Unified electricity market clearing method and device, computer equipment, storage medium and computer program product

Abstract

The invention relates to a clearing method and device for a unified electricity market, computer equipment, a storage medium and a computer program product. Relates to the technical field of power systems. The method comprises the following steps: constructing a tie line variable data set based on historical operation data of a provincial power market in a unified power market; obtaining the current clearing cost based on each tie line variable in combination with the current clearing condition; constructing an explicit expression of tie line variables and clearing cost based on a preset sparse polynomial chaos expansion algorithm, and reconstructing a preset hierarchical clearing model of the unified power market into a single-layer clearing model; constructing a three-dimensional tensor and a tensor feasible region based on historical clearing data and physical constraints, and performing dimension reduction processing on the current high-dimensional clearing variable feasible region in combination with a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie line variable feasible region; and performing optimization processing on the single-layer clearing model to obtain a clearing result of the unified electricity market. By adopting the method, the clearing efficiency of the clearing method of the unified electricity market can be improved.

Description

Technical Field

[0001] This application relates to the field of power system technology, and in particular to a clearing method, apparatus, computer equipment, storage medium and computer program product for a unified electricity market. Background Technology

[0002] Currently, the allocation of electricity resources exhibits a reverse distribution between supply and demand, necessitating inter-provincial electricity trading to achieve resource surpluses and shortages across different regions. Therefore, constructing a unified electricity market and promoting coordinated clearing across provincial and regional markets has become an important research direction.

[0003] However, current electricity market clearing methods typically construct a hierarchical clearing model for the unified electricity market. The lower-level model solves the optimal clearing problem for each provincial market, while the upper-level model solves the overall coordination problem of the unified electricity market. The optimal solution is found through iterative optimization, which often makes it difficult to guarantee the convergence of the algorithm during the iteration process. It requires multiple iterations to find a relatively good feasible solution, resulting in insufficient computational efficiency.

[0004] Therefore, the current clearing method for the unified electricity market suffers from low clearing efficiency. Summary of the Invention

[0005] Therefore, it is necessary to address the technical problem of low clearing efficiency in the current unified electricity market clearing methods by providing an electricity market clearing method, apparatus, computer equipment, computer-readable storage medium, and computer program product.

[0006] Firstly, this application provides a clearing method for a unified electricity market, including:

[0007] Based on the historical operation data of the provincial power markets in the unified power market, a tie-line variable dataset is constructed for the provincial power markets, which includes at least the power and voltage variables at the tie-line points.

[0008] Based on each tie-line variable in the tie-line variable dataset, and combined with the current clearing conditions of the provincial power market, the current clearing cost of the provincial power market is obtained;

[0009] An explicit expression for the tie-line variables and the clearing cost is constructed based on a preset sparse polynomial chaotic expansion algorithm. Based on the explicit expression, the preset hierarchical clearing model of the unified electricity market is reconstructed into a single-layer clearing model.

[0010] Based on the historical clearing data and physical constraints of the provincial power market obtained in advance, a three-dimensional tensor and the corresponding tensor feasible region are constructed. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie-line variable feasible region.

[0011] The single-layer clearing model is optimized based on the feasible region of the low-dimensional tie-line variables to obtain the clearing result of the unified electricity market.

[0012] In one embodiment, the step of constructing an explicit expression for the tie-line variable and the clearing cost based on a preset sparse polynomial chaotic expansion algorithm includes: processing target power data based on a preset generalized polynomial chaotic expansion algorithm to construct a sparse basis function set; obtaining the polynomial coefficients of the sparse basis function set using a preset subspace tracking algorithm based on the tie-line variable and the clearing cost; and constructing an explicit expression for the tie-line variable and the clearing cost based on the polynomial coefficients and the sparse basis function set.

[0013] In one embodiment, obtaining the polynomial coefficients of the sparse basis function set through a preset subspace tracking algorithm includes: obtaining the correlation between the sparse basis function set and the clearing cost, and obtaining an index set corresponding to at least one basis function whose absolute value of the correlation is greater than a preset value; determining the initial residual of the index set, and continuously iterating to calculate the correlation between the residual and the basis function based on the initial residual to obtain an updated index set; updating the residual based on the updated index set and pre-obtained temporary coefficients and repeating the iteration; stopping the iteration when the residual meets a preset threshold or the number of iterations reaches a preset number, and obtaining the polynomial coefficients of the sparse basis function set.

[0014] In one embodiment, the step of constructing a three-dimensional tensor and the corresponding tensor feasible region based on the pre-acquired historical clearing data and physical constraints of the provincial power market includes: constructing the three-dimensional tensor and the corresponding tensor feasible region based on the node information, physical quantity information and time information included in the pre-acquired historical clearing data of the provincial power market, and the physical constraints of the provincial power market including ramp constraints, unit output constraints, power transmission distribution factor matrix, transmission line capacity upper limit, reactance of inter-node connecting lines and upper and lower limits of node voltage angle.

[0015] In one embodiment, the step of reducing the dimensionality of the current high-dimensional clearing variable feasible region based on the three-dimensional tensor and the corresponding tensor feasible region, combined with a preset dynamic tensor decomposition algorithm, to obtain a low-dimensional connection variable feasible region includes: extracting the spatial and temporal correlation factors, physical quantity and temporal correlation factors, and spatial and physical quantity correlation factors of the low-dimensional mapping model corresponding to the three-dimensional tensor; solving the tensor low-dimensional mapping model based on a preset proximal alternating minimization algorithm, and outputting the low-dimensional factors when the results meet preset convergence conditions; and obtaining the low-dimensional connection variable feasible region based on the low-dimensional factors.

[0016] In one embodiment, constructing a tie-line variable dataset for the provincial power market, which includes at least power and voltage variables at tie-line locations, based on historical operating data of the provincial power market within the unified power market, includes: obtaining the power supply periods of the provincial power market and the number of tie-lines in each region during the power supply periods based on historical operating data of the provincial power market within the unified power market; extracting power and voltage data corresponding to each tie-line during each power supply period from the historical operating data; and integrating the correspondence between multiple power and voltage data and each tie-line to obtain the tie-line variable dataset.

[0017] Secondly, this application also provides a clearing device for a unified electricity market, comprising:

[0018] The dataset construction module is used to construct a tie-line variable dataset for the provincial power market, which includes at least power and voltage variables at the tie-line, based on the historical operating data of the provincial power market in the unified power market.

