Energy consumption management method and system based on cloud computing

Abstract

The invention provides an energy consumption management method and system based on cloud computing. The energy consumption management method based on cloud computing comprises the following steps: acquiring historical operation data and real-time operation data of a target power grid through a distributed cloud; analyzing the historical operation data by using an energy consumption management model to obtain a first topological structure of the target power grid, the topological structure being a graph data structure; determining an objective function of the first topological structure according to the energy consumption management request, and optimizing the first topological structure according to the objective function and an energy consumption management model to obtain a second topological structure; executing control processing on the target power grid through the power grid dispatching system according to the second topological structure; wherein the energy consumption management model is obtained through training. The method has the beneficial effects that the energy consumption management efficiency is improved, and the system resource consumption of energy consumption management is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of electric energy management, and in particular to an energy consumption management method and system based on cloud computing. Background Art

[0002] The increasingly complex distribution network includes different types of distribution stations, energy storage equipment and load terminals, making the topology of the distribution network complex.

[0003] Graph neural networks are usually used to process topological relationships. However, traditional methods cannot handle the graph structure of complex distribution networks, making energy consumption management and scheduling efficiency unable to meet the needs of existing distribution networks. Summary of the invention

[0004] The main purpose of the embodiments of the present invention is to provide an energy consumption management method and system based on cloud computing, which improves the efficiency of energy consumption management and reduces the system resource consumption of energy consumption management.

[0005] One aspect of the present invention provides an energy consumption management method based on cloud computing, comprising: Obtain historical and real-time operation data of the target power grid through the distributed cloud; Analyzing the historical operation data using an energy consumption management model to obtain a first topology structure of the target power grid, wherein the topology structure is a graph data structure; Determine the objective function of the first topology structure according to the energy consumption management request, and optimize the first topology structure according to the objective function and the energy consumption management model to obtain a second topology structure; Executing control processing on the target power grid through a power grid dispatching system according to the second topology structure; The training steps of the energy consumption management model include: Acquire the historical operation data, the historical power grid diagram structure and the real-time operation data, pre-process the historical operation data, and then use a distributed computing framework and a graph neural network to perform distributed training to obtain training results; According to the training results, adjusting the training parameters of the graph neural network and re-training for a preset number of times to obtain the energy consumption management model; Dynamically input the real-time operation data into the energy consumption management model and perform reasoning to obtain an updated power grid diagram structure; The energy consumption management model is optimized according to the updated power grid diagram structure.

[0006] According to the energy consumption management method based on cloud computing, the distributed cloud collects the grid equipment operation data of the target grid, cleans, normalizes and extracts features of the grid equipment operation data to obtain a grid representation; the grid representation is converted into a graph data structure, wherein the graph data structure includes node attributes and edge attributes, the node attributes are the attributes of the grid equipment, and the edge attributes are the association relationship attributes of the grid equipment.

[0007] According to the energy consumption management method based on cloud computing, a distributed computing framework and a graph neural network are used for distributed training, including: Deploying the distributed computing framework via the cloud, and using the graph data structure as input to the graph neural network; The graph data structure is sampled and processed in blocks through the distributed computing framework to obtain sampling and block results; The graph data structure is batched and gradient descended according to the sampling and block results; One of the Adam and SGD optimizers is used for cross-validation and selection of the best parameters.

[0008] According to the energy consumption management method based on cloud computing, the method further includes: The distributed computing framework is one of TensorFlow and PyTorch, and the graph sampling and block processing of the graph data structure adopts one of GraphSAGE and Cluster-GCN.

[0009] According to the cloud computing-based energy consumption management method, the training result, adjusting the training parameters of the graph neural network and re-training a preset number of times to obtain the energy consumption management model, further includes: When training the energy consumption management model, one of a GPU and a TPU is used for reasoning acceleration, and TensorRT is used to perform model compression processing on the energy consumption management model; The energy consumption management model is trained according to the objective function and the evaluation index, and the model hyperparameters are adjusted according to the training results each time.

[0010] According to the energy consumption management method based on cloud computing, real-time operation data is dynamically input into the energy consumption management model and reasoned to obtain an updated power grid diagram structure, including: Deploy the energy consumption management model via the cloud; The real-time operation data is converted into stream data through Flink, and the historical power grid diagram structure is updated according to the stream data to obtain an updated power grid diagram structure; The updated power grid diagram structure is dynamically input into the energy consumption management model for reasoning.

