Line loss analysis method and system based on carbon flow tracking

Abstract

The invention provides a line loss analysis method and system based on carbon flow tracking, and belongs to the technical field of power data anomaly analysis. Comprising the steps that a carbon emission flow model based on power flow analysis is established, CE = CEin-CEo, and CE is the total loss carbon amount of a node i; cEin is the total input carbon quantity of the node i; cEo is the total output carbon amount of the node i; based on a gradient decision tree XGBoost, constructing an XGBoost model used for carbon amount anomaly screening and a target function of the model; analyzing each node in the network topology based on the XGBoost model, and predicting a node with abnormal carbon quantity based on the power and carbon emission density of each node; and electric quantity analysis is carried out on each user corresponding to the carbon quantity abnormal node, the electric quantity analysis comprises high-voltage user abnormal electric quantity analysis and low-voltage user abnormal electric quantity analysis, and abnormal condition types of the users are obtained.

Description

Technical Field

[0001] The present invention relates to the technical field of power data anomaly analysis, and in particular to a line loss analysis method and system based on carbon flow tracing. Background Art

[0002] As an important part of the power system, the operating efficiency and maintenance level of the distribution network directly affect the quality and stability of power supply. Line loss management is one of the key indicators used to measure the economy and operating status of the power grid. It represents the power loss from the power generation end to the user end, including both technical and management losses. Abnormal line loss usually refers to line loss beyond the normal range, which may be caused by equipment failure, line aging, illegal access, etc. Effective monitoring and positioning of line loss abnormalities are of great significance for preventing power theft, identifying abnormal power grid topology, and improving the overall operational efficiency of the power grid.

[0003] The power variation of line nodes is large and the load properties are diverse. In line loss analysis, it is usually necessary to comprehensively detect the power consumption data of all users at all nodes and perform data analysis. This method requires a huge amount of data and a large amount of calculation. Accordingly, the monitoring and identification efficiency of line loss anomalies is not high. Summary of the invention

[0004] In view of this, the present invention provides a line loss analysis method and system based on carbon flow tracking, which screens out abnormal carbon nodes by predicting the abnormal carbon content of nodes, and then identifies the abnormal electricity consumption of users corresponding to the abnormal carbon nodes, which can effectively improve the monitoring and identification efficiency of line loss anomalies and improve the operation and maintenance efficiency of the power system.

[0005] The technical solution adopted by the embodiment of the present invention to solve the technical problem is:

[0006] A line loss analysis method based on carbon flow tracking, comprising:

[0007] Step S1, establishing a carbon emission flow model based on power flow analysis:

[0008]

[0009] Where CE is the total carbon loss of node i; CE in is the total carbon input of node i; CE o is the total output carbon of node i; m∈[1,M], M is the total number of input branches of node i, p m is the input power of input branch m, e m is the node carbon flow density of input branch m; n∈[1,N], N is the total number of output branches of node i, p n is the output power of output branch n, e n is the node carbon flow density of output branch n;

[0010] Step S2, constructing an XGBoost model for carbon content anomaly screening based on the gradient decision tree XGBoost, the XGBoost model is defined as:

[0011]

[0012] Where: F represents the classification and regression tree space, f k is the base learner, f k (x i ) represents the kth tree for sample x i The prediction score of the XGBoost model is the total carbon loss CE of node i. It indicates the prediction result obtained by the XGBoost model based on the total carbon consumption of the input node i, and the conclusion of whether the node i is a carbon abnormal node;

[0013] The objective function of the XGBoost model is defined as:

[0014]

[0015] In the formula, g i and h i is the loss function The first and second partial derivatives of , y i For sample x i The corresponding true value, T represents the number of decision trees, γ and λ are hyperparameters, j is the decision tree count value, j∈[1,T]; I j is the sample x i The set of leaf nodes j that are assigned;

[0016] Step S3, analyzing each node in the network topology based on the XGBoost model, and predicting the carbon emission abnormal nodes based on the power and carbon emission density of each node;

[0017] Step S4, performing power analysis on each user corresponding to the abnormal carbon amount node, wherein the power analysis includes abnormal power analysis of high-voltage users and abnormal power analysis of low-voltage users, and obtaining the type of abnormal situation existing in the user.

[0018] Preferably, the derivation process of the objective function of the XGBoost model in step S2 includes:

[0019] The objective function of the XGBoost model is initially defined as:

[0020]

[0021] Where: y i For sample x iThe corresponding true value, represents the prediction deviation of the i-th sample, I is the total number of samples; Ω(f) represents the model complexity; T represents the number of decision trees, ω represents the weight, and γ and λ are hyperparameters;

[0022] set up To predict the i-th sample result at the t-th iteration, the objective function of t trees is:

[0023]

[0024] The objective function of t trees is further expressed by performing a second-order Taylor expansion on the objective function:

[0025]

[0026] Where: g i and h i is the loss function The first and second partial derivatives of , and is a constant term; j is the decision tree count value, j∈[1,T]; I j is the sample x i The set of leaf nodes j assigned to, ω j is the leaf node weight;

[0027] Derivative the second-order Taylor expansion, and when the derivative is 0, the leaf node weight ω is obtained j * :

[0028]

[0029] ω j * Substitute the original objective function Obj(θ) to get the new objective function Obj t :

[0030]

[0031] Preferably, the abnormal power analysis of high-voltage users in step S4 includes:

[0032] Determination of voltage loss abnormality for three-phase high voltage users:

[0033] For three-phase four-wire high-voltage users, the voltage loss threshold is defined as U H1 , analyze the absolute value of the current at the time of historical voltage loss of each phase. When the three conditions are met at the same time, it is determined that the high-voltage user has voltage loss and current abnormality:

