A linearized optimal power flow generation method and device based on measurement drive

Abstract

The present invention proposes a method and device for generating a linearized optimal power flow based on measurement drive, which belongs to the field of power system operation and control technology. The method includes: constructing a linearized optimal power flow regression model, the constraints of the model include: linearized power flow anti-difference model constraints and linearized node voltage amplitude constraints; identifying the regression coefficients of the linearized power flow anti-difference model constraints and the linearized node voltage amplitude constraints according to historical measurement sections; solving the linearized optimal power flow regression model to generate the optimal power flow according to the identification results of the regression coefficients. The present invention does not rely on the network parameter model, and overcomes the node power mismatch problem that exists when the existing data-driven linearized power flow model is used for optimal power flow calculation, effectively improving the safety of power system operation.

Description

Technical Field

[0001] The present invention belongs to the technical field of power system operation and control, and in particular relates to a linearized optimal power flow generation method and device based on measurement drive. Background Art

[0002] Power system optimal power flow is the process of finding the optimal power flow distribution that satisfies all specified constraints and optimizes one or more system performance indicators by optimizing control variables, while satisfying system operational and safety constraints. Power system optimal power flow is the foundation for optimized distribution network operation. It guides the safety, economy, and reliability of power system operation and has become an indispensable network analysis and optimization tool.

[0003] The nonlinearity of the traditional AC power flow equation leads to the non-convexity of the optimal power flow model. In practice, the distribution network has many devices and covers a wide range, and it is difficult to obtain an accurate physical model. The linearized power flow model based on the physical model is difficult to obtain accurate and feasible results when used for optimal power flow calculations; although the data-driven linearized power flow model does not require precise model parameters, it ignores Kirchhoff's current law during the modeling process. When used for optimal power flow calculations, it will lead to infeasible solutions, that is, when operating according to the optimal power flow results, the system will have overvoltage or overcurrent problems; in addition, some literature uses neural network training to generate the relationship between node injection power and the optimal solution, but it requires a large number of historical optimal solution samples, which are often difficult to obtain in actual engineering. Summary of the Invention

[0004] The present invention aims to overcome the shortcomings of existing technologies by proposing a measurement-driven, linearized optimal power flow calculation method and apparatus. Based on system measurement data, the present invention constructs a linearized power flow model and embeds Kirchhoff's current law constraints. This method is independent of network parameter models and overcomes the node power mismatch problem that exists when existing data-driven, linearized power flow models are used for optimal power flow calculations, effectively improving the safety of power system operation.

[0005] The first embodiment of the present invention provides a linearized optimal power flow generation method based on measurement drive, comprising:

[0006] Constructing a linearized optimal power flow regression model, wherein the constraints of the model include: linearized power flow robustness model constraints and linearized node voltage amplitude constraints;

[0007] Identifying the regression coefficients of the linearized power flow robustness model constraint and the linearized node voltage amplitude constraint based on historical measurement sections;

[0008] According to the identification result of the regression coefficient, the linearized optimal power flow regression model is solved to generate the optimal power flow.

[0009] In a specific embodiment of the present invention, the objective function expression of the linearized optimal power flow regression model is as follows:

[0010]

[0011] Where, Represents the main grid and distributed power generation The prime minister has made contributions.

[0012] In a specific embodiment of the present invention, the constraint expression of the linearized power flow robustness model is as follows:

[0013]

[0014]

[0015] Where, It is the phase sequence, They are The phase nodes inject active power vector and reactive power vector, For the main grid and distributed power generation The phase has no effort to exert, Including main grid and distributed power supply Phase active load and reactive load; They are Active power vector and reactive power vector of phase line They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix;

[0016] The linearized node voltage amplitude constraint expression is as follows:

[0017]

[0018] Where, for Phase voltage magnitude square vector, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix;

[0019] in,

[0020]

[0021]

[0022] Where, They are The lower and upper limits of Is the root node Phase voltage amplitude; for Predefined voltage reference values ​​for the phases.

