of Monte Carlo simulation sampling in one embodiment;

[0036] Figure 5 This is a schematic diagram illustrating the construction of a probability distribution feature vector based on the cumulative distribution function in one embodiment;

[0037] Figure 6 This is a schematic diagram of the prediction results of a deep learning probability distribution prediction model in one embodiment;

[0038] Figure 7 This is a schematic diagram illustrating the use of a sensitivity method to correct data results in one embodiment;

[0039] Figure 8 This is a schematic diagram of an IEEE-118 node system with a high proportion of new energy sources in one embodiment.

[0040] Figure 9 This is a schematic diagram comparing a deep learning probability distribution prediction model with a DC power flow convolution method in one embodiment;

[0041] Figure 10This is a structural block diagram of a deep learning-based power probability distribution prediction device for key power sections of a power grid, as shown in one embodiment.

[0042] Figure 11 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0044] This application provides a deep learning-based method for predicting the power probability distribution of critical sections in a power grid, which can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104 or located in the cloud or on other network servers. Terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, smart in-vehicle devices, etc. Portable wearable devices can include smartwatches, smart bracelets, head-mounted devices, etc. Server 104 can be implemented using a standalone server or a server cluster consisting of multiple servers.

[0045] In one embodiment, such as Figure 2 As shown, a deep learning-based method for predicting the power probability distribution of critical sections of a power grid is presented, and this method is applied to... Figure 1 Taking the server in the example, the following steps are included:

[0046] S201: Obtain the probability distribution of new energy output and load power.

[0047] In this embodiment of the application, firstly, a large amount of probability distribution data of new energy output and load power is collected.

[0048] S203: Represent the probability distribution of new energy output and load power as the effective input features required by the deep learning model, and represent the probability distribution of power at key sections of the power grid as the output features of the deep learning model.

[0049] In the embodiments of this application, such as Figure 3 The diagram shown illustrates the process of generating samples with input and output in the form of a probability distribution. P new and Q new These represent the active and reactive power outputs from new energy sources, P. load and Qload The active and reactive power of the load are respectively used for load power calculation, and the node voltage V is obtained after the power flow calculation. bus Phase angle δ bus Active power P at key sections of the power grid branch and reactive power Q branch The cumulative distribution function of each new energy source's output, load power, and power at key grid sections is used to select M points in the range (0,1]. Based on the domain values ​​corresponding to these M points, the power quantiles are obtained, forming a system with dimension (K). r +K l +L)×L eigenvectors, where K r K l L represents the number of uncertain new energy sources, loads, and key grid sections in the power grid, respectively; K r ×M, K l ×M represents the dimension of the input feature vector composed of the power output and load probability distributions of new energy sources, respectively, and L×M represents the dimension of the output feature vector composed of the nonparametric probability distribution of power at key sections of the power grid. This is then used to generate a feature vector with dimension S×(K). r +K l The input feature sample set is S×L×M, and the output feature sample set is S×L×M. For example... Figure 5 The diagram shown is a schematic of the construction of probability distribution feature vectors based on the cumulative distribution function.

[0050] S205: Using the power output and load power probability distribution of the new energy sources, determine the nonparametric probability distribution of power at key sections of the power grid, and establish a sample set containing input and output feature vectors.

[0051] In this embodiment of the application, for each probability distribution of new energy output and load power, a nonparametric probability distribution of power at key sections of the power grid is established using the Monte Carlo method. For example... Figure 4 The image shown is a schematic diagram of Monte Carlo simulation sampling.

[0052] In one embodiment of this application, the number of new energy sources and loads with hypothetical design and uncertainty is K, and their cumulative distribution function is Y. k =F k (X k Given k = 1, 2, ..., K, divide the range [0, 1] of each cumulative distribution function into N equally spaced intervals. Sample from each interval and select the midpoint y of each interval. kn As Y k The sampled values ​​are used to calculate X based on the inverse function of the cumulative distribution function. k The nth sample value is as follows:

[0053]

[0054] Where, xkn To obtain the new energy output and load power from the sampling, N samplings of K input random variables constitute an initial sampling matrix of dimension K×N, which is then randomly arranged to form the sampling matrix.

[0055] Based on the power output of new energy sources and the power of loads in each column of the sampling matrix, the power of L key sections of the power grid is obtained through power flow calculation. After N power flow calculations, a key section power matrix with dimension L×N is formed. Nonparametric estimation is then performed to obtain the power probability distribution of L key sections of the power grid.