[0034] Condition C11: There is a daily abnormal moment in a day, the daily abnormal moment is the voltage loss point moment, and at least one phase sequence satisfies that the absolute value of the current value is not less than the high voltage current threshold;

[0035] Condition C12: The number of daily abnormal points of at least one phase sequence is not less than N1, and the number of daily abnormal points is the number of daily abnormal moments;

[0036] Condition C13: The number of consecutive days of daily abnormalities is not less than N2, and the number of consecutive days of daily abnormalities refers to the number of consecutive days that meet condition C12;

[0037] Determination of abnormal current imbalance for three-phase four-wire high-voltage users:

[0038] For three-phase four-wire high-voltage users, the daily average absolute values ​​of the currents I1I2I3 are not 0A, and the daily average absolute value of the current of one phase is less than the daily average absolute values ​​of the currents of the other two phases. The phase current unbalance corresponding to the minimum average value is defined, and the correlation coefficient P of the current time point data of the other two phases is calculated according to the Pearson correlation coefficient formula. H , when P∈[0.8,1], it is confirmed to meet the positive strong correlation; analyze the daily sampling values ​​of the two-phase currents that are positively correlated with each other, sort the absolute values ​​of the two-phase currents from high to low, and the time points corresponding to the first three values ​​of each phase are used as the three maximum time points of each phase. When the four conditions are met at the same time, it is determined that the low-voltage user has a current imbalance abnormality:

[0039] Condition C21: The first three values ​​of each phase are greater than the high voltage current threshold;

[0040] Condition C22: The three maximum time points of each phase occur on three identical sections;

[0041] Condition C23: The absolute value of active power at any maximum value time point of any phase is greater than the absolute value of reactive power;

[0042] Condition C24: At any maximum value point in any phase, the ratio of the current at the maximum value point to the phase sequence current corresponding to the minimum average value is higher than the preset ratio P H ;

[0043] Determination of abnormal current imbalance for three-phase three-wire high-voltage users:

[0044] For three-phase three-wire high-voltage users, the daily average values ​​of the absolute values ​​of currents I1 and I3 are not 0A, which means that the phase sequence of the current with the smaller daily average value of the absolute value of the current is unbalanced; the absolute values ​​of the current with the larger daily average value of the absolute value of the current are sorted in descending order, and the time points corresponding to the first three values ​​are taken as the three maximum time points of the phase sequence. When all three conditions are met at the same time, it is determined that the three-phase three-wire high-voltage user has a current imbalance abnormality:

[0045] Condition C31: the first three values ​​are all greater than the high voltage current threshold;

[0046] Condition C32: The current ratio of the slice at each moment is always greater than the preset ratio P L ; Among them, in the current ratio, the phase sequence current of the current with the larger daily average absolute value of the current is used as the denominator;

[0047] Condition C33: The absolute value of active power at any maximum value time point is greater than the absolute value of reactive power.

[0048] Preferably, the abnormal power analysis of high-voltage users in step S4 also includes the determination of abnormal public-private transformer users:

[0049] Preferably, the abnormal power analysis of high-voltage users in step S4 also includes the determination of abnormal public-private transformer users: abnormal determination of private transformer users: using the Pearson correlation coefficient calculation formula to calculate the correlation r1 between the power data of the private transformer user and the line loss data; calculating the DTW distance D1 between the power data of the private transformer user and the line loss data; wherein, when r1∈[0.8,1] and D1∈[0,D t1 ) to confirm that the metering of the dedicated transformer user is abnormal;

[0050] Public transformer user abnormality judgment: Use the Pearson correlation coefficient calculation formula to calculate the correlation r2 between the public transformer area line loss data and the line line loss data; calculate the DTW distance D2 between the public transformer area line loss data and the line line loss data; Among them, when r2∈[-1,-0.8] and D2∈(D t2 ,+∞), confirming that the public transformer user metering is abnormal.

[0051] Preferably, the formula for calculating the correlation r in the Pearson correlation coefficient calculation formula in claim 4 is:

[0052]

[0053] Among them, X i Indicates the power consumption data of private transformer users or line loss data of public transformer areas, Y i Indicates line loss data. For X i The mean of Y i The mean of r∈[0.8,1] indicates that the two are strongly positively correlated; r∈[-1,-0.8] indicates that the two are strongly negatively correlated.

[0054] Preferably, the DTW dynamic time warping distance calculation process in claim 4 is:

[0055] Given two time series X = {x1, x2, ..., x m} and Y={y1,y2,...,y n}, construct the cumulative distance D(i,j), where X is the time series set of the dedicated transformer user power data or the public transformer area line loss data, and Y is the time series set of the line line loss data:

[0056] D(i,j)=d(x i ,y j )+min{D(i-1,j),D(i,j-1),D(i-1,j-1)}

[0057] Where: d(x i ,y j )=(x i -y j ) 2 ,1≤i≤m,1≤j≤n, the initial condition is D(1,1)=d(x i ,y j );

[0058] The range reached by the curved path W in the distance matrix is ​​called the curved window, and the final metric distance is defined as:

[0059]

[0060] Where K is the length of the curved path W; set the threshold D t1 , D t2 , D t1 <D t2 ,D∈[0,D t1 ) indicates that the two are strongly positively correlated; D∈(D t2 ,+∞) indicates that the two are strongly negatively correlated.