[0023] In a specific embodiment of the present invention, the constraints of the linearized optimal power flow regression model further include:

[0024] Kirchhoff's Current Law Constraints:

[0025]

[0026]

[0027] Where, for The node-branch correlation matrix of the phase; for Phase node-inflow branch correlation matrix is the Hadamard product;

[0028] System power balance constraints:

[0029]

[0030]

[0031] Where i is the bus number, bi is the line number; They are Active power loss vector and reactive power loss vector of the phase line;

[0032] Line operating capacity constraints:

[0033]

[0034]

[0035]

[0036]

[0037] Where, It's a line Phase transfer capacity vector;

[0038] Generator output constraints:

[0039]

[0040]

[0041] Where, They are The lower and upper limits of They are lower and upper limits.

[0042] In a specific embodiment of the present invention, identifying the regression coefficients of the linearized power flow robustness model constraint and the linearized node voltage amplitude constraint based on the historical measurement section includes:

[0043] 1) Obtain n historical measurement sections of the distribution network as original measurement samples to form an original measurement sample set denoted as X n , n>0, Among them, the jth original measurement sample x j include:

[0044] 2) Using principal component analysis to analyze the original measurement sample set X n The measurements selected as regression independent variables are subjected to feature extraction and dimensionality reduction, including:

[0045] Select independent variables from the original measurement sample set and construct the independent variable set is the independent variable of the jth original measurement sample;

[0046] right After standardization, the expression is as follows:

[0047]

[0048] Where, for The expectation of The standard deviation of

[0049] After standardization, we get The standardized set is recorded as

[0050] calculate The covariance matrix of

[0051] Covariance matrix Perform eigenvalue decomposition and take the unit eigenvectors v1, v2, ..., v corresponding to the largest m eigenvalues m , we get the projection matrix V=(v1,v2,...,v m ) and the updated set of independent variables

[0052] 3) Establishing a parameter identification model based on the features extracted in step 2) and the independent variables after dimensionality reduction;

[0053]

[0054] sty i -w T x i -b≤ε+ξ i ,w T x i +by i ≤ε+ξ i * ,ξ i ,ξ i * ≥0

[0055] in, is the model parameter to be identified; ξ i 、 is a variable; y i is the dependent variable selected from the original measurement sample set, β, ε, and λ are hyperparameters obtained by S-fold cross validation;

[0056] 4) Solve the parameter identification model established in step 3) and obtain The optimal solution of .

[0057] In a specific embodiment of the present invention, the solution of the linearized optimal power flow regression model adopts an interior point method, and the optimal power flow includes: The optimal solution of .

[0058] In a specific embodiment of the present invention, the independent variable is selected as The dependent variable is selected as and

[0059] A second embodiment of the present invention provides a linearized optimal power flow generation method based on measurement drive, comprising:

[0060] Obtain active power and reactive power parameters of nodes in the power system, and useful power and reactive power parameters of lines;

[0061] The parameters are input into a preset linearized optimal power flow regression model, and the model outputs the generation result of the optimal power flow; wherein, the constraints of the model include: linearized power flow anti-difference model constraints and linearized node voltage amplitude constraints, and the regression coefficients of the linearized power flow anti-difference model constraints and the linearized node voltage amplitude constraints are obtained by identifying historical measurement sections.

[0062] A third embodiment of the present invention provides a linearized optimal power flow generation device based on measurement drive, comprising:

[0063] A power flow regression model building module is used to build a linearized optimal power flow regression model, wherein the constraints of the model include: linearized power flow robustness model constraints and linearized node voltage amplitude constraints;

[0064] A regression coefficient identification module, used to identify the regression coefficients of the linearized power flow robustness model constraint and the linearized node voltage amplitude constraint based on historical measurement sections;

[0065] The optimal power flow generation module is used to solve the linearized optimal power flow regression model according to the identification result of the regression coefficient to generate the optimal power flow.

[0066] In a specific embodiment of the present invention, the objective function expression of the linearized optimal power flow regression model is as follows:

[0067]

[0068] Where, Represents the main grid and distributed power generation The prime minister has made contributions.

[0069] In a specific embodiment of the present invention, the constraint expression of the linearized power flow robustness model is as follows:

[0070]

[0071]

[0072] Where, It is the phase sequence, They are The phase nodes inject active power vector and reactive power vector, For the main grid and distributed power generation The phase has no effort to exert, Including main grid and distributed power supply Phase active load and reactive load; They are Active power vector and reactive power vector of phase line They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix;

[0073] The linearized node voltage amplitude constraint expression is as follows:

[0074]

[0075] Where, for Phase voltage magnitude square vector, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix;

[0076] in,

[0077]

[0078]

[0079] Where, They are The lower and upper limits of Is the root node Phase voltage amplitude; for Predefined voltage reference values ​​for the phases.