[0011] According to the energy consumption management method based on cloud computing, the energy consumption management model performs reasoning, including: Obtain the node attributes and edge attributes of the historical power grid graph structure, and update the node attributes of the historical power grid graph structure according to the updated power grid graph structure; wherein the historical power grid graph structure is represented as , V is the node set of the historical power grid diagram, representing the equipment in the power grid, and the matrix of node attributes is , d is the dimension of node attributes, and the matrix of edge attributes is , m is the dimension of edge attributes; the node attributes of the historical power grid graph structure are updated to , the updated node attributes are , where f is the process function for updating node attributes in the power grid graph structure; Graph convolution and attention calculation are used to generate node embedding according to the updated node attributes, and the edge attributes are updated. The operation of graph convolution is: in, H ( l ) is the l The node feature matrix of the layer, A ~= A + I is the adjacency matrix with self-connection added, yes The degree matrix of W ( l ) is the l The learnable weight matrix of the layer, σ is the activation function, and the attention mechanism calculates the attention weights between nodes through the following formula: in, Yes Yes Node i and nodes j The attention score between a is a learnable weight vector, and ||| represents the connection operation of the vector; The node embedding Z can be obtained by weighted summation is the normalized attention weight; The edge attributes based on node embedding are , Function for updating edge attributes; Generate an optimization plan based on the updated node attributes, edge attributes and objective function, where the objective function includes one of minimizing losses, maximizing power supply reliability and minimizing energy consumption, including: is the objective function, is the constraint condition of power balance constraint; The optimization scheme is used to represent the start and stop of the power grid equipment and the transmission lines between the power grid equipment in the target power grid, specifically: .

[0012] According to the energy consumption management method based on cloud computing, the objective function also includes: The objective function of loss minimization is: in is the resistance of edge (i,j), is the square value of the power flow at edge (i, j); The objective function of the highest reliability of power supply is: in Is a node i reliability; The objective function of minimizing energy consumption is: in Is a node i The unit energy cost, Is a node i of power output.

[0013] Another aspect of the present invention further discloses an energy consumption management system based on cloud computing, which is characterized by comprising: The first module is used to obtain historical operation data and real-time operation data of the target power grid through the distributed cloud; The second module is used to analyze the historical operation data using an energy consumption management model to obtain a first topological structure of the target power grid, wherein the topological structure is a graph data structure; A third module is used to determine the objective function of the first topology structure according to the energy consumption management request, and optimize the first topology structure according to the objective function and the management model to obtain a second topology structure; A fourth module, configured to perform control processing on the target power grid according to the second topology structure; The fifth module is used to obtain the historical operation data, the historical power grid diagram structure and the real-time operation data, and after pre-processing the historical operation data, use the distributed computing framework and the graph neural network to perform distributed training to obtain the training results; The sixth module is used to adjust the training parameters of the graph neural network according to the training results and re-train a preset number of times to obtain the energy consumption management model; The seventh module is used to dynamically input the real-time operation data into the energy consumption management model and perform reasoning to obtain an updated power grid diagram structure; The eighth module is used to optimize the energy consumption management model according to the updated power grid diagram structure.

[0014] The beneficial effects of the present invention are as follows: through cloud computing (cloud deployment), the training and reasoning of graph neural networks in a cloud computing environment can efficiently handle power grid topology optimization problems, thereby improving the efficiency of energy consumption management; at the same time, power grid data is collected in real time in the cloud, and GNN is used to generate the optimal topology reconstruction solution, which can adapt to large power grids (such as the State Grid) containing millions of nodes and edges, and has lower complexity than existing technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The above and / or additional aspects and advantages of the present invention will become apparent and easily understood from the description of the embodiments in conjunction with the following drawings, in which: Figure 1 Schematic diagram of system composition of an embodiment of the present invention.

[0016] Figure 2 It is a schematic diagram of the flow of the energy consumption management method for cloud computing according to an embodiment of the present invention.

[0017] Figure 3 It is a schematic diagram of the distributed training process of an embodiment of the present invention.

[0018] Figure 4 It is a schematic diagram of the energy consumption management model training process of an embodiment of the present invention.