[0061] Preferably, the step S4 of analyzing abnormal power consumption of low-voltage users includes:

[0062] Determination of voltage loss and current abnormality for three-phase four-wire low-voltage users:

[0063] For three-phase four-wire low-voltage users, the voltage lower limit threshold is defined as U L1 , analyze the current value at the time of historical voltage loss, and when the three conditions are met at the same time, it is determined that the low-voltage user has voltage loss and current abnormality:

[0064] Condition C41: There is a daily abnormal moment in a day, the daily abnormal moment is the voltage loss point moment, and at least one phase sequence satisfies that the absolute value of the current value is not less than the low-voltage current threshold;

[0065] Condition C42: The proportion of the number of daily abnormal points in at least one phase sequence to the total number of daily sampling points is not less than the preset percentage P1, and the number of daily abnormal points is the number of daily abnormal moments;

[0066] Condition C43: The number of consecutive days of daily abnormalities is not less than N3, and the number of consecutive days of daily abnormalities refers to the number of consecutive days that meet condition C42;

[0067] Judgment of abnormality of single-phase voltage exceeding lower limit for low-voltage users:

[0068] For single-phase low-voltage users, the voltage lower limit threshold is defined as U L2 When two conditions are met at the same time, it is determined that the low-voltage user has a single-phase voltage exceeding the lower limit abnormality:

[0069] Condition C51: The proportion of daily abnormal points to the total number of daily sampling points is not less than the preset percentage P1;

[0070] Condition C52: The number of consecutive days of daily abnormalities is not less than N3, and the number of consecutive days of daily abnormalities refers to the number of consecutive days that meet condition C51;

[0071] Abnormal judgment of single-phase voltage exceeding upper limit for low-voltage users:

[0072] For single-phase low-voltage users, the voltage upper limit threshold is defined as U L3 , analyze the current value at the time of historical voltage loss, and when two conditions are met at the same time, it is determined that the low-voltage user has a single-phase voltage exceeding the upper limit abnormality:

[0073] Condition C61: The proportion of daily abnormal points to the total number of daily sampling points is not less than the preset percentage P1;

[0074] Condition C62: The number of consecutive days of daily abnormalities is not less than N3, and the number of consecutive days of daily abnormalities refers to the number of consecutive days that meet condition C61;

[0075] Determination of abnormal current imbalance for three-phase four-wire low-voltage users:

[0076] For three-phase four-wire low-voltage users, the daily average values ​​of the absolute values ​​of the three-phase four-wire currents are not 0A, and the daily average value of the absolute value of the current of one phase is less than the daily average value of the absolute values ​​of the current of the other two phases. The phase current corresponding to the minimum average value is defined as unbalanced. The correlation coefficient P of the other two phase current time point data is calculated according to the Pearson correlation coefficient formula. When P∈[0.8,1], it is confirmed to meet the positive strong correlation. The daily sampling values ​​of the two-phase currents that are positively correlated with each other are analyzed, and the absolute values ​​of the two-phase currents are sorted from high to low. The time points corresponding to the first three values ​​of each phase are used as the three maximum time points of each phase. When the four conditions are met at the same time, it is determined that the low-voltage user has a current imbalance abnormality:

[0077] Condition C71: The first three values ​​of each phase are greater than the low voltage current threshold;

[0078] Condition C72: Among the three maximum time points of the two phases, there exists a maximum time point that occurs on the same cutting surface;

[0079] Condition C73: The absolute value of the active power at any maximum point in time of any phase is greater than the absolute value of the reactive power;

[0080] Condition C74: At any maximum value point in any phase, the ratio of the current at the maximum value point to the phase sequence current corresponding to the minimum average value is higher than the preset ratio P L .

[0081] A line loss analysis system based on carbon flow tracking is used to execute the above method.

[0082] It can be seen from the above technical scheme that the line loss analysis method and system based on carbon flow tracking provided by the embodiment of the present invention first establishes a carbon emission flow model based on power flow analysis; constructs an XGBoost model for carbon quantity anomaly screening and the objective function of the model based on the gradient decision tree XGBoost; analyzes each node in the network topology based on the XGBoost model, and predicts the carbon quantity abnormal nodes based on the power and carbon emission density of each node; performs power analysis on each user corresponding to the carbon quantity abnormal node, and the power analysis includes abnormal power analysis of high-voltage users and abnormal power analysis of low-voltage users, and obtains the type of abnormal situation of the user. The present invention first screens out abnormal carbon nodes by predicting the abnormal carbon quantity of the nodes, and then identifies the abnormal power consumption of users based on line loss for the abnormal carbon quantity nodes, which can effectively improve the monitoring and identification efficiency of abnormal line loss in the substation area. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 The figure is a flow chart of the line loss analysis method based on carbon flow tracking of the present invention.

[0084] Figure 2 Schematic diagram of node carbon emission input and output. DETAILED DESCRIPTION

[0085] The technical scheme and technical effects of the present invention are further elaborated in detail below in conjunction with the accompanying drawings of the present invention.

[0086] refer to Figure 1 As shown, the present invention provides a line loss analysis method based on carbon flow tracking, and the specific implementation includes:

[0087] Step S1, establishing a carbon emission flow model based on power flow analysis:

[0088]

[0089] Where CE is the total carbon loss of node i; CE in is the total carbon input of node i; CE o is the total output carbon of node i; m∈[1,M], M is the total number of input branches of node i, p mis the input power of input branch m, e m is the node carbon flow density of input branch m; n∈[1,N], N is the total number of output branches of node i, p n is the output power of output branch n, e n is the node carbon flow density of output branch n;

[0090] Step S2, constructing an XGBoost model for carbon content anomaly screening based on the gradient decision tree XGBoost, the XGBoost model is defined as:

[0091]

[0092] Where: F represents the classification and regression tree (CART) space, f k is the base learner, f k (x i ) represents the kth tree for sample x i The prediction score of the XGBoost model is the total carbon loss CE of node i. It indicates the prediction result obtained by the XGBoost model based on the total carbon consumption of the input node i, and the conclusion of whether the node i is a carbon abnormal node;

[0093] The objective function of the XGBoost model is initially defined as:

[0094]

[0095] Where: y i For sample x i The corresponding true value, represents the prediction deviation of the i-th sample, I is the total number of samples; Ω(f) represents the model complexity; T represents the number of decision trees, ω represents the weight, and γ and λ are hyperparameters;

[0096] set up To predict the i-th sample result at the t-th iteration, the objective function of t trees is:

[0097]

[0098] The objective function of t trees is further expressed by performing a second-order Taylor expansion on the objective function:

[0099]

[0100] Where: g i and h i is the loss function The first and second partial derivatives of , and is a constant term; j is the decision tree count value, j∈[1,T]; I j is the sample x i The set of leaf nodes j assigned to, ω j is the leaf node weight; f(x i ) represents the allocation function of a node in a decision tree model, which is used to indicate the input instance x i To which leaf node j is it assigned. Therefore, the Ij set contains all instances assigned to leaf node j.

