Power distribution network coordination evaluation method based on graph convolutional neural network

Abstract

The invention provides a power distribution network coordination evaluation method based on a graph convolutional neural network, and relates to the technical field of power system optimization control, and the method comprises the steps: carrying out the coordination evaluation of a power distribution network containing a high-proportion distributed power supply, so as to obtain a coordination evaluation index of the power distribution network; obtaining power flow data of the power distribution network, and processing the power flow data through the graph convolutional neural network; calculating the index weight of the coordination evaluation index by using an analytic hierarchy process, and adding the index weight as a hidden layer into the graph convolutional neural network to construct an improved graph convolutional neural network; inputting the relation characteristics between the nodes and the topological structure information of the network into an improved graph convolutional neural network model to obtain an evaluation index of the power distribution network; performing coordination evaluation according to the evaluation indexes and the index weights; according to the method, comprehensive and accurate coordination evaluation is carried out on the power distribution network connected with the high-proportion distributed power supply, the evaluation accuracy is improved, the real-time processing capacity is improved, and the coordination state of the power distribution network is comprehensively reflected.

Description

Technical Field

[0001] The present invention relates to the technical field of power system optimization control, and in particular to a distribution network coordination evaluation method based on graph convolutional neural network. Background Art

[0002] With the widespread global attention to renewable energy, the application of distributed power sources (such as solar energy, wind energy, etc.) in power systems is becoming increasingly popular. The access of distributed power sources has brought opportunities for sustainable development to the power system, but it has also triggered a series of challenges. These challenges include how to effectively coordinate the relationship between various power sources and loads to ensure the safe, stable and efficient operation of the distribution network. Therefore, it is particularly important to carry out research on the coordination evaluation of distribution networks with a high proportion of distributed power sources. In recent years, the rapid development of distributed power sources has significantly increased their proportion in distribution networks. These power sources often have volatility and uncertainty, which poses challenges to traditional power system scheduling and control methods. For example, photovoltaic power generation will experience output power fluctuations when the sunshine conditions change, while wind power generation is significantly affected by climatic conditions. This volatility not only affects the stability of power supply, but may also cause voltage fluctuations and power quality degradation. The structure of the modern distribution network is no longer a simple linear network, but a complex multi-level and multi-type network system; it consists of multiple substations, distributed power sources and user loads, and there are complex interactions between these components. In this context, how to effectively evaluate the coordination of the distribution network and ensure the coordinated work of each part has become an important research topic. Coordination evaluation is an important means to evaluate the mutual cooperation and overall performance of the various components in the distribution network. Effective coordination evaluation can not only timely discover potential operating problems of the system, but also provide a scientific basis for the dispatch optimization and fault handling of the distribution network. By establishing a reasonable evaluation index system, distribution network operators can better understand the operating status of the power system and improve the accuracy and timeliness of decision-making.

[0003] At present, most of the coordination evaluation methods of distribution networks are based on traditional mathematical models and empirical formulas. These methods cannot fully reflect the complexity and dynamic characteristics of distribution networks to some extent. Many methods only rely on quantitative indicators, ignoring the spatial relationship and mutual influence between nodes, resulting in limitations in the evaluation results. In addition, the existing methods have low computational efficiency when facing large-scale data, making it difficult to achieve real-time response and unable to adapt to the needs of modern distribution networks. Summary of the invention

[0004] The purpose of the present invention is to provide a distribution network coordination evaluation method based on graph convolutional neural network, which improves the accuracy of evaluation, enhances the real-time processing capability, and can more comprehensively reflect the coordination status of the distribution network by conducting a comprehensive and accurate coordination evaluation on the distribution network with a high proportion of distributed power sources connected.

[0005] The technical solution of the present invention is:

[0006] In a first aspect, the present application provides a distribution network coordination evaluation method based on a graph convolutional neural network, which comprises the following steps:

[0007] S1. Conduct coordination evaluation on the distribution network containing a high proportion of distributed generation to obtain coordination evaluation indicators of the distribution network;

[0008] S2, obtaining the power flow data of the distribution network, and processing the power flow data through a graph convolutional neural network to extract the relationship characteristics between nodes and the topological structure information of the network;

[0009] S3. Calculate the index weights of the coordination evaluation index using the hierarchical analysis method and add them as hidden layers to the graph convolutional neural network to construct an improved graph convolutional neural network.