[0080] In a specific embodiment of the present invention, the constraints of the linearized optimal power flow regression model further include:

[0081] Kirchhoff's Current Law Constraints:

[0082]

[0083]

[0084] Where, for The node-branch correlation matrix of the phase; for Phase node-inflow branch correlation matrix is the Hadamard product;

[0085] System power balance constraints:

[0086]

[0087]

[0088] Where i is the bus number, bi is the line number; They are Active power loss vector and reactive power loss vector of the phase line;

[0089] Line operating capacity constraints:

[0090]

[0091]

[0092]

[0093]

[0094] Where, It's a line Phase transfer capacity vector;

[0095] Generator output constraints:

[0096]

[0097]

[0098] Where, They are The lower and upper limits of They are lower and upper limits.

[0099] In a specific embodiment of the present invention, identifying the regression coefficients of the linearized power flow robustness model constraint and the linearized node voltage amplitude constraint based on the historical measurement section includes:

[0100] 1) Obtain n historical measurement sections of the distribution network as original measurement samples to form an original measurement sample set denoted as X n , n>0, Among them, the jth original measurement sample x j include:

[0101] 2) Using principal component analysis to analyze the original measurement sample set X n The measurements selected as regression independent variables are subjected to feature extraction and dimensionality reduction, including:

[0102] Select independent variables from the original measurement sample set and construct the independent variable set is the independent variable of the jth original measurement sample;

[0103] right After standardization, the expression is as follows:

[0104]

[0105] Where, for The expectation of The standard deviation of

[0106] After standardization, we get The standardized set is recorded as

[0107] calculate The covariance matrix of

[0108] Pair covariance matrix Perform eigenvalue decomposition and take the unit eigenvectors v1, v2, ..., v corresponding to the largest m eigenvalues m , we get the projection matrix V=(v1,v2,...,v m ) and the updated set of independent variables

[0109] 3) Establishing a parameter identification model based on the features extracted in step 2) and the independent variables after dimensionality reduction;

[0110]

[0111] sty i -w T x i -b≤ε+ξ i ,w T x i +by i ≤ε+ξ i * ,ξ i ,ξ i * ≥0

[0112] in, is the model parameter to be identified; ξ i 、 is a variable; y i is the dependent variable selected from the original measurement sample set, β, ε, and λ are hyperparameters obtained by S-fold cross validation;

[0113] 4) Solve the parameter identification model established in step 3) and obtain The optimal solution of .

[0114] In a specific embodiment of the present invention, the solution of the linearized optimal power flow regression model adopts an interior point method, and the optimal power flow includes: The optimal solution of .

[0115] In a specific embodiment of the present invention, the independent variable is selected as The dependent variable is selected as and

[0116] A fourth embodiment of the present invention provides a linearized optimal power flow generation device based on measurement drive, comprising:

[0117] Parameter acquisition module, used to obtain the active power and reactive power parameters of nodes in the power system, and the useful power and reactive power parameters of lines;

[0118] The optimal power flow generation module is used to input the parameters into a preset linearized optimal power flow regression model, and the model outputs the generation result of the optimal power flow; wherein, the constraints of the model include: linearized power flow anti-difference model constraints and linearized node voltage amplitude constraints, and the regression coefficients of the linearized power flow anti-difference model constraints and the linearized node voltage amplitude constraints are obtained by identifying historical measurement sections.

[0119] A fifth embodiment of the present invention provides an electronic device, including:

[0120] at least one processor; and a memory communicatively coupled to the at least one processor;

[0121] The memory stores instructions that can be executed by the at least one processor, and the instructions are configured to execute the above-mentioned linearized optimal power flow generation method based on measurement drive.

[0122] A sixth aspect of the present invention provides a computer-readable storage medium storing computer instructions for enabling the computer to execute the aforementioned measurement-driven linearized optimal power flow generation method.

[0123] The characteristics and beneficial effects of the present invention are:

[0124] 1. This paper proposes for the first time a linearized optimal power flow model without node power mismatch. By converting the optimal power flow into a convex optimization problem and strictly satisfying Kirchhoff's current law, it solves the node power mismatch problem existing in existing data-driven linearized power flow models.