[0056] S207: Use a sample set to train a deep learning probability distribution prediction model.

[0057] S209: Input the feature vectors of the new energy output and load power probability distribution corresponding to the power probability distribution of the key section of the power grid to be predicted into the trained deep learning probability distribution prediction model, and the model outputs the power probability distribution of the target key section of the power grid to be predicted.

[0058] In this embodiment, the sample sets of input and output feature vectors are divided into training sample sets and test sample sets. A deep learning probability distribution prediction model is trained based on the training sample set, and the deep learning probability distribution prediction model is tested based on the test sample set.

[0059] The training sample set is input into the deep learning probability distribution prediction model, and the deep learning probability distribution prediction model is continuously optimized using the loss function and activation function to complete the training of the deep learning probability distribution prediction model. The deep learning probability distribution model is then tested based on the test sample set.

[0060] The mean squared error function is selected as the loss function, and the specific formula is shown below.

[0061]

[0062] Where d is y k and dimensionality, y k The true value is the output feature vector. This is the output of the deep learning probability distribution prediction model, i.e., the initial output prediction result.

[0063] The model uses ReLU as its activation function. ReLU is a piecewise function that outputs 0 on the negative half of the horizontal axis and equals the x-coordinate value on the positive half. When ReLU is used as the activation function, the gradient vanishing phenomenon does not occur when the neuron is activated on the positive half. The specific formula is shown below.

[0064]

[0065] Where x represents the horizontal axis.

[0066] In one embodiment of this application, the method further includes:

[0067] Based on the sensitivity perturbation value of synchronous motor output to the active power of key sections of the power grid, the power probability distribution of the key sections of the power grid to be predicted obtained from the model output is corrected.

[0068] In one embodiment of this application, after model training is completed, the operating mode of the synchronous generators in the power grid during prediction may differ from the operating mode used during training. Therefore, based on the sensitivity perturbation value of the synchronous generator output to the active power of key sections of the power grid, the probability distribution of the power of the key sections of the power grid to be predicted obtained from the model output is corrected. Based on the power grid operating mode when the model is applied, power flow calculation is performed to obtain the power of L key sections.

[0069] Apply a change (a change in active power) to the output of the i-th synchronous generator:

[0070] P′ G,i =P G,i +ε

[0071] Among them, P G,i Let ε represent the active power output of the i-th generator, where i = 1, 2, ..., I, I is the total number of synchronous generators, and ε is the change in active power of the synchronous generators. Power flow calculations are performed to obtain the power at L key cross-sections after the change in active power of the synchronous generators.

[0072] In one embodiment of this application, the sensitivity perturbation value of the synchronous motor output to the active power of a critical section of the power grid can be expressed as:

[0073]

[0074] Among them, P L,l Let P' be the active power at the l-th critical section, where l = 1, 2, ..., L. L,l ε represents the power at the l-th critical section after the change in the active power of the synchronous generator, and ε is the change in the active power of the synchronous generator.

[0075] By applying a change in the output of each synchronous generator, the power change at each critical section of the power grid before and after the change is calculated, resulting in a sensitivity matrix λ with dimension L×I.

[0076]

[0077] The power probability distribution of key sections of the power grid obtained by the deep learning probability distribution prediction model is corrected based on the sensitivity matrix. The specific correction formula is shown below.

[0078] P′ branch =P branch +λΔP

[0079] Among them, P branch This represents the power probability distribution of key sections of the power grid obtained from a deep learning probability distribution prediction model, where ΔP is the difference in synchronous generator output between the application and training phases. For example... Figure 6 , 7 The figures shown are schematic diagrams illustrating the prediction results of the deep learning probability distribution prediction model and the results of the sensitivity correction data calculated using the perturbation method, respectively. (See the table below.) Figure 6 , Figure 7 It can be seen that, based on the power prediction range of the critical section of the power grid with a confidence level of 95%, the actual power fall rate of the critical section of the power grid is higher after sensitivity correction.

[0080]

[0081] In this embodiment, the power probability distribution of the key section of the power grid is corrected by using the sensitivity calculated by the perturbation method, which solves the problem that the operating mode of the synchronous generator of the power grid is different from the operating mode used when training the deep learning probability distribution prediction model.