[0010] S4, inputting the relationship characteristics between nodes and the topological structure information of the network into the improved graph convolutional neural network model to obtain the evaluation index of the distribution network;

[0011] S5. Perform a coordination evaluation of the distribution network according to the evaluation indexes and index weights of the distribution network to obtain a coordination evaluation result of the distribution network.

[0012] Furthermore, in step S1, the coordination evaluation index of the distribution network includes a three-layer index system established by the power source and power grid coordination evaluation index, the power grid and user coordination evaluation index and the isolated network operation capability evaluation index.

[0013] Furthermore, the evaluation indicators for power supply and grid coordination include the matching degree between distributed power supply and grid, power supply penetration rate, the proportion of distributed power supply in total power supply, power supply output prediction accuracy and the deviation between prediction and actual output. The calculation formula includes:

[0014]

[0015] p=|ppred-pact|

[0016] In the formula, A represents power matching, P represents the output power of distributed power generation, Q represents the grid load, B represents the penetration rate of distributed power generation, C represents the installed capacity of distributed power generation, H represents the total load peak, X represents the proportion of distributed power generation in the total power supply, and C represents the total load peak. 总represents the total installed capacity of the power system, Accuracy represents the relative error, MAE represents the average absolute error between the predicted value and the actual value, represents the average actual value, n represents the number of samples, y i Indicates the actual value, represents the predicted value, p represents the deviation between the predicted value and the actual value, and p pred It represents the predicted output of distributed generation in a certain period of time, p act Indicates the actual output of distributed generation in a certain period of time.

[0017] Furthermore, the evaluation indicators of power grid and user coordination include user demand response, user response rate during peak and valley periods, proportion of users participating in peak and valley scheduling, demand side management effect and power saving of demand side management projects; among them, the calculation formula of user response during peak and valley periods is:

[0018]

[0019] In the formula, S represents the user response rate, V' represents the reduced power consumption during the peak and valley periods, and V represents the original power consumption during the peak and valley periods;

[0020] The calculation formula for electricity saving of demand side management project is:

[0021] U=U1-U2

[0022] In the formula, U represents the amount of electricity saved, U1 represents the amount of electricity not consumed, and U2 represents the amount of electricity consumed.

[0023] Furthermore, the evaluation indicators of the isolated grid operation capability include the isolated grid stable operation index, the isolated grid duration and the longest isolated grid operation time, the voltage stability during the isolated grid period and the voltage fluctuation range during the isolated grid period;

[0024] Among them, the calculation formula for the voltage fluctuation range during the isolated grid period is:

[0025]

[0026] In the formula, S 波 Indicates the voltage fluctuation range during isolated grid, V max Indicates the maximum voltage monitored, V min Indicates the minimum voltage detected, V rated Indicates the reference voltage of the system.

[0027] Furthermore, in step S2, the power flow data includes voltage, current amplitude, phase angle, active power generation and reactive power generation.

[0028] Furthermore, in step S3, the calculation process of calculating the index weight of the coordination evaluation index using the hierarchical analysis method includes:

[0029] Construct a judgment matrix:

[0030]

[0031] Hierarchical single sorting:

[0032] ZW=λmaxW

[0033]

[0034] Consistency check:

[0035] LW=λW

[0036]

[0037] In the formula, Z represents the judgment matrix, a ij represents the element in the i-th row and j-th column of the judgment matrix, W represents the eigenvector, and λ max represents the largest characteristic root of Z, W i represents the weight, L is the n-order judgment matrix, λ represents the characteristic root, CI represents the consistency index, RI represents the consistency expectation value, and CR represents the index of whether the judgment matrix has satisfactory consistency.

[0038] In a second aspect, the present application provides an electronic device, including:

[0039] A memory for storing one or more programs;

[0040] processor;

[0041] When the one or more programs are executed by the processor, a distribution network coordination evaluation method based on a graph convolutional neural network as described in any one of the first aspects is implemented.

