Transform consistency optimization method and device for economic dispatching of super-large-scale multi-region interconnected power system, computer equipment and medium

Abstract

The invention provides a Transform consistency optimization method and device for economic dispatching of a super-large-scale multi-region interconnected power system, computer equipment and a medium, and the method comprises the steps: constructing a provincial-level economic dispatching optimization model and a municipal-level economic dispatching optimization model, taking power generation consumption and carbon emission as optimization targets, and considering upper and lower limits of output of a unit and constraint of an operation prohibition region; designing a provincial-level and municipal-level Transform model architecture, wherein the provincial-level and municipal-level models are respectively provided with encoders and decoders with the same layer number; and processing the two-layer economic dispatching optimization model by adopting a consistency optimization algorithm to obtain a power generation consumption matrix, inputting the power generation consumption matrix into a Transform model to predict the optimal power generation amount, and processing the municipal model in the same way. The method provided by the invention can solve the problems of low robustness and poor privacy of the system when the scale of the interconnected power system is expanded and the number of the intelligent agents is increased, and improves the overall solution speed and the solution result reliability of the system.

Description

Technical Field

[0001] The present invention belongs to the fields of optimization of ultra-large-scale power generation, new energy, artificial intelligence and large models, and relates to an optimization method for economic dispatch of ultra-large-scale multi-region interconnected power systems, which is suitable for grid energy dispatch in power systems. Background Art

[0002] Traditional economic dispatch models generally adopt a centralized energy dispatch method. The energy management center obtains all necessary information about all operating equipment in the entire system, analyzes and processes the information of the entire system based on the required power generation value, and feeds back the optimized information to each unit of the system in the form of power generation instructions. This method has problems such as poor robustness, insufficient scalability, slow speed and poor privacy.

[0003] As the scale of systems continues to increase and the complexity of systems continues to increase, distributed energy scheduling methods are receiving more and more attention. Distributed energy scheduling methods regard each generator as an independent intelligent agent. Each intelligent agent only exchanges information with adjacent intelligent agents. The various intelligent agents jointly achieve the solution of the system through frequent information exchange. Although distributed energy scheduling methods can ensure privacy and robustness, when the scale reaches a certain level, the computing speed of distributed energy scheduling methods will be significantly reduced. Summary of the Invention

[0004] Based on this, it is necessary to provide a Transformer consistency optimization method, device, computer equipment, computer-readable storage medium and computer program product for economic dispatch of ultra-large-scale multi-regional interconnected power systems to address the above technical problems.

[0005] In a first aspect, the present application provides a Transformer consistency optimization method for economic dispatch of ultra-large-scale multi-region interconnected power systems. The method comprises:

[0006] Construct a provincial economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the provincial Transformer model to N. Construct a municipal economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the municipal Transformer model to N, and set the global number of iterations. , set the global maximum number of iterations value;

[0007] The provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system is optimized using a consistency method to obtain the power generation and consumption matrix of the provincial generator units in the ultra-large-scale multi-regional interconnected power system. The obtained power generation and consumption matrix of the provincial generator units is input into the provincial Transformer model to predict the optimal power generation of the provincial generator units.

[0008] The consistency method is used to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and the power generation consumption matrix of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system is obtained. The obtained power generation consumption matrix of the municipal generator sets is input into the municipal Transformer model to predict the optimal power generation of the municipal generator sets.

[0009] In a second aspect, the present application also provides a Transformer consistency optimization device for economic dispatch of ultra-large-scale multi-region interconnected power systems. The device comprises:

[0010] Initialization module, used to build a provincial economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the provincial Transformer model to N. Build a municipal economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the municipal Transformer model to N. Set the number of global iterations. , set the global maximum number of iterations value;

[0011] The provincial economic dispatch module is used to optimize the provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system using a consistency method, obtain the power generation and consumption matrix of the provincial generator units in the ultra-large-scale multi-regional interconnected power system, and input the obtained power generation and consumption matrix of the provincial generator units into the provincial Transformer model to predict the optimal power generation of the provincial generator units;

[0012] The municipal economic dispatch module uses a consistency method to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, obtains the power generation and consumption matrix of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system, and inputs the obtained power generation and consumption matrix of the municipal generator sets into the municipal Transformer model to predict the optimal power generation of the municipal generator sets.