[0019] The data acquisition module is used to obtain the current clearing cost of the provincial power market based on each tie-line variable in the tie-line variable dataset and in combination with the current clearing conditions of the provincial power market.

[0020] The data processing module is used to construct an explicit expression for the tie-line variables and the clearing cost based on a preset sparse polynomial chaotic expansion algorithm, and to reconstruct the preset hierarchical clearing model of the unified electricity market into a single-layer clearing model based on the explicit expression.

[0021] The data processing module is also used to construct a three-dimensional tensor and the corresponding tensor feasible region based on the historical clearing data and physical constraints of the provincial power market obtained in advance. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie-line variable feasible region.

[0022] The result acquisition module is used to optimize the single-layer clearing model based on the feasible domain of the low-dimensional tie-line variables to obtain the clearing result of the unified electricity market.

[0023] Thirdly, this application also provides a computer device, the computer device including a memory and a processor, the memory storing a computer program, and the processor executing the computer program to perform the following steps:

[0024] Based on the historical operation data of the provincial power markets in the unified power market, a tie-line variable dataset is constructed for the provincial power markets, which includes at least the power and voltage variables at the tie-line points.

[0025] Based on each tie-line variable in the tie-line variable dataset, and combined with the current clearing conditions of the provincial power market, the current clearing cost of the provincial power market is obtained;

[0026] An explicit expression for the tie-line variables and the clearing cost is constructed based on a preset sparse polynomial chaotic expansion algorithm. Based on the explicit expression, the preset hierarchical clearing model of the unified electricity market is reconstructed into a single-layer clearing model.

[0027] Based on the historical clearing data and physical constraints of the provincial power market obtained in advance, a three-dimensional tensor and the corresponding tensor feasible region are constructed. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie-line variable feasible region.

[0028] The single-layer clearing model is optimized based on the feasible region of the low-dimensional tie-line variables to obtain the clearing result of the unified electricity market.

[0029] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, the computer program performing the following steps when executed by a processor:

[0030] Based on the historical operation data of the provincial power markets in the unified power market, a tie-line variable dataset is constructed for the provincial power markets, which includes at least the power and voltage variables at the tie-line points.

[0031] Based on each tie-line variable in the tie-line variable dataset, and combined with the current clearing conditions of the provincial power market, the current clearing cost of the provincial power market is obtained;

[0032] An explicit expression for the tie-line variables and the clearing cost is constructed based on a preset sparse polynomial chaotic expansion algorithm. Based on the explicit expression, the preset hierarchical clearing model of the unified electricity market is reconstructed into a single-layer clearing model.

[0033] Based on the historical clearing data and physical constraints of the provincial power market obtained in advance, a three-dimensional tensor and the corresponding tensor feasible region are constructed. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie-line variable feasible region.

[0034] The single-layer clearing model is optimized based on the feasible region of the low-dimensional tie-line variables to obtain the clearing result of the unified electricity market.

[0035] Fifthly, this application also provides a computer program product, which includes a computer program that, when executed by a processor, performs the following steps:

[0036] Based on the historical operation data of the provincial power markets in the unified power market, a tie-line variable dataset is constructed for the provincial power markets, which includes at least the power and voltage variables at the tie-line points.

[0037] Based on each tie-line variable in the tie-line variable dataset, and combined with the current clearing conditions of the provincial power market, the current clearing cost of the provincial power market is obtained;

[0038] An explicit expression for the tie-line variables and the clearing cost is constructed based on a preset sparse polynomial chaotic expansion algorithm. Based on the explicit expression, the preset hierarchical clearing model of the unified electricity market is reconstructed into a single-layer clearing model.

[0039] Based on the historical clearing data and physical constraints of the provincial power market obtained in advance, a three-dimensional tensor and the corresponding tensor feasible region are constructed. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie-line variable feasible region.

[0040] The single-layer clearing model is optimized based on the feasible region of the low-dimensional tie-line variables to obtain the clearing result of the unified electricity market.

[0041] The aforementioned unified electricity market clearing method, apparatus, computer equipment, storage medium, and computer program product, in the process of clearing the unified electricity market, firstly, based on the historical operating data of the provincial electricity markets within the unified electricity market, constructs a tie-line variable dataset for the provincial electricity markets, including at least the power and voltage variables at tie-line locations; then, based on each tie-line variable in the tie-line variable dataset, combined with the current clearing conditions of the provincial electricity market, the current clearing cost of the provincial electricity market is obtained; next, based on a preset sparse polynomial chaotic expansion algorithm, an explicit expression for the tie-line variables and clearing cost is constructed; based on the explicit expression, the preset hierarchical clearing model of the unified electricity market is reconstructed into a single-layer clearing model; then, based on the pre-acquired historical clearing data and physical constraints of the provincial electricity markets, a three-dimensional tensor and its corresponding tensor feasible region are constructed; based on the three-dimensional tensor and its corresponding tensor feasible region, combined with a preset dynamic tensor decomposition algorithm, the current high-dimensional clearing variable feasible region is dimensionality-reduced to obtain a low-dimensional tie-line variable feasible region; finally, based on the low-dimensional tie-line variable feasible region, the single-layer clearing model is optimized to obtain the clearing result of the unified electricity market. In the above process, by constructing a tie-line variable dataset based on historical data and by constructing explicit expressions based on sparse polynomial chaotic expansion, the hierarchical clearing model was reconstructed into a single-layer clearing model, simplifying the model structure and reducing the solution complexity. Furthermore, by using the dynamic tensor decomposition algorithm to reduce the dimensionality of the feasible region of high-dimensional clearing variables to obtain the feasible region of low-dimensional tie-line variables, the computational efficiency of high-dimensional variables can be effectively improved, thereby improving the clearing efficiency of the clearing method for the unified electricity market. Attached Figure Description

[0042] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0043] Figure 1 This is a flowchart illustrating a clearing method for a unified electricity market in one embodiment;

[0044] Figure 2 This is a schematic diagram illustrating the construction process of the tie-line variable dataset in the clearing method of a unified electricity market in one embodiment;

[0045] Figure 3 This is a schematic diagram illustrating the construction process of explicit expressions in the clearing method of a unified electricity market in one embodiment;

[0046] Figure 4This is a schematic diagram illustrating the process of obtaining polynomial coefficients in a unified electricity market clearing method in one embodiment;

[0047] Figure 5 This is a schematic diagram illustrating the process of reducing the dimensionality of the feasible domain of the current high-dimensional clearing variable in the clearing method of a unified electricity market in one embodiment.