[0125] 2. The linearized optimal power flow model proposed in the present invention does not require the use of system network parameters, but only needs to maintain the system network topology, thus solving the problem of no model and difficulty in obtaining network parameters.

[0126] 3. This invention eliminates node power mismatch issues in its application, avoids ignoring Kirchhoff's current law during the linearized power flow model regression process, and avoids overvoltage or overcurrent issues when the system operates according to the optimal power flow results, thereby improving the safety of power system operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0127] Figure 1 This is an overall flow chart of a linearized optimal power flow generation method based on measurement drive in an embodiment of the present invention. DETAILED DESCRIPTION

[0128] The present invention proposes a method and device for generating a linearized optimal power flow based on measurement drive, which is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0129] The first embodiment of the present invention provides a linearized optimal power flow generation method based on measurement drive, comprising:

[0130] In a specific embodiment of the present invention, the linear optimal power flow generation method based on measurement drive has the following overall process: Figure 1 As shown, the following steps are included:

[0131] 1) Establish a linearized optimal power flow regression model embedded with Kirchhoff's current law (KCL) constraints. The specific steps are as follows:

[0132] 1-1) Establish the objective function of the linearized optimal power flow regression model;

[0133] In a specific embodiment of the present invention, the objective function is to minimize the total power generation cost, which is expressed as follows:

[0134]

[0135] Where, Represents the main grid and distributed power generation The Prime Minister has made contributions;

[0136] 1-2) Establish the constraints of the linearized optimal power flow regression model, as follows:

[0137] 1-2-1) Constraints of the linearized power flow robustness model:

[0138]

[0139]

[0140] Where, It is the phase sequence, They are The phase nodes inject active power vector and reactive power vector, Including the main grid and distributed generation (DGs) Phase active power output and reactive power output, Including main grid and distributed power supply Phase active load and reactive load. They are The active power vector and reactive power vector of the phase line, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix.

[0141] 1-2-2) Linearized node voltage amplitude constraint:

[0142]

[0143] Where, for Phase voltage magnitude square vector, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix.

[0144] Also need to meet:

[0145]

[0146]

[0147] Where, They are The lower and upper limits of Is the root node of the predefined reference value Phase voltage amplitude; for The predefined voltage reference value of the phase is taken as 1 in this embodiment.

[0148] 1-2-3) Kirchhoff's Current Law (KCL) Constraints:

[0149]

[0150]

[0151] Where, for The node-branch correlation matrix of the phase, which represents the network topology; for Phase node-inflow branch correlation matrix in is the Hadamard product.

[0152] 1-2-4) System power balance constraints:

[0153]

[0154]

[0155] Where i is the bus number, bi is the line number; They are Active power loss vector and reactive power loss vector of the phase line.

[0156] 1-2-5) Line operation capacity constraints:

[0157]

[0158]

[0159]

[0160]

[0161] Where, It's a line Phase transfer capacity vector.

[0162] 1-2-6) Generator output constraints:

[0163]

[0164]

[0165] Where, They are The lower and upper limits of They are lower and upper limits.

[0166] The objective function formula (1) in step 1-1) and the constraint formulas (2) to (14) in step 1-2) form the linearized optimal power flow regression model of this embodiment.

[0167] 2) Identify the regression coefficient matrix in step 1) based on historical measurement sections;

[0168] In this embodiment, the robust regression method combining support vector regression and ridge regression is used to identify the model parameters in steps 1-2-1) and 1-2-2). The specific steps are as follows:

[0169] 2-1) Obtain n historical measurement sections of the distribution network as original measurement samples to form an original measurement sample set denoted as X n , n>0, in this embodiment n=1571, The jth original measurement sample x i Including but not limited to

[0170] 2-2) Based on principal component analysis (PCA) and orthogonal transformation, the original measurement sample set X n The measurements selected as regression independent variables are subjected to feature extraction and dimensionality reduction to obtain measurement data without collinearity issues. The specific steps are as follows:

[0171] 2-2-1) Select independent variables from the original measurement sample set and construct the independent variable set is the independent variable of the jth original measurement sample.