[0013] In a third aspect, the present application further provides a computer device. The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are performed:

[0014] Construct a provincial economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the provincial Transformer model to N. Construct a municipal economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the municipal Transformer model to N, and set the global number of iterations. , set the global maximum number of iterations value;

[0015] The provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system is optimized using a consistency method to obtain the power generation and consumption matrix of the provincial generator units in the ultra-large-scale multi-regional interconnected power system. The obtained power generation and consumption matrix of the provincial generator units is input into the provincial Transformer model to predict the optimal power generation of the provincial generator units.

[0016] The consistency method is used to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and the power generation consumption matrix of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system is obtained. The obtained power generation consumption matrix of the municipal generator sets is input into the municipal Transformer model to predict the optimal power generation of the municipal generator sets.

[0017] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the following steps:

[0018] Construct a provincial economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the provincial Transformer model to N. Construct a municipal economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the municipal Transformer model to N, and set the global number of iterations. , set the global maximum number of iterations value;

[0019] The provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system is optimized using a consistency method to obtain the power generation and consumption matrix of the provincial generator units in the ultra-large-scale multi-regional interconnected power system. The obtained power generation and consumption matrix of the provincial generator units is input into the provincial Transformer model to predict the optimal power generation of the provincial generator units.

[0020] The consistency method is used to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and the power generation consumption matrix of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system is obtained. The obtained power generation consumption matrix of the municipal generator sets is input into the municipal Transformer model to predict the optimal power generation of the municipal generator sets.

[0021] In a fifth aspect, the present application further provides a computer program product. The computer program product includes a computer program that, when executed by a processor, implements the following steps:

[0022] Construct a provincial economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the provincial Transformer model to N. Construct a municipal economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the municipal Transformer model to N, and set the global number of iterations. , set the global maximum number of iterations value;

[0023] The provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system is optimized using a consistency method to obtain the power generation and consumption matrix of the provincial generator units in the ultra-large-scale multi-regional interconnected power system. The obtained power generation and consumption matrix of the provincial generator units is input into the provincial Transformer model to predict the optimal power generation of the provincial generator units.

[0024] The consistency method is used to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and the power generation consumption matrix of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system is obtained. The obtained power generation consumption matrix of the municipal generator sets is input into the municipal Transformer model to predict the optimal power generation of the municipal generator sets.

[0025] The above-mentioned Transformer consistency optimization method, device, computer equipment, storage medium and computer program product for the economic dispatch of ultra-large-scale multi-regional interconnected power systems constructs a provincial economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints, and sets the number of encoder and decoder layers of the provincial Transformer model to N. It constructs a municipal economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints, and sets the number of encoder and decoder layers of the municipal Transformer model to N, and sets the number of global iterations. , set the global maximum number of iterations value; the consistency method is used to optimize the provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and the power generation consumption matrix of the provincial generator sets in the ultra-large-scale multi-regional interconnected power system is obtained. The power generation consumption matrix of the provincial generator sets is input into the provincial Transformer model to predict the optimal power generation of the provincial generator sets; the consistency method is used to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and the power generation consumption matrix of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system is obtained. The power generation consumption matrix of the municipal generator sets is input into the municipal Transformer model to predict the optimal power generation of the municipal generator sets; it can improve the operation speed and solution accuracy, improve privacy and security, and improve the robustness of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 The figure is a flow chart of a Transformer consistency optimization method in one embodiment.