[0048] Figure 6 This is a flowchart illustrating the clearing method for a unified electricity market in another embodiment;

[0049] Figure 7 This is a structural block diagram of a clearing device for a unified electricity market in one embodiment;

[0050] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0051] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0052] Because power resource allocation exhibits an inverse supply-demand distribution, inter-provincial power trading is necessary to achieve resource surplus and shortage mitigation among different regions. Constructing a unified power market and promoting coordinated clearing across provincial and regional markets has become an important research direction. Unified power market clearing is essentially a decentralized optimization problem. Currently, unified power markets are typically constructed as hierarchical clearing models. The lower-level model addresses the optimal clearing problem for each provincial market, while the upper-level model addresses the overall coordination problem of the unified power market, finding the optimal solution through iterative optimization. However, while existing methods reduce problem complexity through hierarchical decomposition, they often struggle to guarantee convergence during iteration, requiring multiple iterations to find a suitable feasible solution, resulting in insufficient computational efficiency. Furthermore, this method requires strong real-time communication capabilities between provincial power markets to ensure iterative updates, which is often difficult to achieve in actual clearing processes. Therefore, current power market clearing methods suffer from low clearing efficiency.

[0053] To address the problem of low clearing efficiency in the aforementioned electricity market clearing methods, in one embodiment, such as Figure 1 As shown, a method for clearing a unified electricity market is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0054] Step S102: Based on the historical operation data of the provincial power markets in the unified power market, construct a tie-line variable dataset for the provincial power markets, which includes at least the power and voltage variables at the tie-line points.

[0055] Historical operating data comprises various types of data generated during the operation of the provincial power market over a historical period, including generation output, load consumption, grid equipment operating parameters, and market transaction data. The tie-line variable dataset is a structured dataset built upon historical operating data, focusing on key operating parameters of tie-lines. It can be used to integrate tie-line-related variables, providing a data carrier for subsequent analysis of the correlation between tie-line variables and clearing costs, and defining the constraint range of tie-line variables. The power at the tie-line is the electrical power transmitted by the tie-line connecting different regional power grids. It is a parameter reflecting the transmission capacity and operating status of the tie-line, and can include active power and reactive power. The voltage variable can be the voltage amplitude at both ends or key nodes of the tie-line. Voltage changes affect power transmission efficiency.

[0056] Step S104: Based on each tie-line variable in the tie-line variable dataset and combined with the current clearing conditions of the provincial electricity market, obtain the current clearing cost of the provincial electricity market.

[0057] Among them, the current clearing conditions are the various constraints and conditions at the current moment when the provincial power market is calculating the clearing, which may include the load forecast value for the current period, the bidding data of power generation companies, the operating status of power grid equipment, etc.; the current clearing cost is the total cost required for the provincial power market to achieve supply and demand balance under the current clearing conditions, which may include the generation cost on the generation side and the network loss cost in the power grid transmission link, etc.

[0058] Step S106: Construct explicit expressions for tie-line variables and clearing costs based on a preset sparse polynomial chaotic expansion algorithm. Based on the explicit expressions, reconstruct the preset hierarchical clearing model of the unified electricity market into a single-layer clearing model.

[0059] Among them, the sparse polynomial chaotic expansion algorithm is an uncertainty quantification and function approximation algorithm based on polynomial chaos theory. It selects basis functions that have a significant impact on the result through sparsification, and then constructs a concise functional relationship between input and output variables. In the application, the sparse polynomial chaotic expansion algorithm is used to transform the complex implicit relationship between connection variables and clearing costs into an explicit expression that can be directly calculated. The explicit expression is an expression that directly uses connection variables as independent variables and clearing costs as dependent variables. It can directly obtain the output from the input without a complex model solution process. The hierarchical clearing model is a multi-stage, hierarchical optimization model. The single-level clearing model integrates the multi-stage, multi-level constraints and objectives in the hierarchical clearing model into a single-level optimization model through explicit expressions.

[0060] Step S108: Based on the historical clearing data and physical constraints of the provincial power market obtained in advance, a three-dimensional tensor and the corresponding tensor feasible region are constructed. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining the preset dynamic tensor decomposition algorithm to obtain the low-dimensional tie-line variable feasible region.

[0061] Among them, historical clearing data refers to the total amount of data generated during the historical clearing process of the provincial power market, including the constraints and input variables of each clearing, such as tie line variables and load data, clearing results, and clearing costs; physical constraints can include grid security constraints, such as the upper limit of line transmission power and voltage amplitude range, and can also include equipment operation constraints, such as the upper and lower limits of generator output and ramp rate limits; three-dimensional tensors are high-order data structures that can be used to efficiently integrate historical clearing data with multi-dimensional correlations during the clearing process of the provincial power market; the tensor feasible region is the range of values ​​for each dimension variable defined based on the three-dimensional tensor data and physical constraints.

[0062] Furthermore, the dynamic tensor decomposition algorithm is a dimensionality reduction algorithm for dynamically changing high-dimensional tensor data. It can map high-dimensional tensor data to a low-dimensional space while preserving the core correlation information of the data. The feasible region of high-dimensional cleared variables is the high-dimensional space composed of the value range of all cleared related variables. Dimensionality reduction is the process of reducing the number of variables in the high-dimensional feasible region and eliminating redundant variables by, for example, using the dynamic tensor decomposition algorithm, thus transforming the high-dimensional feasible region into a low-dimensional feasible region. The feasible region of low-dimensional connection variable is the value range obtained after dimensionality reduction.

[0063] Step S110: Optimize the single-layer clearing model based on the feasible region of low-dimensional tie-line variables to obtain the clearing result of the unified electricity market.

[0064] The optimization process involves solving and optimizing a single-layer clearing model within the constrained boundaries of the feasible region of low-dimensional tie-line variables. This can be achieved using optimization algorithms such as linear programming and nonlinear programming. The clearing result is the output of the clearing optimization, including the clearing price of each power generation company, the allocation of tie-line transmission power, and the total clearing cost.