[0101] Derivative the second-order Taylor expansion, and when the derivative is 0, the leaf node weight ω is obtained j * :

[0102]

[0103] ω j * Substitute the original objective function Obj(θ) to get the new objective function Obj t As the objective function of the XGBoost model:

[0104]

[0105] The historical data can be used to build a i Data x i The database is further divided into training set, validation set and test set. The training set is used to train the model, and the hyperparameters γ and λ are tuned by particle optimization and other methods to minimize the prediction loss (the iteration cutoff condition can be the maximum number of iterations or loss convergence). After model training, model validation and testing, a model with better prediction effect is obtained.

[0106] Step S3, analyzing each node in the network topology based on the XGBoost model, and predicting the carbon emission abnormal nodes based on the power and carbon emission density of each node;

[0107] Step S4, performing power analysis on each user corresponding to the carbon quantity abnormal node, the power analysis includes abnormal power analysis of high-voltage users and abnormal power analysis of low-voltage users, and obtaining the type of abnormal situation existing in the user.

[0108] Step S4: Analysis of abnormal power consumption of high-voltage users includes:

[0109] Determination of voltage loss abnormality for three-phase high voltage users:

[0110] For three-phase four-wire high-voltage users, the voltage loss threshold is defined as U H1, analyze the absolute value of the current at the time of historical voltage loss of each phase. When the three conditions are met at the same time, it is determined that the high-voltage user has voltage loss and current abnormality:

[0111] Condition C11: There is a daily abnormal moment in a day, which is the moment of voltage loss, and at least one phase sequence satisfies that the absolute value of the current value is not less than the high voltage current threshold;

[0112] Condition C12: The number of daily abnormal points of at least one phase sequence is not less than N1, and the number of daily abnormal points is the number of daily abnormal moments;

[0113] Condition C13: The number of consecutive days of daily anomalies is not less than N2. The number of consecutive days of daily anomalies refers to the number of consecutive days that meet condition C12;

[0114] For example, according to the standard voltage specification, if it is 3*220V, the threshold U H1 is 176V; if the standard voltage specification is 3*100V, the threshold U H1 is 80V; if the standard voltage specification is 3*57.7V, the threshold U H1 46V; at least one of the phase sequence voltage loss moments meets the current absolute value maximum value ≥ 0.1A; the number of daily abnormal points in any phase sequence is ≥ 12 (including the situation where both two-phase and three-phase meet the voltage loss condition), and the number of consecutive days of daily abnormality is ≥ 3 days. If the above conditions are met, it is determined that the user has voltage loss abnormality.

[0115] Determination of abnormal current imbalance for three-phase four-wire high-voltage users:

[0116] For three-phase four-wire high-voltage users, the daily average absolute values ​​of the currents I1I2I3 are not 0A, and the daily average absolute value of the current of one phase is less than the daily average absolute values ​​of the currents of the other two phases. The phase current unbalance corresponding to the minimum average value is defined, and the correlation coefficient P of the current time point data of the other two phases is calculated according to the Pearson correlation coefficient formula. H , when P∈[0.8,1], it is confirmed to meet the positive strong correlation; analyze the daily sampling values ​​of the two-phase currents that are positively correlated with each other, sort the absolute values ​​of the two-phase currents from high to low, and the time points corresponding to the first three values ​​of each phase are taken as the three maximum time points of each phase. When all four conditions are met at the same time, it is determined that the low-voltage user has a current imbalance abnormality:

[0117] Condition C21: The first three values ​​of each phase are greater than the high voltage current threshold;

[0118] Condition C22: The three maximum time points of each phase occur on three identical slices; (that is, the three maximum time points of the three phases are distributed in slice 1, slice 2, and slice 3 respectively);

[0119] Condition C23: The absolute value of active power at any maximum value time point of any phase is greater than the absolute value of reactive power;

[0120] Condition C24: At any maximum value point in any phase, the ratio of the current at the maximum value point to the phase sequence current corresponding to the minimum average value is higher than the preset ratio P H ;

[0121] For example, the daily average absolute values ​​of currents I1I2I3 are not 0A (daily data refers to 96 data points). If the daily average absolute values ​​of I2 and I3 are greater than the daily average absolute value of I1 for three-phase four-wire, the current in the phase sequence of I1 is defined as unbalanced; the absolute values ​​of the currents of the two phases I2I3 of the three-phase four-wire are sorted, and the top three maximum values ​​of each phase meet the following conditions: the top three maximum values ​​of the absolute values ​​are all greater than 0.1A; the three maximum time points of any phase correspond to the same cutting plane; the absolute values ​​of active power corresponding to the three maximum time points are respectively greater than the absolute values ​​of reactive power; the three maximum time points satisfy |I3 / I1|>1.5 and |I2 / I1|>1.5; P I2,I3 The correlation coefficient is greater than 0.8; if the above conditions are met, it is determined that the user has a current imbalance abnormality. Here, the correlation coefficient P I2,I3 The calculation formula is:

[0122]

[0123] Where P I2,I3 represents the correlation between the phase sequence of users I2 and I3, Cov(I2,I3) represents the covariance of the phase sequence of users I2 and I3, Var(I2) and Var(I3) represent the variance, and P I2,I3 ∈[0.8,1], when P I2,I3 The larger the value, the stronger the positive correlation;

[0124] Determination of abnormal current imbalance for three-phase three-wire high-voltage users:

[0125] For three-phase three-wire high-voltage users, the daily average values ​​of the absolute values ​​of currents I1 and I3 are not 0A, which means that the phase sequence of the current with the smaller daily average value of the absolute value of the current is unbalanced; the absolute values ​​of the current with the larger daily average value of the absolute value of the current are sorted in descending order, and the time points corresponding to the first three values ​​are taken as the three maximum time points of the phase sequence. When all three conditions are met at the same time, it is determined that the three-phase three-wire high-voltage user has a current imbalance abnormality:

[0126] Condition C31: The first three values ​​are all greater than the high voltage current threshold;

[0127] Condition C32: The current ratio of the slice at each moment is always greater than the preset ratio P L ; Among them, in the current ratio, the phase sequence current of the current with the larger daily average absolute value of the current is used as the denominator;

[0128] Condition C33: The absolute value of active power at any maximum value time point is greater than the absolute value of reactive power.

[0129] For example, the daily average values ​​of the absolute values ​​of the currents I1 and I3 are not 0A. If the daily average value of the absolute value of I3 of the three-phase three-wire is greater than the daily average value of the absolute value of I1, then the current of the phase sequence of I1 is defined to be unbalanced. The absolute values ​​of the currents of I3 of the three-phase three-wire are sorted (96-point sorting), and the top three maximum values ​​are taken to meet condition 1: the top three maximum values ​​of the absolute values ​​are all greater than 0.1A; condition 2: |I3 / I1|>1.5; condition 3: the absolute values ​​of the active power corresponding to the three maximum time points of I3 are respectively greater than the absolute values ​​of the reactive power. If the above three conditions are met, it is determined that the user has a current imbalance abnormality.

[0130] Preferably, the abnormal power analysis of high-voltage users in step S4 also includes the determination of abnormal public-private transformer users: abnormal determination of private transformer users: using the Pearson correlation coefficient calculation formula to calculate the correlation r1 between the power data of the private transformer user and the line loss data; calculating the DTW distance D1 between the power data of the private transformer user and the line loss data; wherein, when r1∈[0.8,1] and D1∈[0,D t1 ) to confirm that the metering of the dedicated transformer user is abnormal;

[0131] Public transformer user abnormality judgment: Use the Pearson correlation coefficient calculation formula to calculate the correlation r2 between the public transformer area line loss data and the line line loss data; calculate the DTW distance D2 between the public transformer area line loss data and the line line loss data; Among them, when r2∈[-1,-0.8] and D2∈(D t2 ,+∞), confirming that the public transformer user metering is abnormal.

[0132] Pearson correlation coefficient calculation formula The formula for calculating the correlation r is:

[0133]

[0134] Among them, X i Indicates the power consumption data of private transformer users or line loss data of public transformer areas, Y i Indicates line loss data. For X i The mean of Y i The mean of r∈[0.8,1] indicates that the two are strongly positively correlated; r∈[-1,-0.8] indicates that the two are strongly negatively correlated.

[0135] The DTW dynamic time warping distance calculation process is:

[0136] Given two time series X = {x1, x2, ..., x m} and Y={y1,y2,...,y n}, construct the cumulative distance D(i,j), where X is the time series set of the dedicated transformer user power data or the public transformer area line loss data, and Y is the time series set of the line line loss data:

[0137] D(i,j)=d(x i ,y j )+min{D(i-1,j),D(i,j-1),D(i-1,j-1)} (11)

[0138] Where: d(x i ,y j )=(x i -y j ) 2 ,1≤i≤m,1≤j≤n, the initial condition is D(1,1)=d(x i ,y j );

[0139] The range reached by the curved path W in the distance matrix is ​​called the curved window, and the final metric distance is defined as:

[0140]

[0141] Where K is the length of the curved path W; set the threshold D t1 , D t2 , D t1 <D t2 ,D∈[0,D t1 ) indicates that the two are strongly positively correlated; D∈(D t2 ,+∞) indicates that the two are strongly negatively correlated.

[0142] Preferably, the abnormal power consumption analysis of low-voltage users in step S4 includes:

[0143] Determination of voltage loss and current abnormality for three-phase four-wire low-voltage users:

[0144] For three-phase four-wire low-voltage users, including direct-type and transformer meters, the voltage lower limit threshold is defined as U L1 , analyze the current value at the time of historical voltage loss, and when the three conditions are met at the same time, it is determined that the low-voltage user has voltage loss and current abnormality:

[0145] Condition C41: There is a daily abnormal moment in a day, which is the voltage loss point moment, and at least one phase sequence satisfies that the absolute value of the current value is not less than the low-voltage current threshold;

[0146] Condition C42: The proportion of the number of daily abnormal points in at least one phase sequence to the total number of daily sampling points is not less than the preset percentage P1, and the number of daily abnormal points is the number of daily abnormal moments;

[0147] Condition C43: The number of consecutive days of daily anomalies is not less than N3. The number of consecutive days of daily anomalies refers to the number of consecutive days that meet condition C42;

[0148] For example, the lower threshold of the three-phase four-wire voltage is 190V; at least one of the voltage loss points satisfies the current absolute value maximum value ≥ 0.1A; the number of abnormal points in any phase sequence is ≥ 50% (including the situation where both two-phase and three-phase meet the voltage loss conditions), and the number of abnormal days is ≥ 3 days. If the above conditions are met, it is determined that the user has voltage loss and current abnormality.