[0027] Figure 2 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0029] In one embodiment, Figure 1As shown, a flow chart of the Transformer consistency optimization method is provided. This embodiment uses the method applied to a terminal as an example. It is understandable that the method can also be applied to a server, and can also be applied to a system including a terminal and a server, and implemented through the interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0030] A Transformer consistency optimization method for economic dispatch of ultra-large-scale multi-regional interconnected power systems combines the Transformer model and the consistency optimization method for economic dispatch of ultra-large-scale multi-regional interconnected power systems. It has the ability to improve the solution speed and reliability of ultra-large-scale generator set systems. The steps in the use process are:

[0031] Step (1) constructs a provincial economic dispatch optimization model for a large-scale multi-regional interconnected power system with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints, and sets the number of encoder and decoder layers of the provincial Transformer model to be N. Constructs a municipal economic dispatch optimization model for a large-scale multi-regional interconnected power system with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints, and sets the number of encoder and decoder layers of the municipal Transformer model to be N, and sets the number of global iterations. , set the global maximum number of iterations value;

[0032] Construct a provincial economic dispatch optimization model for a large-scale multi-regional interconnected power system with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation areas as constraints;

[0033] Power consumption for:

[0034]

[0035] Where, To obtain the instantaneous active power of each generator set; For the Unit No. Active power output during the time period; 、 and Respectively The power consumption coefficients of the quadratic, linear and constant terms of each unit;

[0036] Carbon emissions for:

[0037]

[0038] Where, 、 and Respectively Carbon emission coefficients of the quadratic, linear and constant terms of each unit;

[0039] No. Multi-objective values after linear weighting of units for:

[0040]

[0041] Where, is the weight factor;

[0042] The upper and lower output limits of the unit and the prohibited operation area constraints of the generator are:

[0043]

[0044]

[0045]

[0046] Where, Refers to the minimum value; Refers to the maximum value; The total number of provincial-level generating units in the ultra-large-scale multi-regional interconnected power system; The index number of the generator set; For the The number of generator indexes in the unit; is the lower limit of output; The output limit; for the lower realm; For the upper bound; For the The lower limit of the first prohibited operation area of the unit; For the Unit No. The upper bound of the prohibited operation area; For the Unit No. The lower bound of the prohibited operation area; For the Unit No. The upper bound of the prohibited operation area; It is a prohibited operation area; For the Total number of prohibited operation areas for each unit;

[0047] Construct a provincial economic dispatch optimization model for a large-scale multi-regional interconnected power system with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation areas as constraints;

[0048] Step (2) uses a consistency method to optimize the provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system, obtains the provincial power generation consumption matrix of the ultra-large-scale multi-regional interconnected power system, and inputs the obtained provincial power generation consumption matrix into the provincial Transformer model to predict the optimal power generation of the provincial power generation unit: set the leader intelligent agent and follower intelligent agent of the provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and uses a consistency method to optimize the provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system with power generation consumption and carbon emissions as targets and with upper and lower output constraints of the unit and prohibited operation area constraints of the generator as constraints, and obtains the provincial power generation consumption matrix of the ultra-large-scale multi-regional interconnected power system, and inputs the obtained provincial power generation consumption matrix into the provincial Transformer model to predict the optimal power generation of the provincial power generation unit;

[0049] set up value, The provincial economic dispatch model for ultra-large-scale multi-regional interconnected power systems The total system load demand during the period, setting the number of iterations of the provincial economic dispatch model for ultra-large-scale multi-regional interconnected power systems and the maximum number of iterations of the provincial economic dispatch model for ultra-large-scale multi-regional interconnected power systems ;

[0050] Set up leader and follower agents for the provincial economic dispatch model of a large-scale multi-regional interconnected power system;

[0051] Set the adjacency matrix for the consistency method:

[0052]

[0053] in,

[0054]

[0055]

[0056]

[0057]

[0058]

[0059] Where, is the adjacency matrix, a matrix with values of 0 or 1; for In the matrix The elements at is the total number of elements in matrix A; and are the Laplace matrices Elements in is the row random matrix of the agent; and 1st and Multi-objective values after linear weighting of the units; and 1st and Unit No. Active power output during the time period; for The consistency variable of the agent group at the moment; the consistency variable of the follower in the agent is calculated as ; The consistency variable of the leader among the agents is calculated as ; for Moment The intelligent unit is in the The consistency variable of the iteration; for At the moment, the jth intelligent agent unit is The consistency variable of the iteration; is the power balance adjustment factor of the distributed consensus algorithm; for Moment The power deviation of the iteration; for Moment The intelligent unit is in the The active power output of the iteration; is the number of hierarchical indexes; is the partition index number; is the total number of regions in this layer; is the number of layers; is the total number of layers; For the Tier The active power ceiling of a region is the sum of the active power ceilings of provincial-level generators in the ultra-large-scale multi-regional interconnected power system divided by the region;