[0082] In one embodiment of this application, such as Figure 8 The diagram shows a schematic of an IEEE-118 node system with a high proportion of renewable energy. Traditional turbines at nodes 59 and 61 are replaced by photovoltaic power plants, while traditional turbines at nodes 25, 26, 49, 65, 66, 100, 103, and 111 are replaced by wind turbines. The renewable energy penetration rate is 49.44%. The improved IEEE-118 node system with renewable energy access, featuring a complete topology, generated a total of 10,000 samples, with 80% used as the training set and 20% as the test set. An AC-based Monte Carlo (ACMC) simulation method was used as a benchmark. Comparisons were made between DC-Conv (DC power flow and convolution analysis), MLP (Multilayer Perceptron-based Key Section Power Probability Distribution), CNN (Convolutional Neural Network-based Key Section Power Probability Distribution), LSTM (Long Short-Term Memory Network-based Key Section Power Probability Distribution), and GCN (Graph Convolutional Network-based Key Section Power Probability Distribution). The comparison results are shown below. Figure 9 As shown in the table below. Figure 9 It can be seen that the method based on the deep learning probability distribution prediction model has higher calculation accuracy than the probabilistic power flow convolution method in high-proportion new energy power systems.

[0083]

[0084] Based on the same inventive concept, this application also provides a deep learning-based device for predicting the power probability distribution of critical sections of a power grid, used to implement the aforementioned deep learning-based method for predicting the power probability distribution of critical sections of a power grid. The solution provided by this device is similar to the solution described in the above method. Therefore, the specific limitations of one or more embodiments of the deep learning-based device for predicting the power probability distribution of critical sections of a power grid provided below can be found in the limitations of the deep learning-based method for predicting the power probability distribution of critical sections of a power grid described above, and will not be repeated here.

[0085] In one embodiment, such as Figure 10 As shown, a deep learning-based power probability distribution prediction device 1000 for key power sections of a power grid is provided, comprising: a data acquisition and processing module 1001, an input and output feature vector determination module 1003, a power nonparametric probability distribution establishment module 1005, a deep learning probability distribution prediction model training module 1007, and a power probability distribution determination module 1009, wherein:

[0086] The data acquisition and processing module 1001 is used to acquire the probability distribution of new energy output and load power.

[0087] The input and output feature vector determination module 1003 is used to represent the probability distribution of new energy output and load power as the effective input features required by the deep learning model, and to represent the nonparametric probability distribution of power at key sections of the power grid as the output features of the deep learning model.

[0088] The power nonparametric probability distribution establishment module 1005 for key power sections of the power grid is used to determine the power nonparametric probability distribution of key power sections of the power grid by utilizing the power output and load power probability distribution of the new energy sources, and to establish a sample set containing input and output feature vectors.

[0089] The Deep Learning Probability Distribution Prediction Model Training Module 1007 is used to train a deep learning probability distribution prediction model using a sample set containing input and output feature vectors.

[0090] The power probability distribution determination module 1009 for key power sections of the power grid is used to input the feature vectors of the new energy output and load power probability distribution corresponding to the power probability distribution of the key power sections of the power grid to be predicted into the trained deep learning probability distribution prediction model, and the model outputs the power probability distribution of the key power sections of the power grid to be predicted.

[0091] In one embodiment of this application, the data acquisition and processing module is further configured to:

[0092] Take M points in the range (0, 1] for the cumulative distribution function of each new energy source output, load power, and power at key sections of the power grid. Based on the domain values ​​corresponding to these M points, obtain the power quantiles, thus constructing a system with dimension (K). r +K l The eigenvectors of +L)×M;

[0093] Among them, K r ×M, K l L×L represents the dimension of the input feature vector formed by the probability distributions of new energy output and load power, respectively; L×M represents the dimension of the output feature vector formed by the probability distributions of power at key sections of the power grid.

[0094] In one embodiment of this application, the power nonparametric probability distribution establishment module for key power sections of the power grid is further configured to:

[0095] Using the power output and load power probability distribution of the new energy sources, the power of key sections of the power grid is obtained by sampling using the Monte Carlo method, and then the nonparametric probability distribution of the power of key sections of the power grid is obtained by nonparametric estimation.

[0096] In one embodiment of this application, the deep learning probability distribution prediction model training module is further configured to:

[0097] The sample set containing input and output feature vectors is divided into a training sample set and a test sample set;

[0098] The training sample set is input into the deep learning probability distribution prediction model to obtain the initial output prediction result.