[0042] In a third aspect, the present application provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a distribution network coordination evaluation method based on a graph convolutional neural network as described in any one of the first aspects above.

[0043] Compared with the prior art, the present invention has at least the following advantages or beneficial effects:

[0044] (1) The present invention provides a distribution network coordination evaluation method based on graph convolutional neural network. By combining graph convolutional neural network and hierarchical analysis method to construct an improved graph convolutional neural network model, the coordination evaluation of the distribution network with a high proportion of distributed power generation access is realized, providing decision support and scientific basis for operation management;

[0045] (2) The present invention defines multi-dimensional coordination evaluation indicators at the same time to form a comprehensive evaluation system. The spatial characteristics of the distribution network are extracted through GCN, and weight allocation and comprehensive evaluation are combined with AHP to perform indicator scoring calculation, which significantly improves the accuracy and reliability of the model. The coordination of the distribution network is evaluated in real time through the trained model, which not only improves the accuracy of the evaluation, but also improves the real-time processing capability, providing a new solution for the operation and management of the distribution network. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 The present invention provides a distribution network coordination evaluation method based on graph convolutional neural network;

[0048] Figure 2 Schematic diagram for modeling the graph structure of the distribution network;

[0049] Figure 3 This is the hierarchical structure diagram of the distribution network coordination evaluation index;

[0050] Figure 4 Schematic diagram of the graph convolutional neural network structure;

[0051] Figure 5 This is a schematic diagram of the improved graph convolutional neural network structure of the present invention;

[0052] Figure 6 The figure is a schematic structural block diagram of an electronic device according to an embodiment of the present invention.

[0053] Icon: 101, memory; 102, processor; 103, communication interface. DETAILED DESCRIPTION

[0054] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. The components of the embodiments of the present application described and shown in the drawings here can be arranged and designed in various different configurations.

[0055] Therefore, the following detailed description of the embodiments of the present application provided in the accompanying drawings is not intended to limit the scope of the present application for which protection is sought, but merely represents selected embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in the field without creative work are within the scope of protection of the present application.

[0056] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0057] It should be noted that, in this article, the term "comprises" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprises..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0058] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0059] Example 1

[0060] See also Figure 1 , Figure 1 Shown is a step diagram of a distribution network coordination evaluation method based on a graph convolutional neural network provided in an embodiment of the present application.

[0061] The present application provides a distribution network coordination evaluation method based on a graph convolutional neural network, which comprises the following steps:

[0062] S1. Conduct coordination evaluation on the distribution network containing a high proportion of distributed generation to obtain coordination evaluation indicators of the distribution network;

[0063] S2, obtaining the power flow data of the distribution network, and processing the power flow data through a graph convolutional neural network to extract the relationship characteristics between nodes and the topological structure information of the network;

[0064] S3. Calculate the index weights of the coordination evaluation index using the hierarchical analysis method and add them as hidden layers to the graph convolutional neural network to construct an improved graph convolutional neural network.

[0065] S4, inputting the relationship characteristics between nodes and the topological structure information of the network into the improved graph convolutional neural network model to obtain the evaluation index of the distribution network;

[0066] S5. Perform a coordination evaluation of the distribution network according to the evaluation indexes and index weights of the distribution network to obtain a coordination evaluation result of the distribution network.

[0067] As a preferred implementation, in step S1, the coordination evaluation index of the distribution network includes a three-layer index system established by the power supply and power grid coordination evaluation index, the power grid and user coordination evaluation index and the isolated network operation capability evaluation index.