[0172] In a specific embodiment of the present invention, the independent variable is selected as To build a set of independent variables The independent variable of the jth original measurement sample is in is the jth measurement section The active power vector injected into the phase node is, is the jth measurement section The reactive power vector injected into the phase node is

[0173] 2-2-2) Yes After standardization, the expression is as follows:

[0174]

[0175] Where, for The expectation of After standardization, we get The standardized set is recorded as

[0176] 2-2-3) Calculate the updated The covariance matrix of

[0177] 2-2-4) Covariance matrix Perform eigenvalue decomposition and take the unit eigenvectors v1, v2, ..., v corresponding to the largest m eigenvalues m ; In this embodiment, the values ​​of m are ((41,70,70), (41,70,70), (41,70,70)) corresponding to the three phases respectively and and is the dependent variable selected in this embodiment.

[0178] 2-2-5) Output projection matrix V = (v1, v2, ..., v m ) and the updated set of independent variables That is, feature extraction and dimensionality reduction processing of the selected independent variable measurement are realized.

[0179] 2-3) Based on the results of step 2-2), a parameter identification model is established as follows:

[0180]

[0181] in, are all model parameters to be identified; ξ i 、 is a variable, obtained by solving the model; y i is the dependent variable selected from the original measurement sample set, β, ε, and λ are hyperparameters, which can be obtained by S-fold cross validation. In a specific embodiment of the present invention, the sample data is divided into s=10 parts in the S-fold cross validation.

[0182] 2-4) The parameter identification model established in step 2-3) is a quadratic programming (QP) problem, which can be solved using solvers such as Cplex and Gurobi. The optimal solution of .

[0183] 3) Based on the regression coefficient matrix identification results obtained in step 2), the interior point method is used to solve the linearized optimal power flow regression model established in step 1) to obtain the system node voltage Line active power Line reactive power Line active power loss Line reactive loss The optimal solution is to generate a linearized optimal power flow based on measurement drive.

[0184] It should be noted that the independent variables and dependent variables in step 2) of this embodiment are not limited to Both the independent variables and the dependent variables can be determined based on actual quantity measurements and demand; the objective function of the linearized optimal power flow regression model is not limited to formula (1) and can be set as needed to make a certain performance indicator of the system (such as minimum power generation cost) reach the optimal value.

[0185] To implement the above embodiment, a second embodiment of the present invention proposes a linearized optimal power flow generation method based on measurement drive, including:

[0186] Obtain active power and reactive power parameters of nodes in the power system, and useful power and reactive power parameters of lines;

[0187] The parameters are input into a preset linearized optimal power flow regression model, and the model outputs the generation result of the optimal power flow; wherein, the constraints of the model include: linearized power flow anti-difference model constraints and linearized node voltage amplitude constraints, and the regression coefficients of the linearized power flow anti-difference model constraints and the linearized node voltage amplitude constraints are obtained by identifying historical measurement sections.

[0188] It should be noted that the aforementioned explanation of the embodiment of a linearized optimal power flow generation method based on measurement drive in the first aspect is also applicable to the linearized optimal power flow generation method based on measurement drive in this embodiment, and will not be repeated here.

[0189] To implement the above embodiment, a third embodiment of the present invention provides a linearized optimal power flow generation device based on measurement drive, comprising:

[0190] A power flow regression model building module is used to build a linearized optimal power flow regression model, wherein the constraints of the model include: linearized power flow robustness model constraints and linearized node voltage amplitude constraints;

[0191] A regression coefficient identification module, used to identify the regression coefficients of the linearized power flow robustness model constraint and the linearized node voltage amplitude constraint based on historical measurement sections;

[0192] The optimal power flow generation module is used to solve the linearized optimal power flow regression model according to the identification result of the regression coefficient to generate the optimal power flow.

[0193] In a specific embodiment of the present invention, the power flow regression model building module is specifically used to establish a linearized optimal power flow regression model embedded with Kirchhoff's current law (KCL) constraints, and the specific steps are as follows:

[0194] 1) Establish a linearized optimal power flow regression model embedded with Kirchhoff's current law (KCL) constraints. The specific steps are as follows:

[0195] 1-1) Establish the objective function of the linearized optimal power flow regression model;

[0196] In a specific embodiment of the present invention, the objective function is to minimize the total power generation cost, which is expressed as follows:

[0197]

[0198] Where, Represents the main grid and distributed power generation The Prime Minister has made contributions;

[0199] 1-2) Establish the constraints of the linearized optimal power flow regression model, as follows:

[0200] 1-2-1) Constraints of the linearized power flow robustness model:

[0201]

[0202]

[0203] Where, It is the phase sequence, They are The phase nodes inject active power vector and reactive power vector, Including the main grid and distributed generation (DGs) Phase active power output and reactive power output, Including main grid and distributed power supply Phase active load and reactive load. They are The active power vector and reactive power vector of the phase line, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix.