[0060] The upper and lower output limits of the unit and the prohibited operation area constraints of the generator are:

[0061]

[0062] for Moment Unit No. The prohibited operation area is in The active power output of the iteration; The index number range of provincial-level generator units for ultra-large-scale multi-regional interconnected power systems;

[0063] The active power deviation is calculated based on the corrected equation result. The active power deviation is:

[0064]

[0065] Where, To solve for the active power deviation;

[0066] Update the power generation consumption matrix of the generator set , determine the power deviation Is it less than the maximum allowable power deviation? ;like , transmitting the updated provincial optimal power generation of the ultra-large-scale multi-regional interconnected power system As the first city-level ultra-large-scale multi-regional interconnected power system The total system load demand during the period continues to update the provincial power generation consumption matrix of the ultra-large-scale multi-regional interconnected power system; if , making the provincial iteration number of ultra-large-scale multi-regional interconnected power systems , determine the number of provincial iterations for ultra-large-scale multi-regional interconnected power systems Is it greater than or equal to the maximum provincial iteration number of the ultra-large-scale multi-regional interconnected power system? ;

[0067] like , continue to update the power generation and consumption matrix of the provincial-level generator sets in the ultra-large-scale multi-regional interconnected power system, and obtain the power generation and consumption matrix of the provincial-level generator sets in the ultra-large-scale multi-regional interconnected power system;

[0068] like , the obtained provincial generator power consumption matrix is input into the provincial Transformer model to predict the optimal power generation of provincial generators in the ultra-large-scale multi-regional interconnected power system;

[0069] The provincial Transformer model includes: an embedding layer, an encoder and a decoder; the encoder includes multiple encoder layers, each encoder layer includes a multi-head self-attention mechanism and a feedforward neural network; the decoder includes multiple decoder layers, each decoder layer includes a multi-head self-attention mechanism, an encoder-decoder attention mechanism and a feedforward neural network; the multi-head self-attention mechanism includes multiple parallel attention heads, each attention head performs a linear transformation on the input through a query matrix, a key matrix and a value matrix, and calculates the attention weight; the encoder-decoder attention mechanism allows the decoder to pay attention to the output of the encoder; the feedforward neural network includes two linear transformation layers and a nonlinear activation function; the encoder and decoder also include residual connections and layer normalization operations.

[0070] Generator set power consumption matrix Enter the provincial Transformer model. The embedding layer mathematical model of the provincial Transformer model is:

[0071]

[0072]

[0073]

[0074]

[0075] Where, The number of generator sets connected in real time for participation; is the dimension of the embedding vector; The power consumption matrix of each generator set is calculated from the real-time active power of each generator set before adding the position coding; is the set of real numbers, The size of ; The power consumption matrix of each generator set is calculated from the real-time active power of each generator set after adding position coding. The size of ; It is the position coding matrix of the real-time power consumption of each generator set in the power plant; for The position index in the sequence; for Half the dimension index of the position encoding vector in the sequence; is a sine function; is the cosine function; for The positional encoding value of even-numbered bits in the sequence; for The positional encoding value of the odd-numbered bits in the sequence; For input elements The corresponding unit complex power embedding vector; For input elements The corresponding unit complex power embedding vector; For input elements The corresponding unit complex power embedding vector;

[0076] The encoder mathematical model of the provincial Transformer model is:

[0077]

[0078]

[0079]

[0080]

[0081]

[0082]

[0083]

[0084]

[0085]

[0086]