[0019] Figure 5 It is a schematic diagram of the reasoning process of an energy consumption management model according to an embodiment of the present invention.

[0020] Figure 6 It is another schematic diagram of the energy consumption management model reasoning process of an embodiment of the present invention.

[0021] Figure 7 It is a system block diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0022] The embodiments of the present invention are described in detail below, and examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. In the subsequent description, the use of suffixes such as "module", "component" or "unit" used to represent elements is only for the purpose of facilitating the description of the present invention, and has no specific meaning in itself. Therefore, "module", "component" or "unit" can be used in a mixed manner. "First", "second" and the like are only used for the purpose of distinguishing technical features, and cannot be understood as indicating or implying relative importance or implicitly indicating the number of indicated technical features or implicitly indicating the order of the indicated technical features. In this subsequent description, the continuous numbering of the method steps is for the convenience of review and understanding. In combination with the overall technical solution of the present invention and the logical relationship between the various steps, adjusting the implementation order between the steps will not affect the technical effect achieved by the technical solution of the present invention. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and cannot be understood as limiting the present invention.

[0023] refer to Figure 1 ,in Figure 1 : is a schematic diagram of the system composition of an embodiment of the present invention. It includes a target power grid 100, a cloud server 200 and a power grid dispatching system 300. Among them, the cloud server 200 is the cloud, which is set according to the complexity of the target power grid. The cloud server 200 is used to obtain historical operation data, historical power grid graph structure and real-time operation data. After pre-processing the historical operation data, a distributed computing framework and a graph neural network are used for distributed training to obtain training results; according to the training results, the training parameters of the graph neural network are adjusted and re-trained for a preset number of times to obtain an energy consumption management model; the real-time operation data is dynamically input into the energy consumption management model and reasoned to obtain an updated power grid graph structure; the energy consumption management model is optimized according to the updated power grid graph structure.

[0024] The power grid dispatching system 300 is used to receive the optimization plan and perform topology optimization on the target power grid 100.

[0025] refer to Figure 2 ,in Figure 2 1 is a schematic diagram of a method for managing energy consumption of cloud computing according to an embodiment of the present invention, which includes but is not limited to steps S100 to S800, wherein S100 to S400 are a method for managing energy consumption, including: S100,obtains historical and real-time operation data of the target power grid through the,distributed cloud.

[0026] In some embodiments, the distributed cloud collects the grid equipment operation data of the target grid, cleans, normalizes and extracts features of the grid equipment operation data to obtain a grid representation; the grid representation is converted into a graph data structure, wherein the graph data structure includes node attributes and edge attributes, the node attributes are the attributes of the grid equipment, and the edge attributes are the association relationship attributes of the grid equipment.

[0027] S200, analyzing historical operation data using an energy consumption management model to obtain a first topology structure of a target power grid, wherein the topology structure is a graph data structure.

[0028] S300, determining an objective function of a first topology structure according to an energy consumption management request, optimizing the first topology structure according to the objective function and an energy consumption management model, and obtaining a second topology structure.

[0029] In some embodiments, it can be understood that the first topology is the topology of the target power grid in history, and the second topology is the topology updated according to real-time operation data. For example, when any distribution device or load device fails or is actively shut down, the topology is updated.

[0030] S400: Execute control processing on the target power grid through the power grid dispatching system according to the second topology structure.

[0031] In some embodiments, it can be understood that when the power grid dispatching system changes the power grid from a first topology to a second topology, any nodes and edges in the power grid that have changed will be modified accordingly, and a dynamic graph neural network (Dynamic GNN) can be used to handle the dynamic changes in the power grid topology.

[0032] Among them, S500~S800 is the training process of the energy consumption management model, including: S500, obtain historical operation data, historical power grid diagram structure and real-time operation data, pre-process the historical operation data, use distributed computing framework and graph neural network to perform distributed training, and obtain training results.

[0033] refer to Figure 3 The distributed training process diagram shown includes but is not limited to steps S510 to S540: S510, deploys a distributed computing framework through the cloud, using the graph data structure as the input of the graph neural network.

[0034] S520, sampling and partitioning the graph data structure through a distributed computing framework to obtain sampling and partitioning results.

[0035] S530, batch-dividing and gradient-descent the graph data structure according to the sampling and block results.