[0065] The aforementioned clearing method for the unified electricity market first constructs a tie-line variable dataset for each provincial electricity market, based on historical operational data of the provincial electricity markets within the unified electricity market. This dataset includes at least power and voltage variables at tie-line points. Then, based on each tie-line variable in the dataset and the current clearing conditions of the provincial electricity market, the current clearing cost is obtained. Next, an explicit expression for the tie-line variables and clearing cost is constructed using a pre-defined sparse polynomial chaotic expansion algorithm. Based on this explicit expression, the pre-defined hierarchical clearing model of the unified electricity market is reconstructed into a single-layer clearing model. Then, based on pre-acquired historical clearing data and physical constraints of the provincial electricity markets, a three-dimensional tensor and its corresponding tensor feasible region are constructed. Using the three-dimensional tensor and its corresponding feasible region, a pre-defined dynamic tensor decomposition algorithm is used to reduce the dimensionality of the current high-dimensional clearing variable feasible region, resulting in a low-dimensional tie-line variable feasible region. Finally, the single-layer clearing model is optimized based on the low-dimensional tie-line variable feasible region to obtain the clearing result of the unified electricity market. In the above process, by constructing a tie-line variable dataset based on historical data and by constructing explicit expressions based on sparse polynomial chaotic expansion, the hierarchical clearing model was reconstructed into a single-layer clearing model, simplifying the model structure and reducing the solution complexity. Furthermore, by using the dynamic tensor decomposition algorithm to reduce the dimensionality of the feasible region of high-dimensional clearing variables to obtain the feasible region of low-dimensional tie-line variables, the computational efficiency of high-dimensional variables can be effectively improved, thereby improving the clearing efficiency of the electricity market clearing method.

[0066] More importantly, in this application, the electricity market includes provincial electricity markets and a unified electricity market, where the unified electricity market includes multiple provincial electricity markets. This application uses sparse polynomial chaotic expansion technology to fit the relationship between the clearing cost of the provincial electricity market and the tie-line variables of the provincial electricity market, thereby transforming the original hierarchical clearing of the unified electricity market into a single-layer clearing model, and realizing non-iterative collaborative clearing of cross-provincial electricity markets. Secondly, the dynamic tensor decomposition algorithm also reduces the dimensionality of the feasible region of the clearing variables of the provincial electricity market, representing it as a low-dimensional feasible region of tie-line variables, and ensures the feasibility of the clearing scheme within the provincial electricity market by judging the tie-line variables.

[0067] In one embodiment, for the clearing problem of a unified electricity market, existing methods typically solve it by constructing a hierarchical clearing model: the lower-level model aims to achieve market clearing within the province with the goal of optimal market operation; the upper-level model aims to achieve coordination and mutual assistance between provincial electricity markets with the goal of minimizing inter-provincial electricity purchase costs. The specific model description is as follows: the lower-level intra-provincial market clearing model aims to minimize market operation costs, and its objective function and constraints are shown in formulas (1)-(9). Among them, formula (1) is the objective function, formula (2) is the intra-provincial power balance constraint, formulas (3)-(4) are the generator output constraint, formulas (5)-(6) are the generator ramping constraint, formulas (7)-(8) are the generator reserve capacity constraint, and formula (9) is the line power flow constraint.

[0068] Formula (1):

[0069] Formula (2):

[0070] Formula (3):

[0071] Formula (4):

[0072] Formula (5):

[0073] Formula (6):

[0074] Formula (7):

[0075] Formula (8):

[0076] Formula (9):

[0077]

[0078] In the formula, This is a collection of provinces that receive electricity. A set of scheduling times; Indicates province A collection of generators; Indicates province The Middle The generators during the time period contribution; , They represent provinces respectively. The Middle The standby cost coefficients for generators; , They represent provinces respectively. The Middle The upper and lower standby capacities of the generators; For the first Cost calculation formula for one generator; Indicates province During the period Electricity purchased; Indicates province During the period The electricity purchase price; , They represent provinces respectively. During the period The output of wind and solar power; Indicates province The set of load nodes; Indicates province The load nodes during the time period Load demand; , They represent provinces respectively. The Middle The upper and lower limits of the generator's output; , They represent provinces respectively. The Middle The upper and lower limits of the ramp rate of the generator; province The Middle The upper limit of power flow on each line; , , , , These represent the power flow transfer factors for generators, wind power, photovoltaics, loads, and inter-provincial tie line nodes, respectively.

[0079] Furthermore, the upper-level inter-provincial market coordination model aims to minimize the cost of electricity purchase. Its objective function and constraints are shown in formulas (10)-(13), where formula (10) is the objective function, formula (11) is the tie-line power balance constraint, formula (12) is the tie-line transmission capacity constraint, and formula (13) is the sending-end generator capacity constraint.

[0080] Formula (10):

[0081] Formula (11):

[0082] Formula (12):

[0083] Formula (13):

[0084] In the formula, This is a collection of provinces that send electricity; For provinces The generator combination in; Indicates province The Middle Quotation for one generator; Indicates province The Middle The generators during the time period To the province Transmitted power; Indicates province With provinces The power transmission loss coefficient between them; , These represent the upper and lower limits of the power of inter-provincial connecting lines, respectively. , They represent provinces respectively. The Middle The upper and lower limits of the electricity output of the generator.

[0085] The aforementioned tiered clearing model couples inter-provincial electricity demand and inter-provincial electricity prices, and obtains the optimal clearing plan for the unified electricity market through multiple iterations of the upper and lower level models. However, it suffers from convergence issues between the upper and lower level models, resulting in numerous iterations and insufficient computational efficiency. Therefore, this application proposes a non-iterative unified electricity market collaborative clearing algorithm. It uses sparse polynomial chaotic expansion technology to fit the relationship between provincial electricity market clearing costs and tie-line variables, reconstructing the original tiered clearing model of the unified electricity market into a single-level clearing model, thereby achieving non-iterative collaborative clearing of the inter-provincial electricity market. Furthermore, to ensure the feasibility of the clearing scheme within the provincial electricity market, dynamic tensor decomposition technology is used to reduce the dimensionality of the originally high-dimensional provincial electricity market clearing variable feasible region, representing it as a low-dimensional tie-line variable feasible region. The feasibility of the clearing scheme within the provincial electricity market is ensured through the judgment of tie-line variables. The specific process is as follows:

[0086] First, a non-iterative clearing algorithm for a unified electricity market based on sparse polynomial chaotic expansion is proposed, which mainly includes three steps: dataset construction, internal optimal cost calculation, and sparse polynomial chaotic expansion.

[0087] In order to explicitly represent the relationship between the clearing cost of the provincial power market and the tie-line variables, the first step is to construct a reasonable set of tie-line variables. Therefore, based on the historical operating data of the provincial power market, the following set of variables is proposed: (14)-(15)

[0088] Formula (14):

[0089] Formula (15):

[0090] In the formula, This represents the constructed dataset of connection variables; Indicates time period Data set of connection line variables at that time; Indicates time period The first time The set of tie-line variables consists of all tie-line variables in the provincial power market; Indicates time period Time The first case Each tie-line variable consists of power and voltage variables at the tie-line; I represents the number of tie-lines in the provincial power market.