[0068] Among them, it should be noted that in step S2, the process of processing the power flow data through the graph convolutional neural network to extract the relationship characteristics between nodes and the topological structure information of the network includes the graph structure modeling of the distribution network and the simulation of the power flow data. It first constructs the graph model by defining nodes and edges. Nodes represent key equipment such as substations and distributed power sources in the distribution network, and have attributes such as capacity, type and location. Edges represent power transmission lines between nodes, containing information such as impedance and capacity. Next, based on the definitions of these nodes and edges, an adjacency matrix is ​​constructed to represent the connection relationship and strength between nodes. The adjacency matrix is ​​an important input for subsequent analysis and model training. Through this modeling method, the complex relationship of the distribution network can be effectively represented, laying the foundation for further power flow data simulation and coordination evaluation. Finally, the graph structure can provide necessary support for the subsequent graph convolutional neural network (GCN) model to help capture the complex interaction characteristics between nodes;

[0069] The improved IEEE 33-node distribution network model with distributed power generation is selected for modeling and analysis. The IEEE 33-node distribution network model contains 33 electrical nodes, which are interconnected by lines and transformers, simulating the complex structure of the actual power network. Figure 2 As shown, the system voltage level is set to 12.66kV, the frequency is 50Hz, the total active power of the load is 5.4MW, the distributed power source adopts inverter photovoltaic power generation and wind power generation, and the total active power of the injected distributed power source is 5.4MW, of which the photovoltaic power generation is 0.8MW and the wind power generation is 1.5MW. The distribution network is connected to the large power grid at node 1 and connected to 5 distributed power sources;

[0070] Then, PSASP was used for simulation to establish an IEEE33-node distribution network model with 5 distributed power sources. In the power flow calculation stage, the load demand of each node and the output range of the distributed power sources were set. By simulating various load conditions and power generation modes (such as sunny and cloudy days), the power flow data was collected every 15 minutes starting from zero o'clock on a certain day for three months, and 8544 sets of data were finally obtained to form the sample set. According to the ratio of 7:1:1, it was divided into training set, validation set and test set, so there were 6646 sets of training set sample data, and the validation set and test set each contained 949 sets.

[0071] It should be noted that the evaluation index system of the distribution network generally consists of two parts: the name definition of the evaluation index and the numerical calculation of the evaluation index. The name definition of the evaluation index is a description that reflects some essential attributes of the evaluation object, namely the distribution network; the numerical calculation of the evaluation index can quantify the specific performance of a certain aspect of the distribution network under certain conditions. The present invention uses a multi-level attribute graph model as a data storage support, considers the coordination research of the distribution network, and constructs a three-level indicator system for the evaluation of the coordination status of the distribution network. The hierarchical architecture diagram of the distribution network coordination evaluation index of the indicator system is shown in the figure. Figure 3 As shown. Among them, the distribution network coordination evaluation index system is divided into five layers. The first layer is the target layer coordination; the next layer is the criterion layer, which includes three criterion dimensions: power supply and power grid coordination, power grid and user coordination, and isolated network operation capability; followed by the first-level indicator layer, which includes the matching degree between distributed power supply and power grid A1, user demand response A2, and stable operation of isolated network A3; followed by the second-level indicator layer represented by power supply penetration rate B1, user response rate during peak and valley periods B3, and isolated network duration B5, and the third-level basic indicator layer C1-C6 represented by the proportion of distributed power supply in the total power supply C1, the proportion of users participating in peak and valley scheduling C3, and the longest isolated network operation time C5;

[0072] In addition, the graph convolutional neural network (GCN) used in step S3 is an extension of the convolutional neural network (CNN) in non-Euclidean space, such as Figure 4 As shown. GCN is a feature extraction algorithm based on graph data, and usually uses convolution for calculation in the graph structure. The input data of GCN is a graph structure. By performing convolution operations on the adjacent edges of each node, the convolution result is used to update this node, and then the activation function is operated. The above two steps are repeated until the expected effect of network training is achieved, and finally the relevant label output is obtained through the output function.

[0073] As a preferred implementation, the power supply and grid coordination evaluation index includes the matching degree between the distributed power supply and the grid, the power supply penetration rate, the proportion of the distributed power supply in the total power supply, the power supply output prediction accuracy and the deviation between the prediction and the actual output, and the calculation formula includes:

[0074]

[0075] In the formula, A represents power matching, P represents the output power of distributed power generation, Q represents the grid load, B represents the penetration rate of distributed power generation, C represents the installed capacity of distributed power generation, H represents the total load peak, X represents the proportion of distributed power generation in the total power supply, and C represents the total load peak. 总 represents the total installed capacity of the power system, Accuracy represents the relative error, MAE represents the average absolute error between the predicted value and the actual value, y represents the average actual value, n represents the number of samples, and y i Indicates the actual value, represents the predicted value, p represents the deviation between the predicted value and the actual value, and p pred It represents the predicted output of distributed generation in a certain period of time, p act Indicates the actual output of distributed generation in a certain period of time.