[0036] S540, using one of the Adam and SGD optimizers for cross-validation and selection of optimal parameters.

[0037] In some embodiments, the distributed computing framework is one of TensorFlow and PyTorch, and the graph data structure is sampled and partitioned using one of GraphSAGE and Cluster-GCN.

[0038] It is understandable that graph sampling using GraphSAGE or Cluster-GCN is used to reduce the amount of computation.

[0039] In some embodiments, reference Figure 4 The energy consumption management model training process diagram shown in FIG. 1 includes but is not limited to steps S550 to S560: S600: According to the training results, the training parameters of the graph neural network are adjusted and the training is repeated for a preset number of times to obtain an energy consumption management model.

[0040] S650 uses GPU or TPU for inference acceleration when training the energy management model, and uses TensorRT to compress the energy management model. S660, the energy consumption management model is trained according to the objective function and the evaluation index, and the model hyperparameters are adjusted according to the training results during each training.

[0041] In some embodiments, the evaluation indicators include accuracy, recall, F1 score, and can also be optimized according to the objective function of the energy consumption management task (such as minimizing energy consumption).

[0042] S700,dynamically inputs the real-time operation data into the energy consumption management,model and performs reasoning to obtain an updated grid diagram structure.

[0043] In some embodiments, reference Figure 5 The energy consumption management model reasoning process diagram shown in the figure includes but is not limited to steps S710 to S730: S710, deploys energy consumption management model through the cloud; S720, converting the real-time operation data into stream data through Flink, and updating the historical power grid graph structure according to the stream data to obtain an updated power grid graph structure; S730, dynamically input the updated power grid diagram structure into the energy consumption management model for reasoning.

[0044] The objective function includes one of minimizing losses, maximizing power supply reliability, and minimizing energy consumption, and the optimization scheme is used to characterize the start and stop of power grid equipment and transmission lines between power grid equipment in the target power grid.

[0045] In some embodiments, the energy consumption management model performs reasoning, including: Obtain the node attributes and edge attributes of the historical power grid graph structure, and update the node attributes of the historical power grid graph structure according to the updated power grid graph structure; wherein the historical power grid graph structure is represented as , V is the node set of the historical power grid diagram, representing the equipment in the power grid, and the matrix of node attributes is , d is the dimension of node attributes, and the matrix of edge attributes is , m is the dimension of edge attributes; the node attributes of the historical power grid graph structure are updated to , the updated node attributes are , where f is the process function for updating node attributes in the power grid graph structure; Graph convolution and attention calculation are used to generate node embedding according to the updated node attributes, and the edge attributes are updated. The operation of graph convolution is: in, H ( l ) is the l The node feature matrix of the layer, A ~= A + I is the adjacency matrix with self-connection added, yes The degree matrix of W ( l ) is the l The learnable weight matrix of the layer, σ is the activation function, and the attention mechanism calculates the attention weights between nodes through the following formula: in, Yes Yes Node i and nodes j The attention score between a is a learnable weight vector, and ||| represents the connection operation of the vector; The node embedding Z can be obtained by weighted summation is the normalized attention weight; The edge attributes based on node embedding are , Function for updating edge attributes; Generate an optimization plan based on the updated node attributes, edge attributes and objective function, where the objective function includes one of minimizing losses, maximizing power supply reliability and minimizing energy consumption, including: is the objective function, is the constraint condition of power balance constraint; The optimization scheme is used to represent the start and stop of the power grid equipment and the transmission lines between the power grid equipment in the target power grid, specifically: .

[0046] In some embodiments, the objective function is as follows: The objective function of loss minimization is: in is the resistance of edge (i,j), is the square value of the power flow at edge (i, j); The objective function of the highest reliability of power supply is: in Is a node i reliability; The objective function of minimizing energy consumption is: in Is a node i The unit energy cost, Is a node i of power output.

[0047] It is understandable that the objective function may be selected in other ways according to the power grid energy consumption management.

[0048] In some embodiments, referring to another energy consumption management model reasoning process diagram, it includes but is not limited to steps S731-S733: S731, obtaining node attributes and edge attributes of the historical power grid graph structure, and updating the node attributes of the historical power grid graph structure according to the updated power grid graph structure; S732, using graph convolution and attention calculation to generate node embedding according to the updated node attributes, and update edge attributes; S733, generating an optimization solution according to the updated node attributes, edge attributes and objective function.