[0091] Therefore, the above process corresponds to, for example Figure 2 Step S202, as shown, obtains the power supply period of the provincial power market and the number of tie lines in each region during the power supply period based on the historical operation data of the provincial power market in the unified power market; Step S204, extracts the power data and voltage data corresponding to each tie line in each power supply period from the historical operation data; Step S206, integrates the correspondence between multiple power data and voltage data and each tie line to obtain the tie line variable dataset.

[0092] The internal optimal cost calculation can be analyzed by equations (1) and (10). For the provincial power market, the clearing cost generally includes the operating cost of thermal power units, the standby cost, and the cost of purchasing and selling electricity. When the tie line variables are known, the clearing cost of the provincial power market can be solved by the optimal power flow problem. The specific solution model is as follows: Model (16):

[0093]

[0094]

[0095]

[0096]

[0097]

[0098]

[0099]

[0100]

[0101]

[0102]

[0103]

[0104] in, For the first Provincial electricity markets in different time periods under various tie-line variables The optimal clearing cost; For the first Thermal power units under various tie line variables During the period The unit output; , They represent the first Thermal power units under various tie line variables During the period The reduction and increase of reserve capacity; Indicates the first Time period under various connection line variables Electricity purchased; For the first Thermal power units under various tie line variables During the period Electricity sales output; For generator sets; , The first The standby cost coefficients for generators; , The first Time period under various connection line variables The output of wind and solar power; For the set of load nodes; For the first Under the condition of various tie line variables, the load node in the time period Load demand; , The first The upper and lower limits of the generator's output; , The first The upper and lower limits of the ramp rate of the generator; For the first The upper limit of power flow on each line; , , , , These are the power flow transfer factors for generators, wind power, photovoltaics, loads, and inter-provincial tie line nodes, respectively. , The first The upper and lower limits of the power output of the generator. After solving the above model, the following formula (17) can be constructed to form the set of tie-line variables and clearing costs: .

[0105] Furthermore, the sparse polynomial chaotic expansion is used to construct an explicit expression for the tie-line variables and the provincial power market clearing cost. A sparse basis function set is constructed using hyperbolic truncation, and the polynomial coefficients are solved using a subspace tracking algorithm. This effectively addresses the problems of limited fitting accuracy and insufficient computational efficiency of traditional algorithms. Specific steps include: Generalized polynomial chaotic expansion representing the clearing cost function: Based on the theory of generalized polynomial chaotic expansion, the clearing cost function can be approximately constructed as a combination of sparse orthogonal polynomials, as shown in formula (18): In the formula, The number of sparse basis functions is determined by the hyperbolic truncation; Let be the coefficients of the polynomial to be determined; The basis functions are convex orthogonal to ensure the convexity of the fitting results, and their specific expressions are shown in formula (19): In the formula, It is about of Legendre polynomials must satisfy orthogonality, i.e., formula (20): In the formula, For the first Uniform probability density of each link variable; For the Kronecker function, i.e. when When equals 0, Time equals 1; is a standardized constant.

[0106] In addition, the hyperbolic truncation method for constructing a sparse basis function set is to generate a sparse candidate polynomial set by using the hyperbolic truncation scheme, eliminating redundant high-order basis functions and alleviating the curse of dimensionality. The specific sparse basis set is shown in formula (21). In the formula, The truncation order is the polynomial order. The order of the expansion is used to control sparsity; For the first The order of the polynomials corresponding to the connection line variables, and, based on formula (21), the number of basis functions. This can be expressed as formula (22): .

[0107] Furthermore, the subspace tracking algorithm solves for the polynomial coefficients in order to obtain an explicit expression for the clearing cost function, and then further solves for the polynomial coefficients. The process is based on compressed sensing theory and uses sparse approximation vectors. To equivalently replace polynomial coefficients Furthermore, the subspace tracing algorithm is used to solve the problem, which significantly reduces the difficulty of the algorithm's solution. Sparse approximation vectors It is possible The optimization problem is solved using the following model (23): In the formula, Vector of clearing costs in the provincial electricity market; For the basis function matrix, each element ; For coefficient vector The clearing cost vector obtained by fitting; This is the residual threshold.

[0108] To solve the above model, a subspace tracking algorithm is proposed. The specific process includes: ① calculating the correlation between the basis function and the provincial electricity market clearing cost vector. ; ②Choose forward The set of basis function indices corresponding to the maximum absolute values As the initial input of the algorithm; ③ Calculate the initial residual using the following formulas (24)-(25); In the formula, The initial residual; for of List of subsets; for The pseudo-inverse can be calculated using the following formula (25): ④while( )or( )do: ⑤ Calculate the correlation between residuals and basis functions. ⑥ Choose forward The index set corresponding to the largest absolute value Combined with the previous set, we get ⑦ Projection calculation of temporary coefficients And filter the temporary coefficients before The index set corresponding to the largest absolute value ⑧ Update the residuals using formulas (24)-(25); ⑨ n=n+1; ⑩ Output the approximation coefficients This allows us to explicitly obtain the clearing cost function for each provincial electricity market.

[0109] Therefore, the above process corresponds to, for example Figure 3Step S302 shows obtaining the correlation between the sparse basis function set and the clearing cost, and obtaining the index set corresponding to at least one basis function whose absolute value of the correlation is greater than a preset value; Step S304, determining the initial residual of the index set, and continuously iterating to calculate the correlation between the residual and the basis function based on the initial residual to obtain the updated index set; Step S306, updating the residual based on the updated index set and the pre-obtained temporary coefficients and repeating the iteration; Step S308, stopping the iteration when the residual meets the preset threshold or the number of iterations reaches the preset number, and obtaining the polynomial coefficients of the sparse basis function set.

[0110] Based on the above, the hierarchical clearing model shown in the original formulas (1)-(13) is reconstructed into a single-layer clearing model to achieve non-iterative collaborative clearing. The specific model is as follows: Model (26):

[0111]

[0112] in, For provincial electricity market During the period The optimal clearing cost at that time; For provincial electricity market All clearing variables; For provincial electricity market The feasible region of the market clearing variable is defined. When the market clearing variable is within this feasible region, the market clearing constraint is satisfied. To clear relevant constraints in the model, including power flow constraints, generator constraints, etc.; , These represent the provincial electricity markets. The set of internal and boundary variables; , These are the power of the interconnection lines between adjacent provincial power markets, and the power balance of the interconnection lines needs to be satisfied. , These are the voltage angles of the provincial power market boundary busbars. It is the tie line reactance.