[0087] Where, is the query matrix; is the bond matrix; is the value matrix; 、 and are the projection matrices for query, key, and value, respectively, 、 and The values are obtained by gradient descent algorithm. 、 and The dimensions are ; is the number of attention heads; is the middle-level dimension; is an activation function used to normalize the numerical vector into a probability distribution vector, and The sum of the probabilities of each probability distribution vector in the function is 1; is the self-attention mechanism function, Sequentially extract the activation function The variables with the highest coefficients are reordered; is the bond matrix The transpose of For the The calculation results of the self-attention mechanism function; 、 and Respectively The projection matrices of the query, key, and value of each head, 、 and The values are obtained by gradient descent algorithm. 、 and The dimensions are ; 、 and After linear transformation, 、 and Vector, used for attention calculation of the self-attention mechanism function head; is the multi-head attention mechanism function, Allows the model to learn different attention patterns in multiple subspaces; is the splicing function; 、 and The first, second and The calculation results of the self-attention mechanism function; is the linear transformation matrix of the spliced output, The value is obtained by gradient descent algorithm. The dimension is ; is the output after residual connection and normalization operation; is the normalization function; and The weight matrices of the first and second layers are initialized by the input dimension; for Activation function; and are the bias terms for the first and second layers respectively; It is a feedforward neural network function, which is used to increase the nonlinear expression ability of the model; The final output after the feedforward network, that is, the optimized data matrix of the power generation and consumption of the generator set processed by the encoder;

[0088] Output As the decoder area Input;

[0089] The decoder mathematical model of the provincial Transformer model is:

[0090]

[0091]

[0092]

[0093]

[0094]

[0095]

[0096]

[0097]

[0098]

[0099]

[0100]

[0101] Where, To mask the multi-head self-attention function, by shielding or modifying the attention weights between certain positions, the attention mechanism can be constrained and guided to enhance the logic of the model; The result of the second sub-layer of the decoder combined with the encoder output; is the final output after the feedforward network, and is the power generation and consumption prediction data matrix of the generator set processed by the encoder and decoder; is the mask matrix;

[0102] The output after encoder and decoder processing Perform linearization processing and obtain the optimal power generation of provincial generators in ultra-large-scale multi-regional interconnected power systems through the fully connected layer;

[0103] Step (3): Use the consistency method to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, obtain the power generation consumption matrix of the municipal generator set of the ultra-large-scale multi-regional interconnected power system, and input the obtained power generation consumption matrix of the municipal generator set into the municipal Transformer model to predict the optimal power generation of the municipal generator set: set the leader intelligent agent and follower intelligent agent of the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, use the consistency method to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system with power generation consumption and carbon emissions as targets, and with the upper and lower limit output constraints of the unit and the generator prohibited operation area constraints as constraints, and obtain the power generation consumption matrix of the municipal generator set of the ultra-large-scale multi-regional interconnected power system, and input the obtained power generation consumption matrix of the municipal generator set into the municipal Transformer model to predict the optimal power generation of the municipal generator set;

[0104] Enter the city-level economic dispatch model of ultra-large-scale multi-regional interconnected power system, set the first Total system load demand during the period ; Set the number of iterations of the municipal economic dispatch model for ultra-large-scale multi-regional interconnected power systems and the maximum number of iterations of the municipal economic dispatch model for ultra-large-scale multi-regional interconnected power systems ; Set up followers and leaders of the city-level economic dispatch model of the ultra-large-scale multi-regional interconnected power system; establish an optimization model of the city-level economic dispatch model of the ultra-large-scale multi-regional interconnected power system with power generation consumption and carbon emissions as targets, and with the upper and lower output limits of the units and the generator prohibited operation area constraints as constraints, and calculate the first part of the city-level economic dispatch model of the ultra-large-scale multi-regional interconnected power system Multi-objective values after linear weighting of units ; Calculate the next iteration active power output of each intelligent agent in the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system; Determine whether the active power output of each unit in the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system exceeds the limit and make corrections; Determine the active power deviation and determine the active power deviation Is it less than the maximum allowable power deviation of the municipal generator set in the ultra-large-scale multi-regional interconnected power system? ,like , the consistency method is used to iteratively obtain the power consumption matrix of the municipal generator units in the ultra-large-scale multi-regional interconnected power system; if , let the number of iterations of the municipal economic dispatch model for ultra-large-scale multi-regional interconnected power systems be , to determine the number of iterations of the municipal economic dispatch model for ultra-large-scale multi-regional interconnected power systems Is it greater than or equal to the maximum number of iterations of the municipal economic dispatch model for ultra-large-scale multi-regional interconnected power systems? ;like , the obtained power consumption matrix of the municipal generator sets is input into the municipal Transformer model to predict the optimal power generation of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system; if , continue to update the power generation and consumption matrix of municipal generators in the ultra-large-scale multi-regional interconnected power system;