[0149] Judgment of abnormality of single-phase voltage exceeding lower limit for low-voltage users:

[0150] For single-phase low-voltage users, the voltage lower limit threshold is defined as U L2 When two conditions are met at the same time, it is determined that the low-voltage user has a single-phase voltage exceeding the lower limit abnormality:

[0151] Condition C51: The proportion of daily abnormal points to the total number of daily sampling points is not less than the preset percentage P1;

[0152] Condition C52: The number of consecutive days of daily anomalies is not less than N3. The number of consecutive days of daily anomalies refers to the number of consecutive days that meet condition C51;

[0153] For example, the voltage lower limit threshold for a single-phase low-voltage user is 132V, the number of daily abnormal points is ≥50%, and the number of abnormal days is ≥3 days. If the above conditions are met, it is determined that the user has a single-phase voltage lower limit abnormality.

[0154] Abnormal judgment of single-phase voltage exceeding upper limit for low-voltage users:

[0155] For single-phase low-voltage users, the voltage upper limit threshold is defined as U L3 , analyze the current value at the time of historical voltage loss, and when two conditions are met at the same time, it is determined that the low-voltage user has a single-phase voltage exceeding the upper limit abnormality:

[0156] Condition C61: The proportion of daily abnormal points to the total number of daily sampling points is not less than the preset percentage P1;

[0157] Condition C62: The number of consecutive days with daily anomalies is not less than N3. The number of consecutive days with daily anomalies refers to the number of consecutive days that meet condition C61;

[0158] For example, the voltage of a single-phase low-voltage user exceeds the upper limit threshold of 300V, the number of daily abnormal points is ≥50%, and the number of abnormal days is ≥3 days. If the above conditions are met, it is determined that the user has a single-phase voltage exceeding the upper limit abnormality.

[0159] Determination of abnormal current imbalance for three-phase four-wire low-voltage users:

[0160] For three-phase four-wire low-voltage users, the daily average values ​​of the absolute values ​​of the three-phase four-wire currents are not 0A, and the daily average value of the absolute value of the current of one phase is less than the daily average value of the absolute values ​​of the current of the other two phases. The phase current imbalance corresponding to the minimum average value is defined, and the correlation coefficient P of the other two phase current time point data is calculated according to the Pearson correlation coefficient formula. When P∈[0.8,1], it is confirmed to meet the positive strong correlation; analyze the daily sampling values ​​of the two-phase currents that are positively correlated with each other, and sort the absolute values ​​of the two-phase currents from high to low. The time points corresponding to the first three values ​​of each phase are used as the three maximum time points of each phase. When all four conditions are met at the same time, it is determined that the low-voltage user has a current imbalance abnormality:

[0161] Condition C71: The first three values ​​of each phase are greater than the low voltage current threshold;

[0162] Condition C72: Among the three maximum time points of the two phases, there exists a maximum time point that occurs on the same cutting surface;

[0163] Condition C73: The absolute value of the active power at any maximum point in time of any phase is greater than the absolute value of the reactive power;

[0164] Condition C74: At any maximum value point in any phase, the ratio of the current at the maximum value point to the phase sequence current corresponding to the minimum average value is higher than the preset ratio P L .

[0165] For example, the daily average values ​​of the absolute values ​​of the three-phase four-wire currents I1, I2, and I3 are not 0A. If the daily average values ​​of the absolute values ​​of I2 and I3 are greater than the daily average value of the absolute value of I1, then the phase sequence current of I1 is defined as unbalanced (similarly, the phase sequence current of I2 and I3 is unbalanced); calculate the correlation coefficient P I2,I3 (Refer to formula (9)), P I2,I3 The correlation coefficient must be greater than 0.8, indicating a strong positive correlation;

[0166] The absolute values ​​of the two-phase currents of the three-phase four-wire I2I3 are sorted, and the top three maximum values ​​of each phase are taken to meet the following conditions: the top three absolute values ​​are all greater than 0.1A; at least one of the three maximum value time points occurs on the same cutting plane; the absolute values ​​of the active power at the three time points are respectively greater than the absolute values ​​of the reactive power; the cutting planes at the three time points respectively satisfy |I3 / I1|>1.5 and |I2 / I1|>1.5; if the above conditions are met, it is determined that the user has a current imbalance abnormality.

[0167] Furthermore, the present invention provides a line loss analysis system based on carbon flow tracking, which is used to execute the method described in steps S1-S4. The system may include a data acquisition module, a model building module, a prediction module and an analysis module:

[0168] The data acquisition module is used to collect the input power of each input branch of each node and the output power of the output branch; obtain the historical power consumption data of high-voltage users and low-voltage users of the node line;

[0169] Model building module, used to build analysis models, such as carbon emission flow model based on trend analysis, XGBoost model, etc.;

[0170] The prediction module first obtains carbon loss data based on the carbon emission flow model, and then inputs it into the XGBoost model for node anomaly prediction to obtain carbon quantity abnormal nodes;

[0171] The analysis module is used to analyze the line loss of the carbon quantity abnormal nodes, including the abnormal power analysis of high-voltage users and the abnormal power analysis of low-voltage users, and to obtain the type of abnormal situation existing in the user.

[0172] The present invention screens out abnormal carbon nodes by predicting abnormal carbon content of nodes, and then identifies abnormal electricity consumption of users corresponding to the abnormal carbon content nodes, which can effectively improve the monitoring and identification efficiency of line loss anomalies and improve the operation and maintenance efficiency of the power system.

[0173] What is disclosed above is only a preferred embodiment of the present invention, which certainly cannot be used to limit the scope of rights of the present invention. A person skilled in the art can understand that all or part of the processes of the above embodiments and equivalent changes made according to the claims of the present invention still fall within the scope of the invention.