[0113] Therefore, the above process corresponds to, for example Figure 4 Step S402, as shown, processes the target power data based on a preset generalized polynomial chaotic expansion algorithm to construct a sparse basis function set; Step S404, based on tie-line variables and clearing costs, obtains the polynomial coefficients of the sparse basis function set through a preset subspace tracking algorithm; Step S406, based on the polynomial coefficients and the sparse basis function set, constructs explicit expressions for the tie-line variables and clearing costs.

[0114] In the above embodiments, the optimal transmission power of the interconnection lines between provincial power markets can be quickly obtained through the single-layer clearing model without multiple iterations, thereby achieving optimal coordinated clearing of provincial power markets and improving the overall solution efficiency.

[0115] More specifically, in one embodiment, a dynamic tensor decomposition algorithm is proposed, which can accurately represent the correlation between provincial power market clearing variables and tie-line variables. This transforms the feasible region of high-dimensional clearing variables in the original provincial power market clearing process into a low-dimensional feasible region of tie-line variables, thereby ensuring the feasibility of clearing schemes within the provincial power market. The dynamic tensor decomposition algorithm proposed in this application can map the feasible region of high-dimensional clearing variables in the provincial power market to a low-dimensional space without constraint linear relaxation. By decomposing the power market tensor into a sum of multiple terms, where each term is a convolution of a BM rank-1 tensor and a three-dimensional filter, the global and local spatiotemporal correlation features in the provincial power market are extracted using the BM rank-1 tensor and the three-dimensional filter, respectively. This decomposes the original high-dimensional power market clearing information into low-dimensional factors, thereby achieving efficient dimensionality reduction. Specific steps include:

[0116] First, a three-dimensional tensor is constructed based on the historical clearing data of the provincial electricity market. The specific process is shown in formulas (27)-(28). , In the formula, A three-dimensional tensor constructed from historical clearing data of the provincial electricity market; The spatial dimension of the data includes all nodes in the provincial electricity market, i.e., internal nodes. With boundary connection nodes ; The physical quantity dimension of the data includes generator output. Voltage angle wait; The time dimension of the data represents the scheduling time precision.

[0117] Then, the tensor feasible region is constructed according to the model constraints shown in formulas (2)-(9), and the specific process is as follows: formulas (29)-(33):

[0118] Formula (29):

[0119] Formula (30):

[0120] Formula (31):

[0121] Formula (32):

[0122] Formula (33):

[0123] In the formula, , for node The upper and lower limits of the uphill climb during different time periods; , for node Upper and lower limits of unit output during specific time periods; This is the power transfer distribution factor matrix; This represents the upper limit of transmission line capacity. for Nodes and Reactance of the line connected to the node; , The above constraints, which are the upper and lower limits of the node voltage angle, together constitute the tensor. feasible domain The feasible region is transformed from the physical constraints of the provincial electricity market clearing model. When the tensor satisfies this feasible region, the feasibility of provincial electricity market clearing can be guaranteed.

[0124] Therefore, the above process corresponds to: based on the node information, physical quantity information and time information included in the historical clearing data of the provincial power market obtained in advance, as well as the physical constraints of the provincial power market including ramp constraints, unit output constraints, power transmission distribution factor matrix, transmission line capacity upper limit, reactance of interconnected lines between nodes and upper and lower limits of node voltage angle, constructing a three-dimensional tensor and the tensor feasible region corresponding to the three-dimensional tensor.

[0125] Furthermore, a low-dimensional tensor mapping model for the provincial electricity market is constructed, as shown in the following formulas (34)-(36):

[0126] Formula (34):

[0127] Formula (35):

[0128] Formula (36):

[0129] In the formula, Let be the rank of the tensor decomposition. The larger the rank, the stronger the tensor decomposition's ability to express complex dynamic patterns, but the higher the computational complexity. The most suitable rank can be obtained through a parameter increment strategy. , , The low-dimensional factor obtained after tensor decomposition of equation (28) is denoted as , where This represents the spatial-temporal correlation factor, indicating the contribution of different nodes in different time slots. This is a time-dependent factor for physical quantities, representing the temporal coupling relationship between physical quantities. This is a spatial and physical quantity correlation factor, representing the coupling relationship between power and voltage angle; For three-dimensional filter factors, representing local physical constraints, the larger the weight, the stronger the correlation; It is less than Odd numbers; For regularization parameters; These are physical constraint terms.

[0130] Next, based on the solution of the mapping model by proximal alternation minimization, auxiliary variables are introduced. Furthermore, by using the semi-quadratic splitting method, the formula (34) can be reformulated as the following formula (37):

[0131]

[0132] In the formula, As the penalty parameter, the above model is solved based on the theory of proximal alternation minimization. The above model can be further transformed into the following model (38) form:

[0133]

[0134] in, The objective function of equation (37) is... , , , , , In the previous iteration Solution to the subproblem; For proximal parameters; The specific expression of the subproblem is shown in the following model (39):

[0135]

[0136] in, .

[0137] Iteratively solve the above The iteration ends when the solution to the subproblem converges to the critical point of formula (37). The convergence criteria include: ① lower half continuity: objective function It is true and lower half continuous; ②K-L condition judgment: objective function Satisfying the KL property; ③ Sufficient descent test: objective function ④ Satisfy the sufficient descent condition; ④ Relative error judgment: Result sequence Satisfying the relative error condition; ⑤ Boundedness judgment: Result sequence It is bounded. When all five conditions above are met, it proves that the current result has converged to the critical point, and the final result is output, that is, the low-dimensional factor obtained by dimensionality reduction.