[0204] 1-2-2) Linearized node voltage amplitude constraint:

[0205]

[0206] Where, for Phase voltage magnitude square vector, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix.

[0207] Also need to meet:

[0208]

[0209]

[0210] Where, They are The lower and upper limits of Is the root node of the predefined reference value Phase voltage amplitude; for The predefined voltage reference value of the phase is taken as 1 in this embodiment.

[0211] 1-2-3) Kirchhoff's Current Law (KCL) Constraints:

[0212]

[0213]

[0214] Where, for The node-branch correlation matrix of the phase, which represents the network topology; for Phase node-inflow branch correlation matrix in is the Hadamard product.

[0215] 1-2-4) System power balance constraints:

[0216]

[0217]

[0218] Where i is the bus number, bi is the line number; They are Active power loss vector and reactive power loss vector of the phase line.

[0219] 1-2-5) Line operation capacity constraints:

[0220]

[0221]

[0222]

[0223]

[0224] Where, It's a line Phase transfer capacity vector.

[0225] 1-2-6) Generator output constraints:

[0226]

[0227]

[0228] Where, They are The lower and upper limits of They are lower and upper limits.

[0229] The objective function formula (1) in step 1-1) and the constraint formulas (2) to (14) in step 1-2) form the linearized optimal power flow regression model of this embodiment.

[0230] In a specific embodiment of the present invention, the regression coefficient identification module is specifically used to identify the regression coefficient matrix based on historical measurement sections.

[0231] In this embodiment, the robust regression method combining support vector regression and ridge regression is used to identify the model parameters. The specific steps are as follows:

[0232] 2-1) Obtain n historical measurement sections of the distribution network as original measurement samples to form an original measurement sample set denoted as X n , n>0, in this embodiment n=1571, The jth original measurement sample x i Including but not limited to

[0233] 2-2) Based on principal component analysis (PCA) and orthogonal transformation, the original measurement sample set X n The measurements selected as regression independent variables are subjected to feature extraction and dimensionality reduction to obtain measurement data without collinearity issues. The specific steps are as follows:

[0234] 2-2-1) Select independent variables from the original measurement sample set and construct the independent variable set is the independent variable of the jth original measurement sample.

[0235] In a specific embodiment of the present invention, the independent variable is selected as To build a set of independent variables The independent variable of the jth original measurement sample is in is the jth measurement section The active power vector injected into the phase node is, is the jth measurement section The reactive power vector injected into the phase node is,

[0236] 2-2-2) Yes After standardization, the expression is as follows:

[0237]

[0238] Where, for The expectation of After standardization, we get The standardized set is recorded as

[0239] 2-2-3) Calculate the updated The covariance matrix of

[0240] 2-2-4) Covariance matrix Perform eigenvalue decomposition and take the unit eigenvectors v1, v2, ..., v corresponding to the largest m eigenvalues m ; In this embodiment, the values ​​of m are ((41,70,70), (41,70,70), (41,70,70)) corresponding to the three phases respectively and and is the dependent variable selected in this embodiment.

[0241] 2-2-5) Output projection matrix V = (v1, v2, ..., v m ) and the updated set of independent variables That is, feature extraction and dimensionality reduction processing of the selected independent variable measurement are realized.

[0242] 2-3) Based on the results of step 2-2), a parameter identification model is established as follows:

[0243]

[0244] in, are all model parameters to be identified; ξ i 、 is a variable, obtained by solving the model; y i is the dependent variable selected from the original measurement sample set, β, ε, and λ are hyperparameters, which can be obtained by S-fold cross validation. In a specific embodiment of the present invention, the sample data is divided into s=10 parts in the S-fold cross validation.

[0245] 2-4) The parameter identification model established in step 2-3) is a quadratic programming (QP) problem, which can be solved using solvers such as Cplex and Gurobi. The optimal solution of .