[0099] The deep learning-based power probability distribution prediction device for critical power grid sections further includes a power probability distribution correction module for critical power grid sections. In one embodiment of this application, the power probability distribution correction module for critical power grid sections is used for:

[0100] During prediction, if the operating mode of the synchronous generator in the power grid is different from the operating mode used when training the deep learning probability distribution prediction model, the power probability distribution of the key section to be predicted obtained from the model output in S4 will be corrected based on the sensitivity perturbation value of the synchronous generator output to the active power of the key section of the power grid.

[0101] In one embodiment of this application, the sensitivity can be expressed as:

[0102]

[0103] Among them, P L,l Let P' be the active power at the l-th critical section, where l = 1, 2, ..., L. L,lε represents the power at the l-th critical section after the change in the active power of the synchronous generator, and ε is the change in the active power of the synchronous generator.

[0104] The modules in the aforementioned deep learning-based power probability distribution prediction device for critical power grid sections can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the computer device's memory as software, so that the processor can call and execute the corresponding operations of each module.

[0105] In one embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 11 As shown, the computer device includes a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a deep learning-based method for predicting the probability distribution of power at critical sections of a power grid. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0106] Those skilled in the art will understand that Figure 11 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0107] In one embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0108] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0109] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0110] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0111] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0112] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0113] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for predicting the probability distribution of power at key sections of a power grid based on deep learning, characterized in that, The method includes: S1. Represent the probability distribution of new energy output and load power as the effective input features required by the deep learning model, and represent the probability distribution of power at key sections of the power grid as the output features of the deep learning model. Includes the following steps: The cumulative distribution function of each of the aforementioned new energy sources' output, load power, and power at key grid sections is defined in the range of... Pick Points, according to the The domain values ​​corresponding to each point are used to obtain the power quantiles, forming a dimension of ( ). eigenvectors; in, , , These represent uncertain new energy sources, loads, and the number of key sections of the power grid, respectively. , These are the dimensions of the input feature vectors formed by the probability distributions of new energy output and load power, respectively. The dimension of the output feature vector formed by the power probability distribution of key sections of the power grid; S2. Using the power output and load power probability distribution of the new energy source, the power of the key section is obtained by sampling through the Monte Carlo method. Then, the non-parametric probability distribution of the power of the key section of the power grid is obtained through non-parametric estimation, and a sample set containing input and output feature vectors is established. S3. Use the sample set to train a deep learning probability distribution prediction model; S4. Input the feature vectors of the new energy output and load power probability distribution corresponding to the power probability distribution of the key section of the power grid to be predicted into the trained deep learning probability distribution prediction model, and the model outputs the power probability distribution of the key section of the power grid to be predicted.

2. The method for predicting the probability distribution of power at key sections of a power grid based on deep learning according to claim 1, characterized in that, If the operating mode of the grid synchronous generator is different from the operating mode used when training the deep learning probability distribution prediction model, the power probability distribution of the grid key section obtained by S4 will be corrected.

3. The method for predicting the probability distribution of power at key sections of a power grid based on deep learning according to claim 2, characterized in that, Based on the sensitivity perturbation value of synchronous motor output to the active power of key sections of the power grid, the probability distribution of power of key sections of the power grid to be predicted obtained from the model output in S4 is corrected.

4. The method for predicting the probability distribution of power at critical sections of a power grid based on deep learning according to claim 3, characterized in that, The sensitivity perturbation value is expressed as: in, For the first Active power at key cross-sections of each target , The first time after the change in active power of the synchronous generator Power at key cross sections This represents the change in the active power of the synchronous generator.

5. A deep learning-based power probability distribution prediction device for critical power sections of a power grid, using the method described in any one of claims 1-4, characterized in that, include: The data acquisition and processing module is used to acquire the probability distribution of new energy output and load power. The input and output feature vector determination module is used to represent the probability distribution of new energy output and load power as the effective input features required by the deep learning model, and to represent the probability distribution of power at key sections of the power grid as the output features of the deep learning model. The power nonparametric probability distribution establishment module for key power sections of the power grid is used to determine the power nonparametric probability distribution of key power sections of the power grid by utilizing the power output and load power probability distribution of the new energy sources, and to establish a sample set containing input and output feature vectors. The deep learning probability distribution prediction model training module is used to train a deep learning probability distribution prediction model using a sample set containing input and output feature vectors. The power probability distribution determination module for key power sections of the power grid is used to input the feature vectors of the new energy output and load power probability distribution corresponding to the power probability distribution of the key power sections of the power grid to be predicted into the trained deep learning probability distribution prediction model, and the model outputs the power probability distribution of the key power sections of the power grid to be predicted.

6. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 4.

Images