[0105] Determine the number of global iterations Is it greater than or equal to the global maximum number of iterations? ;like , output the updated optimal power generation of the generator units in the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system; if ,make , the consistency method is used to optimize the provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and the obtained provincial generator unit power consumption matrix is input into the provincial Transformer model to predict the optimal power generation of the provincial generator units in the ultra-large-scale multi-regional interconnected power system;

[0106] The city-level Transformer model includes: an embedding layer, an encoder and a decoder; the encoder includes multiple encoder layers, each encoder layer includes a multi-head self-attention mechanism and a feedforward neural network; the decoder includes multiple decoder layers, each decoder layer includes a multi-head self-attention mechanism, an encoder-decoder attention mechanism and a feedforward neural network; the multi-head self-attention mechanism includes multiple parallel attention heads, each attention head performs a linear transformation on the input through a query matrix, a key matrix and a value matrix, and calculates the attention weight; the encoder-decoder attention mechanism allows the decoder to pay attention to the output of the encoder; the feedforward neural network includes two linear transformation layers and a nonlinear activation function; the encoder and decoder also include residual connections and layer normalization operations.

[0107] The calculated optimal power generation of the generator sets for provincial and municipal economic dispatch of the ultra-large-scale multi-regional interconnected power system is sent as an instruction to the generator sets of the ultra-large-scale multi-regional interconnected power system, thereby realizing the dispatch of the generator sets of the ultra-large-scale multi-regional interconnected power system.

[0108] The provincial and municipal Transformer models have the same structure, but different inputs and outputs. The provincial Transformer model takes as input the provincial generator unit power consumption matrix; its output is the optimal power generation of provincial generator units in the ultra-large-scale multi-regional interconnected power system. The municipal Transformer model takes as input the municipal generator unit power consumption matrix; its output is the optimal power generation of municipal generator units in the ultra-large-scale multi-regional interconnected power system.

[0109] The present invention has the following advantages and effects compared to the prior art:

[0110] (1) This invention combines the large-scale Transformer model with the consistency optimization method for the first time, which can not only use the large model for acceleration, but also use the consistency optimization method to improve accuracy.

[0111] (2) The present invention adopts a layered approach to solve multi-layer optimization problems.

[0112] (3) The present invention adopts a consistent optimization method to solve distributed optimization problems.

[0113] (4) The present invention can simultaneously solve distributed multi-objective hierarchical optimization problems, ensure the optimization speed, and obtain the optimal solution.

[0114] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 2As shown. The computer device includes a processor, memory, an input / output interface, a communication interface, a display unit, and an input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are connected to the system bus via the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and external devices. The communication interface of the computer device is used to communicate with external terminals via wired or wireless means, and the wireless means can be implemented via Wi-Fi, mobile cellular networks, NFC (near-field communication), or other technologies. When executed by the processor, the computer program implements a Transformer consistency optimization method for economic dispatch of ultra-large-scale multi-regional interconnected power systems. The display unit of the computer device is used to produce visual images and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device casing, or an external keyboard, touchpad or mouse.

[0115] Those skilled in the art will understand that Figure 2 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0116] In one embodiment, a computer device is further provided, including a memory and a processor. The memory stores a computer program, and the processor implements the steps in the above method embodiments when executing the computer program.

[0117] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0118] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0119] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with the relevant laws, regulations and standards of relevant countries and regions.