[0049] In some embodiments, the cloud deployment uses the cloud service AWS SageMaker to deploy the GNN model (energy consumption management model, i.e., GNN model), wherein the cloud service AWS SageMaker uses a RESTful API interface.

[0050] In some embodiments, hardware such as GPU / TPU is used to accelerate reasoning, and ONNX or TensorRT is used for model optimization and compression.

[0051] In some embodiments, real-time reasoning includes stream data processing: including using Kafka or Flink to process real-time data streams, updating the power grid graph structure, and dynamically inputting the GNN model for reasoning.

[0052] In some embodiments, node / edge feature update: update node and edge features based on real-time data, perform graph convolution or attention calculation, generate node embedding, and generate power grid optimization solutions (such as switch operation recommendations).

[0053] S800, optimizes the energy consumption management model according to the updated grid diagram structure to reduce data transmission.

[0054] In some embodiments, the above-obtained inference results can be fed back to the power grid dispatching system to adjust the model parameters according to the actual operation effect.

[0055] In some embodiments, edge computing is used to transmit data to the cloud, power grid dispatching system, and power grid equipment to improve transmission efficiency.

[0056] In some embodiments, the specific process of model training is as follows (1) Test dataset preparation After the training is completed, we divide the data of the IEEE 118-bus system into a training set and a test set: Training set: 80% of the node data, used for training the model.

[0057] Test set: 20% of the node data, used to evaluate model performance.

[0058] Assuming that we have completed model training and made predictions on the test set, the following are the detailed result data of the test set.

[0059] (2) Test result data: The following is the comparison data between the model's prediction results and the true values ​​on the test set, refer to the data table in Table 1: Table 1 Test result data table In some implementations, mean absolute error (MAE) is also used for evaluation, as shown in Table 2 for specific error distribution.

[0060] It can be determined that the power grid optimization solution obtained by adopting the technical solution of the embodiment of the present invention can efficiently handle the power grid topology optimization problem in a shorter time, thereby improving the efficiency of energy consumption management.

[0061] refer to Figure 7 ,in, Figure 7 7 is a system block diagram of an embodiment of the present invention, which includes a first module 710, which is used to obtain historical operation data and real-time operation data of a target power grid through a distributed cloud; a second module 720, which is used to analyze the historical operation data using an energy consumption management model to obtain a first topology structure of the target power grid, wherein the topology structure is a graph data structure; a third module 730, which is used to determine the objective function of the first topology structure according to the energy consumption management request, and optimize the first topology structure according to the objective function and the management model to obtain a second topology structure; a fourth module 740, which is used to perform control processing on the target power grid according to the second topology structure; a fifth module 750, which is used to obtain historical operation data, a historical power grid graph structure and real-time operation data, and after pre-processing the historical operation data, use a distributed computing framework and a graph neural network for distributed training to obtain training results; a sixth module 760, which is used to adjust the training parameters of the graph neural network according to the training results and re-train for a preset number of times to obtain an energy consumption management model; a seventh module 770, which is used to dynamically input the real-time operation data into the energy consumption management model and perform reasoning to obtain an updated power grid graph structure; an eighth module 780, which is used to optimize the energy consumption management model according to the updated power grid graph structure.

[0062] An embodiment of the present invention further provides an electronic device, the electronic device comprising a processor and a memory; The memory stores a program; The processor executes a program to execute the aforementioned cloud computing-based energy consumption management method; the electronic device has the function of carrying and running the cloud computing-based energy consumption management software system provided by an embodiment of the present invention, for example, a personal computer, a minicomputer, a main frame, a workstation, a network or distributed computing environment, a separate or integrated computer platform, or communicating with a charged particle tool or other imaging device, etc.

[0063] An embodiment of the present invention further provides a computer-readable storage medium, wherein the storage medium stores a program, and the program is executed by a processor to implement the energy consumption management method based on cloud computing as described above.