[0076] It should be noted that the coordination between power sources and power grids can increase the proportion of distributed energy, improve energy utilization efficiency, and reasonably allocate peak-shaving power sources; distributed energy refers to solar energy and wind energy, and the power generation or installed capacity of these energy sources will change with time, weather, equipment status and other factors. Therefore, when calculating the proportion of distributed power sources in the total power supply, it is necessary to ensure that the data used is accurate and up-to-date.

[0077] As a preferred implementation, the evaluation indexes for coordination between the power grid and users include user demand response, user response rate during peak and valley periods, proportion of users participating in peak and valley scheduling, demand side management effect, and power saving of demand side management projects; wherein, the calculation formula for user response during peak and valley periods is:

[0078]

[0079] In the formula, S represents the user response rate, V' represents the reduced power consumption during the peak and valley periods, and V represents the original power consumption during the peak and valley periods;

[0080] The calculation formula for electricity saving of demand side management project is:

[0081] U=U1-U2

[0082] In the formula, U represents the amount of electricity saved, U1 represents the amount of electricity not consumed, and U2 represents the amount of electricity consumed.

[0083] It should be noted that the coordination between the power grid and users not only helps to improve the reliability and economy of the power system, but also helps to achieve a higher proportion of distributed power access. Effective coordination between the power grid and users can be achieved through means such as demand response, load management and smart grid technology.

[0084] As a preferred implementation, the isolated grid operation capability evaluation index includes the isolated grid stable operation index, the isolated grid duration and the longest isolated grid operation time, the voltage stability during the isolated grid period and the voltage fluctuation range during the isolated grid period;

[0085] Among them, the calculation formula for the voltage fluctuation range during the isolated grid period is:

[0086]

[0087] In the formula, S 波 Indicates the voltage fluctuation range during isolated grid, V max Indicates the maximum voltage monitored, V min Indicates the minimum voltage detected, V rated Indicates the reference voltage of the system.

[0088] It should be noted that the isolated grid operation capability refers to the ability of the power system to operate independently and provide stable power to specific areas or users after being disconnected from the main power grid. This evaluation indicator is particularly critical when responding to emergencies (such as natural disasters or equipment failures). It can ensure the continuous power supply of key facilities and ensure that the power needs of users are met in the isolated grid mode.

[0089] As a preferred implementation, in step S2, the power flow data includes voltage, current amplitude phase angle, active power generation and reactive power generation.

[0090] As a preferred implementation, in step S3, the process of calculating the index weight of the coordination evaluation index using the hierarchical analysis method includes:

[0091] Construct a judgment matrix:

[0092]

[0093] Hierarchical single sorting:

[0094] ZW=λmaxW

[0095]

[0096] Consistency check:

[0097] LW=λW

[0098]

[0099]

[0100] In the formula, Z represents the judgment matrix, a ij represents the element in the i-th row and j-th column of the judgment matrix, W represents the eigenvector, and λ max represents the largest characteristic root of Z, Wi represents the weight, L is the n-order judgment matrix, λ represents the characteristic root, CI represents the consistency index, RI represents the consistency expectation value, and CR represents the index of whether the judgment matrix has satisfactory consistency.