[0246] In a specific embodiment of the present invention, the optimal power flow generation module is specifically used to solve the linearized optimal power flow regression model using the interior point method according to the regression coefficient matrix identification result to obtain the system node voltage Line active power Line reactive power Line active power loss Line reactive loss The optimal solution is to generate a linearized optimal power flow based on measurement drive.

[0247] According to an embodiment of the present invention, a measurement-driven linearized optimal power flow generation device is proposed. By constructing a linearized optimal power flow regression model, the model's constraints include: linearized power flow robustness model constraints and linearized node voltage amplitude constraints; identifying the regression coefficients of the linearized power flow robustness model constraints and the linearized node voltage amplitude constraints based on historical measurement sections; and solving the linearized optimal power flow regression model based on the identified regression coefficients to generate the optimal power flow. This achieves independence from network parameter models and overcomes the node power mismatch problem that exists when existing data-driven linearized power flow models are used for optimal power flow calculations, effectively improving the safety of power system operation.

[0248] To implement the above embodiment, a fourth embodiment of the present invention provides a linearized optimal power flow generation device based on measurement drive, comprising:

[0249] Parameter acquisition module, used to obtain the active power and reactive power parameters of nodes in the power system, and the useful power and reactive power parameters of lines;

[0250] The optimal power flow generation module is used to input the parameters into a preset linearized optimal power flow regression model, and the model outputs the generation result of the optimal power flow; wherein, the constraints of the model include: linearized power flow anti-difference model constraints and linearized node voltage amplitude constraints, and the regression coefficients of the linearized power flow anti-difference model constraints and the linearized node voltage amplitude constraints are obtained by identifying historical measurement sections.

[0251] It should be noted that the above explanation of the embodiment of a linearized optimal power flow generation method based on measurement drive in the first aspect is also applicable to a linearized optimal power flow generation device based on measurement drive in this embodiment, and will not be repeated here.

[0252] To implement the above embodiment, a fifth aspect of the present invention provides an electronic device, including:

[0253] at least one processor; and a memory communicatively coupled to the at least one processor;

[0254] The memory stores instructions that can be executed by the at least one processor, and the instructions are configured to execute the above-mentioned linearized optimal power flow generation method based on measurement drive.

[0255] To implement the above embodiment, a sixth aspect of the present invention proposes a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the above-mentioned linearized optimal power flow generation method based on measurement drive.

[0256] It should be noted that the computer-readable medium mentioned above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device, or component, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present disclosure, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, device, or component. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, which carries computer-readable program code. Such a propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in conjunction with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium may be transmitted using any suitable medium, including but not limited to wires, optical cables, RF (radio frequency), etc., or any suitable combination thereof.

[0257] The computer-readable medium may be included in the electronic device or may exist independently and not incorporated into the electronic device. The computer-readable medium carries one or more programs. When executed by the electronic device, the one or more programs cause the electronic device to perform the measurement-driven linearized optimal power flow generation method described in the above embodiment.

[0258] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on the remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., through the Internet using an Internet service provider).

[0259] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine different embodiments or examples described in this specification and features of different embodiments or examples without contradiction.

[0260] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of the technical features being referred to. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of such features. Throughout the description of this application, "plurality" means at least two, for example, two, three, etc., unless otherwise specifically defined.

[0261] Any process or method description in a flowchart or otherwise described herein may be understood to represent a module, segment or portion of code comprising one or more executable instructions for implementing the steps of a specific logical function or process, and the scope of the preferred embodiments of the present application includes alternative implementations in which functions may be performed out of the order shown or discussed, including performing functions in a substantially simultaneous manner or in the reverse order depending on the functions involved, which should be understood by those skilled in the art to which the embodiments of the present application belong.

[0262] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and programmable read-only memory (EPROM or flash memory), fiber optic devices, and a portable compact disc read-only memory (CDROM). Furthermore, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing it in a suitable manner if necessary, and then storing it in a computer memory.

[0263] It should be understood that various parts of the present application can be implemented using hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used to implement: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0264] Those skilled in the art will understand that all or part of the steps in the method of the above embodiment can be completed by instructing related hardware through a program, and the program can be stored in a computer-readable storage medium. When the program is executed, it includes one or a combination of the steps of the method embodiment.