[0064] In some selectable embodiments, the function / operation mentioned in the block diagram may not occur in the order mentioned in the operation diagram. For example, depending on the function / operation involved, the two boxes shown in succession can actually be executed substantially simultaneously or the boxes can sometimes be executed in reverse order. In addition, the embodiment presented and described in the flow chart of the present invention is provided by way of example, for the purpose of providing a more comprehensive understanding of technology. The disclosed method is not limited to the operation and logic flow presented herein. Selectable embodiments are expected, wherein the order of various operations is changed and the sub-operation of a part for which is described as a larger operation is performed independently.

[0065] The embodiment of the present invention also discloses a computer program product or a computer program, which includes a computer instruction stored in a computer-readable storage medium. A processor of a computer device can read the computer instruction from the computer-readable storage medium, and the processor executes the computer instruction, so that the computer device executes the aforementioned cloud computing-based energy consumption management method.

[0066] In addition, although the present invention is described in the context of functional modules, it should be understood that, unless otherwise specified, one or more of the functions and / or features described may be integrated into a single physical device and / or software module, or one or more functions and / or features may be implemented in separate physical devices or software modules. It is also understood that a detailed discussion of the actual implementation of each module is unnecessary for understanding the present invention. More specifically, in view of the properties, functions, and internal relationships of the various functional modules in the device disclosed herein, the actual implementation of the module will be understood within the conventional skills of the engineer. Therefore, those skilled in the art can implement the present invention set forth in the claims without excessive experimentation using ordinary techniques. It is also understood that the specific concepts disclosed are merely illustrative and are not intended to limit the scope of the present invention, which is determined by the full scope of the appended claims and their equivalents.

[0067] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0068] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0069] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0070] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0071] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "examples", "specific examples", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0072] Although the embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the claims and their equivalents.

[0073] The above is a specific description of the preferred implementation of the present invention, but the present invention is not limited to the described embodiments. Those skilled in the art may make various equivalent modifications or substitutions without violating the spirit of the present invention. These equivalent modifications or substitutions are all included in the scope defined by the claims of this application.

Claims

1. A cloud computing-based energy consumption management method, characterized in that: include: Obtain historical and real-time operation data of the target power grid through the distributed cloud; Analyzing the historical operation data using an energy consumption management model to obtain a first topology structure of the target power grid, wherein the topology structure is a graph data structure; Determine the objective function of the first topology structure according to the energy consumption management request, optimize the first topology structure according to the objective function and the energy consumption management model to obtain a second topology structure, the first topology structure is the topology structure of the target power grid in history, and the second topology structure is the topology structure updated according to the real-time operation data, wherein the objective function is one of minimizing loss, maximizing power supply reliability, and minimizing energy consumption; Executing control processing on the target power grid through a power grid dispatching system according to the second topology structure; The training steps of the energy consumption management model include: Acquire the historical operation data, the historical power grid diagram structure and the real-time operation data, pre-process the historical operation data, and then use a distributed computing framework and a graph neural network to perform distributed training to obtain training results; According to the training results, adjusting the training parameters of the graph neural network and re-training for a preset number of times to obtain the energy consumption management model; Dynamically input the real-time operation data into the energy consumption management model and perform reasoning to obtain an updated power grid diagram structure; The energy consumption management model is optimized according to the updated power grid diagram structure.

2. The cloud computing-based energy consumption management method according to claim 1, characterized in that: The distributed cloud collects the grid equipment operation data of the target grid, performs data cleaning, normalization and feature extraction on the grid equipment operation data to obtain a grid representation; the grid representation is converted into a graph data structure, wherein the graph data structure includes node attributes and edge attributes, the node attributes are the attributes of the grid equipment, and the edge attributes are the association relationship attributes of the grid equipment.

3. The cloud computing-based energy consumption management method according to claim 2, characterized in that: The distributed training using a distributed computing framework and a graph neural network includes: Deploying the distributed computing framework via the cloud, and using the graph data structure as input to the graph neural network; The graph data structure is sampled and processed in blocks through the distributed computing framework to obtain sampling and block results; The graph data structure is batched and gradient descended according to the sampling and block results; One of the Adam and SGD optimizers is used for cross-validation and selection of the best parameters.

4. The cloud computing-based energy consumption management method according to claim 3, characterized in that: The method further comprises: The distributed computing framework is one of TensorFlow and PyTorch, and the graph sampling and block processing of the graph data structure adopts one of GraphSAGE and Cluster-GCN.