Claims

1. A line loss analysis method based on carbon flow tracking, characterized in that: include: Step S1, establishing a carbon emission flow model based on power flow analysis: Where, CE is the total carbon loss of node i; CE in is the total carbon input of node i; CE o is the total output carbon of node i; m∈[1,M], M is the total number of input branches of node i, p m is the input power of input branch m, e m is the node carbon flow density of input branch m; n∈[1,N], N is the total number of output branches of node i, p n is the output power of output branch n, e n is the node carbon flow density of output branch n; Step S2, constructing an XGBoost model for carbon content anomaly screening based on the gradient decision tree XGBoost, the XGBoost model is defined as: Where: F represents the classification and regression tree space, f k is the base learner, f k (x i ) represents the kth tree for sample x i The prediction score of the XGBoost model is the total carbon loss CE of node i. It indicates the prediction result obtained by the XGBoost model based on the total carbon consumption of the input node i, and the conclusion of whether the node i is a carbon abnormal node; The objective function of the XGBoost model is defined as: In the formula, g i and h i is the loss function The first and second partial derivatives of , y i For sample x i The corresponding true value, T represents the number of decision trees, γ and λ are hyperparameters, j is the decision tree count value, j∈[1,T]; I j is the sample x i The set of leaf nodes j that are assigned; Step S3, analyzing each node in the network topology based on the XGBoost model, and predicting the carbon emission abnormal nodes based on the power and carbon emission density of each node; Step S4, performing power analysis on each user corresponding to the abnormal carbon amount node, wherein the power analysis includes abnormal power analysis of high-voltage users and abnormal power analysis of low-voltage users, and obtaining the type of abnormal situation existing in the user.

2. The line loss analysis method based on carbon flow tracking according to claim 1, characterized in that: The derivation process of the objective function of the XGBoost model in step S2 includes: The objective function of the XGBoost model is initially defined as: Where: y i For sample x i The corresponding true value, represents the prediction deviation of the i-th sample, I is the total number of samples; Ω(f) represents the model complexity; T represents the number of decision trees, ω represents the weight, γ and λ are hyperparameters; let To predict the i-th sample result at the t-th iteration, the objective function of t trees is: The objective function of t trees is further expressed by performing a second-order Taylor expansion on the objective function: Where: g i and h i is the loss function The first and second partial derivatives of , and is a constant term; j is the decision tree count value, j∈[1,T]; I j is the sample x i The set of leaf nodes j assigned to, ω j is the leaf node weight; Derivative the second-order Taylor expansion, and when the derivative is 0, the leaf node weight ω is obtained j * : ω j * Substitute the original objective function Obj(θ) to get the new objective function Obj t :

3. The line loss analysis method based on carbon flow tracking according to claim 1, characterized in that: The step S4 of analyzing abnormal power consumption of high-voltage users includes: Determination of voltage loss abnormality for three-phase high voltage users: For three-phase four-wire high-voltage users, the voltage loss threshold is defined as U H1 , analyze the absolute value of the current at the time of historical voltage loss of each phase. When the three conditions are met at the same time, it is determined that the high-voltage user has voltage loss and current abnormality: Condition C11: There is a daily abnormal moment in a day, the daily abnormal moment is the voltage loss point moment, and at least one phase sequence satisfies that the absolute value of the current value is not less than the high voltage current threshold; Condition C12: The number of daily abnormal points of at least one phase sequence is not less than N1, and the number of daily abnormal points is the number of daily abnormal moments; Condition C13: The number of consecutive days of daily abnormalities is not less than N2, and the number of consecutive days of daily abnormalities refers to the number of consecutive days that meet condition C12; Determination of abnormal current imbalance for three-phase four-wire high-voltage users: For three-phase four-wire high-voltage users, the daily average absolute values ​​of the three-phase I1, I2, and I3 currents are not 0A, and the daily average absolute value of one phase sequence current is less than the daily average absolute value of the other two phase sequence currents. The phase sequence current unbalance corresponding to the minimum average value is defined, and the correlation coefficient P of the other two phase sequence current time point data is calculated according to the Pearson correlation coefficient formula. H , when P∈[0.8,1], it is confirmed to meet the positive strong correlation; analyze the daily sampling values ​​of the two-phase currents that are positively correlated with each other, sort the absolute values ​​of the two-phase currents from high to low, and the time points corresponding to the first three values ​​of each phase are used as the three maximum time points of each phase. When the four conditions are met at the same time, it is determined that the low-voltage user has a current imbalance abnormality: Condition C21: The first three values ​​of each phase are greater than the high voltage current threshold; Condition C22: The three maximum time points of each phase occur on three identical sections; Condition C23: The absolute value of active power at any maximum time point of any phase is greater than the absolute value of reactive power; Condition C24: At any maximum time point of any phase, the ratio of the current at the maximum time point to the phase sequence current corresponding to the minimum average value is higher than the preset ratio P H ; Determination of abnormal current imbalance for three-phase three-wire high-voltage users: For three-phase three-wire high-voltage users, the daily average values ​​of the absolute values ​​of currents I1 and I3 are not 0A, which means that the phase sequence of the current with the smaller daily average value of the absolute value of the current is unbalanced; the absolute values ​​of the current with the larger daily average value of the absolute value of the current are sorted in descending order, and the time points corresponding to the first three values ​​are taken as the three maximum time points of the phase sequence. When all three conditions are met at the same time, it is determined that the three-phase three-wire high-voltage user has a current imbalance abnormality: Condition C31: the first three values ​​are all greater than the high voltage current threshold; Condition C32: The current ratio of the slice at each moment is always greater than the preset ratio P L ; Among them, in the current ratio, the phase sequence current of the current with the larger daily average absolute value of the current is used as the denominator; Condition C33: The absolute value of active power at any maximum value time point is greater than the absolute value of reactive power.