[0101] Example 2

[0102] Aiming at the coordination evaluation problem of distribution networks with a high proportion of distributed power sources, this paper proposes a distribution network coordination evaluation method based on graph convolutional neural network, which combines graph convolutional network (GCN) with analytic hierarchy process (AHP) for comprehensive evaluation. Figure 5 As shown in the figure, the whole process is based on Python and implemented in PyCharm. First, GCN is used to process the collected power flow data, including voltage, current amplitude phase angle, active and reactive power generation, etc. After the convolution operation of three convolutional layers, GCN effectively extracts the relationship characteristics between nodes and the topological structure information of the network to form a more comprehensive feature matrix with the same dimension as the input features. A pooling layer is set after the three convolutional layers to reduce memory consumption and overfitting risks and enhance translation invariance. Next, two hidden layers are added. In the first hidden layer, Python is used to perform indicator operations on the new data after aggregating neighbor features to give a score. In the second hidden layer, the indicator weights obtained by AHP are used to perform weighted average operations on the scores to obtain the final score. Finally, the model is output with the final score as the output layer.

[0103] After the model was built, the model was trained with 6646 sets of training set sample data. During the forward propagation process, the activation function selected the ReLU function; during the back propagation process, the gradient was calculated based on the loss between the result of the forward propagation calculation and the true label, and the Adam optimization algorithm was used to update the weights. This algorithm combines momentum and adaptive learning rate, and has better performance under different gradient update step sizes. To optimize the model, K-fold cross validation (K=8) was used to improve the generalization ability, the initial learning rate was 0.001, and it was decayed by 0.4 every 50 epochs, and the Adam optimizer was combined for training; to prevent overfitting, L2 regularization was added, and the weight decay coefficient was 1e-5. An early stopping mechanism was introduced to terminate the training early when the validation set loss did not improve within 10 epochs.

[0104] Example 3

[0105] See also Figure 6 , Figure 6 A schematic structural block diagram of an electronic device provided in an embodiment of the present application.

[0106] An electronic device includes a memory 101, a processor 102 and a communication interface 103, wherein the memory 101, the processor 102 and the communication interface 103 are electrically connected to each other directly or indirectly to realize data transmission or interaction. For example, these components can be electrically connected to each other through one or more communication buses or signal lines. The memory 101 can be used to store software programs and modules, and the processor 102 executes various functional applications and data processing by executing the software programs and modules stored in the memory 101. The communication interface 103 can be used to communicate signaling or data with other node devices.

[0107] Among them, the memory 101 can be, but is not limited to, a random access memory (RAM), a read only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable read-only memory (EEPROM), etc.

[0108] The processor 102 may be an integrated circuit chip with signal processing capability. The processor 102 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0109] It can be understood that the structure shown in the figure is only for illustration, and a distribution network coordination evaluation method based on graph convolutional neural network may also include more or fewer components than shown in the figure, or have a different configuration than shown in the figure. Each component shown in the figure can be implemented by hardware, software or a combination thereof.

[0110] In the embodiments provided in the present application, it should be understood that the disclosed method can also be implemented in other ways. The embodiments described above are merely schematic, for example, the flowchart or block diagram in the accompanying drawings shows the possible implementation architecture, functions and operations of the method and computer program product according to multiple embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a module, a program segment or a part of a code, and the module, a program segment or a part of a code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or the flowchart, and the combination of the boxes in the block diagram and / or the flowchart can be implemented with a dedicated hardware-based system that performs a specified function or action, or can be implemented with a combination of dedicated hardware and computer instructions.

[0111] In addition, the functional modules in the various embodiments of the present application may be integrated together to form an independent part, or each module may exist separately, or two or more modules may be integrated to form an independent part.

[0112] If the functions are implemented in the form of software function modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable 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 methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0113] The above description is only the preferred embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

[0114] It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, and that the present application can be implemented in other specific forms without departing from the spirit or essential features of the present application. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the present application is defined by the appended claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims be included in the present application. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A distribution network coordination evaluation method based on graph convolutional neural network, characterized in that: The following steps are involved: S1. Conduct coordination evaluation on the distribution network containing a high proportion of distributed generation to obtain coordination evaluation indicators of the distribution network; S2, obtaining the power flow data of the distribution network, and processing the power flow data through a graph convolutional neural network to extract the relationship characteristics between nodes and the topological structure information of the network; S3. Calculate the index weights of the coordination evaluation index using the hierarchical analysis method and add them as hidden layers to the graph convolutional neural network to construct an improved graph convolutional neural network. S4, inputting the relationship characteristics between nodes and the topological structure information of the network into the improved graph convolutional neural network model to obtain the evaluation index of the distribution network; S5. Perform a coordination evaluation of the distribution network according to the evaluation indexes and index weights of the distribution network to obtain a coordination evaluation result of the distribution network.