5. The cloud computing-based energy consumption management method according to claim 3, characterized in that: The step of adjusting the training parameters of the graph neural network according to the training results and re-training for a preset number of times to obtain the energy consumption management model further includes: When training the energy consumption management model, one of a GPU and a TPU is used for reasoning acceleration, and TensorRT is used to perform model compression processing on the energy consumption management model; The energy consumption management model is trained according to the objective function and the evaluation index, and the model hyperparameters are adjusted according to the training results each time.

6. The energy consumption management method based on cloud computing according to claim 3 is characterized in that: The real-time operation data is dynamically input into the energy consumption management model and reasoned to obtain an updated power grid diagram structure, including: Deploy the energy consumption management model via the cloud; The real-time operation data is converted into stream data through Flink, and the historical power grid diagram structure is updated according to the stream data to obtain an updated power grid diagram structure; The updated power grid diagram structure is dynamically input into the energy consumption management model for reasoning.

7. The cloud computing-based energy consumption management method according to claim 6, characterized in that: The energy consumption management model performs reasoning, including: Obtain the node attributes and edge attributes of the historical power grid graph structure, and update the node attributes of the historical power grid graph structure according to the updated power grid graph structure; wherein the historical power grid graph structure is represented as , V is the node set of the historical power grid diagram, representing the equipment in the power grid, and the matrix of node attributes is , d is the dimension of node attributes, and the matrix of edge attributes is , m is the dimension of edge attributes; the node attributes of the historical power grid graph structure are updated to , the updated node attributes are , where f is the process function for updating node attributes in the power grid graph structure; Graph convolution and attention calculation are used to generate node embedding according to the updated node attributes, and the edge attributes are updated. The operation of graph convolution is: in, H ( l ) is the l The node feature matrix of the layer, A ~= A + I is the adjacency matrix with self-connection added, yes The degree matrix of W ( l ) is the l The learnable weight matrix of the layer, σ is the activation function, and the attention mechanism calculates the attention weights between nodes through the following formula: in, Yes Yes Node i and nodes j The attention score between a is a learnable weight vector, and ||| represents the connection operation of the vector; The node embedding Z can be obtained by weighted summation is the normalized attention weight; The edge attributes based on node embedding are , Function for updating edge attributes; Generate an optimization plan based on the updated node attributes, edge attributes and objective function, where the objective function includes one of minimizing losses, maximizing power supply reliability and minimizing energy consumption, including: is the objective function, is the constraint condition of power balance constraint; The optimization scheme is used to represent the start and stop of the power grid equipment and the transmission lines between the power grid equipment in the target power grid, specifically: 。 8. The cloud computing-based energy consumption management method according to claim 7, characterized in that: The objective function also includes: The objective function of loss minimization is: in is the resistance of edge (i,j), is the square value of the power flow at edge (i, j); The objective function of the highest reliability of power supply is: in Is a node i reliability; The objective function of minimizing energy consumption is: in Is a node i The unit energy cost, Is a node i of power output.

9. An energy consumption management system based on cloud computing according to any one of the methods of claims 1-8, characterized in that: include: The first module is used to obtain historical operation data and real-time operation data of the target power grid through the distributed cloud; The second module is used to analyze the historical operation data using an energy consumption management model to obtain a first topological structure of the target power grid, wherein the topological structure is a graph data structure; The third module is used to determine the objective function of the first topology structure according to the energy consumption management request, optimize the first topology structure according to the objective function and the energy consumption management model to obtain a second topology structure, the first topology structure is the topology structure of the target power grid in history, and the second topology structure is the topology structure updated according to the real-time operation data, wherein the objective function is one of minimizing losses, maximizing power supply reliability and minimizing energy consumption; A fourth module, configured to perform control processing on the target power grid according to the second topology structure; The fifth module is used to obtain the historical operation data, the historical power grid diagram structure and the real-time operation data, and after pre-processing the historical operation data, use the distributed computing framework and the graph neural network to perform distributed training to obtain the training results; The sixth module is used to adjust the training parameters of the graph neural network according to the training results and re-train a preset number of times to obtain the energy consumption management model; The seventh module is used to dynamically input the real-time operation data into the energy consumption management model and perform reasoning to obtain an updated power grid diagram structure; The eighth module is used to optimize the energy consumption management model according to the updated power grid diagram structure.

Images