[0120] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the above-mentioned embodiments. In particular, any reference to memory, database, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The databases involved in the various embodiments provided herein may include at least one of a relational database and a non-relational database. Non-relational databases may include, but are not limited to, distributed databases based on blockchains. The processors involved in the various embodiments provided herein may be, but are not limited to, general-purpose processors, central processing units (CPUs), graphics processing units (GPUs), digital signal processors (DSPs), programmable logic devices (PLDs), data processing logic devices based on quantum computing, and the like.

[0121] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0122] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present application. It should be noted that a person of ordinary skill in the art may make various modifications and improvements without departing from the spirit of the present application, and these modifications and improvements fall within the scope of protection of the present application. Therefore, the scope of protection of the present application shall be determined by the appended claims.

Claims

1. A Transformer consistency optimization method for economic dispatch of ultra-large-scale multi-regional interconnected power systems, characterized by: The method comprises: Construct a provincial economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the provincial Transformer model to N. Construct a municipal economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the municipal Transformer model to N, and set the global number of iterations. , set the global maximum number of iterations value; The provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system is optimized using a consistency method to obtain the power generation and consumption matrix of the provincial generator units in the ultra-large-scale multi-regional interconnected power system. The obtained power generation and consumption matrix of the provincial generator units is input into the provincial Transformer model to predict the optimal power generation of the provincial generator units. The consistency method is used to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system, and the power generation consumption matrix of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system is obtained. The obtained power generation consumption matrix of the municipal generator sets is input into the municipal Transformer model to predict the optimal power generation of the municipal generator sets.

2. The Transformer consistency optimization method for economic dispatch of ultra-large-scale multi-regional interconnected power systems according to claim 1 is characterized in that: The provincial Transformer model and the municipal Transformer model have the same structure, namely, they include: an embedding layer, an encoder and a decoder; wherein, the encoder includes multiple encoder layers, each encoder layer includes a multi-head self-attention mechanism and a feedforward neural network; the decoder includes multiple decoder layers, each decoder layer includes a multi-head self-attention mechanism, an encoder-decoder attention mechanism and a feedforward neural network; the multi-head self-attention mechanism includes multiple parallel attention heads, each attention head performs a linear transformation on the input through a query matrix, a key matrix and a value matrix, and calculates the attention weight; the encoder-decoder attention mechanism allows the decoder to pay attention to the output of the encoder; the feedforward neural network includes two linear transformation layers and a nonlinear activation function; the encoder and decoder also include residual connections and layer normalization operations.

3. A Transformer consistency optimization device for economic dispatch of ultra-large-scale multi-region interconnected power systems, characterized by: The device comprises: Initialization module, used to build a provincial economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the provincial Transformer model to N. Build a municipal economic dispatch optimization model for ultra-large-scale multi-regional interconnected power systems with power generation consumption and carbon emissions as targets, and with upper and lower output limits of units and generator prohibited operation area constraints as constraints. Set the number of encoder and decoder layers of the municipal Transformer model to N. Set the number of global iterations. , set the global maximum number of iterations value; The provincial economic dispatch module is used to optimize the provincial economic dispatch model of the ultra-large-scale multi-regional interconnected power system using a consistency method, obtain the power generation and consumption matrix of the provincial generator units in the ultra-large-scale multi-regional interconnected power system, and input the obtained power generation and consumption matrix of the provincial generator units into the provincial Transformer model to predict the optimal power generation of the provincial generator units; The municipal economic dispatch module is used to optimize the municipal economic dispatch model of the ultra-large-scale multi-regional interconnected power system using a consistency method, obtain the power generation and consumption matrix of the municipal generator sets in the ultra-large-scale multi-regional interconnected power system, and input the obtained power generation and consumption matrix of the municipal generator sets into the municipal Transformer model to predict the optimal power generation of the municipal generator sets.

4. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the Transformer consistency optimization method for economic dispatch of ultra-large-scale multi-regional interconnected power systems as described in any one of claims 1 to 2 are implemented.

5. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the Transformer consistency optimization method for economic dispatch of ultra-large-scale multi-regional interconnected power systems described in any one of claims 1 to 2 are implemented.

Images