2. A distribution network coordination evaluation method based on graph convolutional neural network as claimed in claim 1, characterized in that: In step S1, the coordination evaluation index of the distribution network includes a three-layer index system established by the power source and power grid coordination evaluation index, the power grid and user coordination evaluation index and the isolated network operation capability evaluation index.

3. A distribution network coordination evaluation method based on graph convolutional neural network as claimed in claim 2, characterized in that: The evaluation indicators for coordination between power sources and power grids include the matching degree between distributed power sources and power grids, power penetration rate, the proportion of distributed power sources in total power supply, power output prediction accuracy, and the deviation between prediction and actual output. The calculation formulas include: p=|ppred-pact| In the formula, A represents power matching, P represents the output power of distributed power generation, Q represents the grid load, B represents the penetration rate of distributed power generation, C represents the installed capacity of distributed power generation, H represents the total load peak, X represents the proportion of distributed power generation in the total power supply, and C represents the total load peak. 总 represents the total installed capacity of the power system, Accuracy represents the relative error, MAE represents the average absolute error between the predicted value and the actual value, represents the average actual value, n represents the number of samples, y i Indicates the actual value, represents the predicted value, p represents the deviation between the predicted value and the actual value, and p pred It represents the predicted output of distributed generation in a certain period of time, p act Indicates the actual output of distributed generation in a certain period of time.

4. A distribution network coordination evaluation method based on graph convolutional neural network as claimed in claim 2, characterized in that: The evaluation indicators of power grid and user coordination include user demand response, user response rate during peak and valley periods, proportion of users participating in peak and valley dispatching, demand side management effect and power saving of demand side management projects; among them, the calculation formula of user response during peak and valley periods is: In the formula, S represents the user response rate, V' represents the reduced power consumption during the peak and valley periods, and V represents the original power consumption during the peak and valley periods; The calculation formula for electricity saving of demand side management project is: U=U1-U2 In the formula, U represents the amount of electricity saved, U1 represents the amount of electricity not consumed, and U2 represents the amount of electricity consumed.

5. A distribution network coordination evaluation method based on graph convolutional neural network as claimed in claim 2, characterized in that: The evaluation indicators of isolated grid operation capability include isolated grid stable operation index, isolated grid duration and maximum isolated grid operation time, voltage stability during isolated grid period and voltage fluctuation range during isolated grid period; Among them, the calculation formula for the voltage fluctuation range during the isolated grid period is: In the formula, S 波 Indicates the voltage fluctuation range during isolated grid, V max Indicates the maximum voltage monitored, V min Indicates the minimum voltage detected, V rated Indicates the reference voltage of the system.

6. A distribution network coordination evaluation method based on graph convolutional neural network as claimed in claim 1, characterized in that: In step S2, the power flow data includes voltage, current amplitude, phase angle, active power generation and reactive power generation.

7. A distribution network coordination evaluation method based on graph convolutional neural network as claimed in claim 1, characterized in that: In step S3, the calculation process of calculating the index weight of the coordination evaluation index using the hierarchical analysis method includes: Construct a judgment matrix: Hierarchical single sorting: ZW=λmaxW Consistency check: LW=λW In the formula, Z represents the judgment matrix, a ij represents the element in the i-th row and j-th column of the judgment matrix, W represents the eigenvector, and λ max represents the largest characteristic root of Z, W i represents the weight, L is the n-order judgment matrix, λ represents the characteristic root, CI represents the consistency index, RI represents the consistency expectation value, and CR represents the index of whether the judgment matrix has satisfactory consistency.

8. An electronic device, characterized in that: include: A memory for storing one or more programs; processor; When the one or more programs are executed by the processor, a distribution network coordination evaluation method based on a graph convolutional neural network as described in any one of claims 1 to 7 is implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements a distribution network coordination evaluation method based on a graph convolutional neural network as described in any one of claims 1 to 7.

Images