[0265] In addition, the functional units in the various embodiments of the present application may be integrated into a processing module, or each unit may exist physically separately, or two or more units may be integrated into a module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of a software functional module. If the integrated module is implemented in the form of a software functional module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0266] The storage medium mentioned above may be a read-only memory, a magnetic disk, or an optical disk, etc. Although the embodiments of the present application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present application. Persons skilled in the art may make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present application.

Claims

1. A linearized optimal power flow generation method based on measurement drive, characterized in that: include: Constructing a linearized optimal power flow regression model, wherein the constraints of the model include: linearized power flow robustness model constraints and linearized node voltage amplitude constraints; Identifying the regression coefficients of the linearized power flow robustness model constraint and the linearized node voltage amplitude constraint based on historical measurement sections; Solving the linearized optimal power flow regression model to generate an optimal power flow according to the identification result of the regression coefficient; The objective function expression of the linearized optimal power flow regression model is as follows: Where, Represents the main grid and distributed power generation The Prime Minister has made contributions; The constraint expression of the linearized power flow robustness model is as follows: Where, It is the phase sequence, They are The phase nodes inject active power vector and reactive power vector, For the main grid and distributed power generation The phase has no effort to exert, Including main grid and distributed power supply Phase active load and reactive load; They are Active power vector and reactive power vector of the phase line; They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix; The linearized node voltage amplitude constraint expression is as follows: Where, for Phase voltage magnitude square vector, They are The first regression coefficient matrix, the second regression coefficient matrix, and the third regression coefficient matrix; in, Where, They are The lower and upper limits of Is the root node Phase voltage amplitude; for Predefined voltage reference values ​​for the phases; The constraints of the linearized optimal power flow regression model also include: Kirchhoff's Current Law Constraints: Where, for The node-branch correlation matrix of the phase; for Phase node-inflow branch correlation matrix is the Hadamard product; System power balance constraints: Where i is the bus number and bi is the line number; They are Active power loss vector and reactive power loss vector of the phase line; Line operating capacity constraints: Where, It's a line Phase transfer capacity vector; Generator output constraints: Where, They are The lower and upper limits of They are the lower and upper limits of The identifying of the regression coefficients of the linearized power flow robustness model constraint and the linearized node voltage amplitude constraint based on the historical measurement section includes: 1) Obtain n historical measurement sections of the distribution network as original measurement samples to form an original measurement sample set denoted as X n , n>0, Among them, the jth original measurement sample x j include: 2) Using principal component analysis to analyze the original measurement sample set X n The measurements selected as regression independent variables are subjected to feature extraction and dimensionality reduction, including: Select independent variables from the original measurement sample set and construct the independent variable set is the independent variable of the jth original measurement sample; right After standardization, the expression is as follows: Where, for The expectation of The standard deviation of After standardization, we get The standardized set is recorded as calculate The covariance matrix of Pair covariance matrix Perform eigenvalue decomposition and take the unit eigenvectors v1, v2, ..., v corresponding to the largest m eigenvalues m , we get the projection matrix V=(v1,v2,...,v m ) and the updated set of independent variables 3) Establishing a parameter identification model based on the features extracted in step 2) and the independent variables after dimensionality reduction; st y i -w T x i -b≤e+ξ i ,w T x i +by i ≤e+ξ i * ,x i ,x i * ≥0 in, is the model parameter to be identified; i 、 is a variable; y i is the dependent variable selected from the original measurement sample set, β, ε, and λ are hyperparameters obtained by S-fold cross validation; 4) Solve the parameter identification model established in step 3) and obtain The optimal solution of .

2. The method according to claim 1, characterized in that The linearized optimal power flow regression model is solved by using the interior point method, and the optimal power flow includes: The optimal solution of .

3. The method according to claim 1, characterized in that The independent variable is selected as The dependent variable is selected as and 4. A linearized optimal power flow generation device based on measurement drive according to the method of claim 1, characterized in that: include: A power flow regression model building module is used to build a linearized optimal power flow regression model, wherein the constraints of the model include: linearized power flow robustness model constraints and linearized node voltage amplitude constraints; A regression coefficient identification module, used to identify the regression coefficients of the linearized power flow robustness model constraint and the linearized node voltage amplitude constraint based on historical measurement sections; The optimal power flow generation module is used to solve the linearized optimal power flow regression model according to the identification result of the regression coefficient to generate the optimal power flow.

Images