[0138] Finally, the feasible region of low-dimensional tie-line variables is constructed, starting from the low-dimensional factors corresponding to the provincial electricity market clearing variables. The factors corresponding to the connection line variables are extracted, and the extraction results are shown in the following formula (40):

[0139]

[0140] Based on the aforementioned low-dimensional factors, the connection line variables can be reconstructed, as shown in the following formula (41):

[0141]

[0142] Through the above steps, it can be ensured that the provincial electricity market variables are determined by a single tie-line variable, i.e., formula (42): At this point, the feasible region of the high-dimensional clearing variables in the provincial power market is successfully mapped to the feasible region of the low-dimensional tie-line variables, i.e., formula (43):

[0143]

[0144] As can be seen from formula (42), when the tie-line variables of the provincial power market are within the feasible region shown in formula (43), there must be a corresponding power market variable to ensure the feasibility of the market clearing scheme. Therefore, through the above steps, the dynamic tensor decomposition algorithm proposed in this application can reduce the dimension of the high-dimensional clearing variable feasible region in the original provincial power market clearing process to a low-dimensional tie-line variable feasible region, thereby ensuring the feasibility of the provincial power market clearing scheme through the judgment of tie-line variables. On this basis, the unified power market single-layer clearing model constructed by model (26) can be further transformed into the following model (44):

[0145]

[0146] Through the model in (44), this application can achieve non-iterative collaborative clearing of the unified power market on the basis of ensuring the feasibility of the clearing scheme within the provincial power market, thereby improving the clearing efficiency of the unified power market.

[0147] Therefore, the above process corresponds to, for example Figure 5Step S502 shows the extraction of spatial and temporal correlation factors, physical quantity and temporal correlation factors, and spatial and physical quantity correlation factors of the low-dimensional mapping model corresponding to the three-dimensional tensor; Step S504 shows the solution of the tensor low-dimensional mapping model based on the preset proximal alternating minimization algorithm, and the output of low-dimensional factors when the result meets the preset convergence conditions; Step S506 shows the obtaining of the low-dimensional connection line variable feasible region based on the low-dimensional factors.

[0148] In summary, as Figure 6 The overall process of this application includes: ① constructing a tie-line dataset as shown in formula (14) based on historical operating data of the provincial power market; ② calculating the provincial power market clearing cost corresponding to the tie-line variables through model (16); ③ constructing an expression formula between the provincial power market clearing cost and the tie-line variables using sparse polynomial formula chaotic expansion technology, i.e., formula (18); ④ constructing a single-layer clearing model of the unified power market as shown in model (26) based on the clearing cost expression formula of the provincial power market, to realize non-iterative collaborative clearing between provincial power markets; ⑤ further constructing a tie-line dataset as shown in formula (18) based on historical data and physical constraints of the provincial power market. The three-dimensional tensor set and feasible region shown in equations (27)-(33); ⑥ Construct the provincial power market tensor low-dimensional mapping model as shown in equations (34)-(36) using the dynamic tensor algorithm; ⑦ Solve the tensor low-dimensional mapping model using the near-end alternation minimization algorithm shown in equations (37)-(39) to obtain the market low-dimensional factor after dimensional reduction; ⑧ Construct the feasible region of tie-line variables of the provincial power market using the low-dimensional factor obtained by dimensional reduction, i.e., equation (43); ⑨ Use the feasible region of tie-line variables obtained by equation (43) to improve the non-iterative collaborative clearing model of the original model (26), and finally obtain the single-layer clearing model as shown in equation (44). ⑩ Based on the single-layer clearing model shown in equation (44), the tie-line power of each province is quickly solved to achieve efficient clearing.

[0149] Through the above embodiments, this application proposes for the first time a non-iterative clearing algorithm for a unified electricity market based on sparse polynomial chaotic expansion and dynamic tensor decomposition. While ensuring the feasibility of clearing schemes within provincial electricity markets, it achieves non-iterative collaborative clearing between electricity markets. Furthermore, the non-iterative collaborative clearing algorithm proposed in this application effectively solves the problem of insufficient clearing efficiency in the unified electricity market, improves the overall clearing efficiency of the market, and is conducive to realizing the coordination of electricity resources between regions.

[0150] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0151] Based on the same inventive concept, this application also provides a unified electricity market clearing apparatus for implementing the clearing method of the unified electricity market described above. The solution provided by this apparatus is similar to the implementation described in the above method; therefore, the specific limitations in one or more embodiments of the unified electricity market clearing apparatus provided below can be found in the limitations of the unified electricity market clearing method described above, and will not be repeated here.

[0152] In one exemplary embodiment, such as Figure 7 As shown, a clearing device for a unified electricity market is provided, comprising: a dataset construction module 701, a data acquisition module 702, a data processing module 703, and a result acquisition module 704, wherein:

[0153] The dataset construction module 701 is used to construct a dataset of tie-line variables for provincial power markets, which includes at least power and voltage variables at tie-line locations, based on historical operating data of provincial power markets in the unified power market.

[0154] The data acquisition module 702 is used to obtain the current clearing cost of the provincial power market based on each tie-line variable in the tie-line variable dataset and the current clearing conditions of the provincial power market.

[0155] The data processing module 703 is used to construct explicit expressions for tie-line variables and clearing costs based on a preset sparse polynomial chaotic expansion algorithm, and to reconstruct the preset hierarchical clearing model of the unified electricity market into a single-layer clearing model based on the explicit expressions.

[0156] The data processing module 703 is also used to construct a three-dimensional tensor and the corresponding tensor feasible region based on the historical clearing data and physical constraints of the provincial power market obtained in advance. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie-line variable feasible region.

[0157] The result acquisition module 704 is used to optimize the single-layer clearing model based on the feasible region of low-dimensional tie-line variables to obtain the clearing result of the unified electricity market.

[0158] Furthermore, in one embodiment, the data processing module 703 is also used to process the target power data based on a preset generalized polynomial chaotic expansion algorithm to construct a sparse basis function set; based on the tie-line variables and clearing costs, to obtain the polynomial coefficients of the sparse basis function set through a preset subspace tracking algorithm; and based on the polynomial coefficients and the sparse basis function set, to construct an explicit expression for the tie-line variables and clearing costs.

[0159] Furthermore, in one embodiment, the data processing module 703 is also used to obtain the correlation between the sparse basis function set and the clearing cost, and obtain the index set corresponding to at least one basis function whose absolute value of the correlation is greater than a preset value; determine the initial residual of the index set, and continuously iterate to calculate the correlation between the residual and the basis function based on the initial residual to obtain an updated index set; update the residual based on the updated index set and the pre-obtained temporary coefficients and repeat the iteration; stop the iteration when the residual meets a preset threshold or the number of iterations reaches a preset number, and obtain the polynomial coefficients of the sparse basis function set.

[0160] Furthermore, in one embodiment, the data processing module 703 is also used to construct a three-dimensional tensor and the corresponding tensor feasible region based on the node information, physical quantity information and time information included in the historical clearing data of the provincial power market obtained in advance, as well as the physical constraints of the provincial power market including ramp constraints, unit output constraints, power transmission distribution factor matrix, transmission line capacity upper limit, reactance of interconnected lines between nodes and upper and lower limits of node voltage angle.

[0161] Furthermore, in one embodiment, the data processing module 703 is also used to extract the spatial and temporal correlation factors, physical quantity and temporal correlation factors, and spatial and physical quantity correlation factors of the low-dimensional mapping model corresponding to the three-dimensional tensor; solve the low-dimensional mapping model of the tensor based on a preset proximal alternating minimization algorithm; output the low-dimensional factors when the results meet the preset convergence conditions; and obtain the feasible region of the low-dimensional connection line variables based on the low-dimensional factors.

[0162] Furthermore, in one embodiment, the dataset construction module 701 is also used to obtain the power supply period of the provincial power market and the number of interconnecting lines in each region during the power supply period based on the historical operation data of the provincial power market in the unified power market; extract the power data and voltage data corresponding to each interconnecting line during each power supply period from the historical operation data; and integrate the correspondence between multiple power data and voltage data and each interconnecting line to obtain the interconnecting line variable dataset.

[0163] The modules in the aforementioned clearing device for the unified electricity market can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0164] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores clearing data for a unified electricity market. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a clearing method for a unified electricity market.

[0165] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0166] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0167] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps in the above method embodiments.

[0168] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0169] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0170] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0171] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0172] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method of clearing a unified electricity market, characterized by, The method includes: Based on the historical operation data of the provincial power markets in the unified power market, a tie-line variable dataset is constructed for the provincial power markets, which includes at least the power and voltage variables at the tie-line points. Based on each tie-line variable in the tie-line variable dataset, and combined with the current clearing conditions of the provincial power market, the current clearing cost of the provincial power market is obtained; An explicit expression for the tie-line variables and the clearing cost is constructed based on a preset sparse polynomial chaotic expansion algorithm. Based on the explicit expression, the preset hierarchical clearing model of the unified electricity market is reconstructed into a single-layer clearing model. Based on the historical clearing data and physical constraints of the provincial power market obtained in advance, a three-dimensional tensor and the corresponding tensor feasible region are constructed. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie-line variable feasible region. The single-layer clearing model is optimized based on the feasible region of the low-dimensional tie-line variables to obtain the clearing result of the unified electricity market.

2. The method of claim 1, wherein, The explicit expression for the connection variables and the clearing cost, constructed based on a preset sparse polynomial chaotic expansion algorithm, includes: The target power data is processed based on a pre-defined generalized polynomial chaotic expansion algorithm, and a sparse basis function set is constructed. Based on the connection line variables and the clearing cost, the polynomial coefficients of the sparse basis function set are obtained through a preset subspace tracking algorithm; Based on the polynomial coefficients and the sparse basis function set, an explicit expression for the connection line variable and the clearing cost is constructed.

3. The method of claim 2, wherein, The step of obtaining the polynomial coefficients of the sparse basis function set through a preset subspace tracing algorithm includes: Obtain the correlation between the sparse basis function set and the clearing cost, and obtain the index set corresponding to at least one basis function whose absolute value of the correlation is greater than a preset value; Determine the initial residual of the index set, and continuously iterate the correlation between the residual and the basis function based on the initial residual to obtain the updated index set; The residual is updated based on the updated index set and the pre-acquired temporary coefficients, and the iteration is repeated. When the residual satisfies a preset threshold or the number of iterations reaches a preset number, the iteration stops, and the polynomial coefficients of the sparse basis function set are obtained.

4. The method of claim 1, wherein, The construction of a three-dimensional tensor and its corresponding tensor feasible region based on pre-acquired historical clearing data and physical constraints of the provincial power market includes: Based on the node information, physical quantity information, and time information included in the historical clearing data of the provincial power market obtained in advance, as well as the physical constraints of the provincial power market including ramp constraints, unit output constraints, power transmission distribution factor matrix, transmission line capacity upper limit, reactance of inter-node connecting lines, and upper and lower limits of node voltage angle, the three-dimensional tensor and the corresponding tensor feasible region are constructed.

5. The method of claim 4, wherein, The step of using the three-dimensional tensor and the corresponding tensor feasible region, combined with a preset dynamic tensor decomposition algorithm, to perform dimensionality reduction processing on the current high-dimensional clearing variable feasible region, yielding a low-dimensional connection variable feasible region, including: Extract the spatial and temporal correlation factors, physical quantity and temporal correlation factors, and spatial and physical quantity correlation factors of the low-dimensional mapping model corresponding to the three-dimensional tensor; The tensor low-dimensional mapping model is solved based on a preset proximal alternating minimization algorithm, and a low-dimensional factor is output when the result meets a preset convergence condition. The feasible region of the low-dimensional connection line variables is obtained based on the low-dimensional factor.

6. The method of claim 1, wherein, Based on historical operational data from provincial power markets within a unified power market, a tie-line variable dataset is constructed for the provincial power market, including at least power and voltage variables at tie-line locations. Based on historical operational data of provincial power markets within a unified power market, the power supply periods of the provincial power markets and the number of regional interconnection lines during the power supply periods are obtained. Extract the power and voltage data corresponding to each tie line during each power supply period from the historical operation data; By integrating the correspondence between multiple power and voltage data and each of the tie lines, the tie line variable dataset is obtained.

7. An apparatus for clearing a unified electricity market, characterized by: The device includes: The dataset construction module is used to construct a tie-line variable dataset for the provincial power market, which includes at least power and voltage variables at tie-line locations, based on historical operating data of the provincial power markets in the unified power market. The data acquisition module is used to obtain the current clearing cost of the provincial power market based on each tie-line variable in the tie-line variable dataset and the current clearing conditions of the provincial power market. The data processing module is used to construct an explicit expression for the tie-line variables and the clearing cost based on a preset sparse polynomial chaotic expansion algorithm, and to reconstruct the preset hierarchical clearing model of the unified electricity market into a single-layer clearing model based on the explicit expression. The data processing module is also used to construct a three-dimensional tensor and the corresponding tensor feasible region based on the historical clearing data and physical constraints of the provincial power market obtained in advance. Based on the three-dimensional tensor and the corresponding tensor feasible region, the current high-dimensional clearing variable feasible region is reduced in dimensionality by combining a preset dynamic tensor decomposition algorithm to obtain a low-dimensional tie-line variable feasible region. The result acquisition module is used to optimize the single-layer clearing model based on the feasible domain of the low-dimensional tie-line variables to obtain the clearing result of the unified electricity market. 8.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-7. When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images