4. The line loss analysis method based on carbon flow tracking according to claim 3 is characterized in that: The abnormal power analysis of high-voltage users in step S4 also includes the determination of abnormal public-private transformer users: Abnormal determination of special transformer users: The correlation r1 between the special transformer user power data and line loss data is calculated using the Pearson correlation coefficient calculation formula; Calculate the DTW distance D1 between the dedicated transformer user power data and the line loss data; when r1∈[0.8,1] and D1∈[0,D t1 ) to confirm that the metering of the dedicated transformer user is abnormal; Public transformer user abnormality judgment: Use the Pearson correlation coefficient calculation formula to calculate the correlation r2 between the public transformer area line loss data and the line line loss data; calculate the DTW distance D2 between the public transformer area line loss data and the line line loss data; Among them, when r2∈[-1,-0.8] and D2∈(D t2 ,+∞), confirming that the public transformer user metering is abnormal.

5. The line loss analysis method based on carbon flow tracking according to claim 4, characterized in that: The formula for calculating the correlation r in the Pearson correlation coefficient calculation formula in claim 4 is: Among them, X i Indicates the power consumption data of private transformer users or line loss data of public transformer areas, Y i Indicates line loss data. For X i The mean of Y i The mean of r∈[0.8,1] indicates that the two are strongly positively correlated; r∈[-1,-0.8] indicates that the two are strongly negatively correlated.

6. The line loss analysis method based on carbon flow tracking according to claim 4, characterized in that: The DTW dynamic time warping distance calculation process in claim 4 is: Given two time series X = {x1, x2, ..., x m } and Y={y1,y2,...,y n }, construct the cumulative distance D(i,j), where X is the time series set of the dedicated transformer user power data or the public transformer area line loss data, and Y is the time series set of the line line loss data: D(i,j)=d(x i ,y j )+min{D(i-1,j),D(i,j-1),D(i-1,j-1)} Where: d(x i ,y j )=(x i -y j ) 2 ,1≤i≤m,1≤j≤n, the initial condition is D(1,1)=d(x i ,y j ); The range reached by the curved path W in the distance matrix is ​​called the curved window, and the final metric distance is defined as: Where K is the length of the curved path W; set the threshold D t1 , D t2 , D t1 <D t2 ,D∈[0,D t1 ) indicates that the two are strongly positively correlated; D∈(D t2 ,+∞) indicates that the two are strongly negatively correlated.

7. The line loss analysis method based on carbon flow tracking according to claim 1, characterized in that: The step S4 of analyzing the abnormal power consumption of low-voltage users includes: Determination of voltage loss and current abnormality for three-phase four-wire low-voltage users: For three-phase four-wire low-voltage users, the voltage lower limit threshold is defined as U L1 , analyze the current value at the time of historical voltage loss, and when the three conditions are met at the same time, it is determined that the low-voltage user has voltage loss and current abnormality: Condition C41: There is a daily abnormal moment in a day, the daily abnormal moment is the voltage loss point moment, and at least one phase sequence satisfies that the absolute value of the current value is not less than the low-voltage current threshold; Condition C42: The proportion of the number of daily abnormal points in at least one phase sequence to the total number of daily sampling points is not less than the preset percentage P1, and the number of daily abnormal points is the number of daily abnormal moments; Condition C43: The number of consecutive days of daily abnormalities is not less than N3, and the number of consecutive days of daily abnormalities refers to the number of consecutive days that meet condition C42; Judgment of abnormality of single-phase voltage exceeding lower limit for low-voltage users: For single-phase low-voltage users, the voltage lower limit threshold is defined as U L2 When two conditions are met at the same time, it is determined that the low-voltage user has a single-phase voltage exceeding the lower limit abnormality: Condition C51: The proportion of daily abnormal points to the total number of daily sampling points is not less than the preset percentage P1; Condition C52: The number of consecutive days of daily abnormalities is not less than N3, and the number of consecutive days of daily abnormalities refers to the number of consecutive days that meet condition C51; Abnormal judgment of single-phase voltage exceeding upper limit for low-voltage users: For single-phase low-voltage users, the voltage upper limit threshold is defined as U L3 , analyze the current value at the time of historical voltage loss, and when two conditions are met at the same time, it is determined that the low-voltage user has a single-phase voltage exceeding the upper limit abnormality: Condition C61: The proportion of daily abnormal points to the total number of daily sampling points is not less than the preset percentage P1; Condition C62: The number of consecutive days of daily abnormalities is not less than N3, and the number of consecutive days of daily abnormalities refers to the number of consecutive days that meet condition C61; Determination of abnormal current imbalance for three-phase four-wire low-voltage users: For three-phase four-wire low-voltage users, the daily average values ​​of the absolute values ​​of the three-phase four-wire currents I1, I2, and I3 are not 0A, and the daily average value of the absolute value of the current of one phase is less than the daily average value of the absolute value of the current of the other two phases. The phase current corresponding to the minimum average value is defined as unbalanced. The correlation coefficient P of the time point data of the other two phase currents is calculated according to the Pearson correlation coefficient formula. When P∈[0.8,1], it is confirmed to meet the positive strong correlation. The daily sampling values ​​of the two-phase currents that are positively correlated with each other are analyzed, and the absolute values ​​of the two-phase currents are sorted from high to low. The time points corresponding to the first three values ​​of each phase are used as the three maximum time points of each phase. When the four conditions are met at the same time, it is determined that the low-voltage user has a current imbalance abnormality: Condition C71: The first three values ​​of each phase are greater than the low voltage current threshold; Condition C72: Among the three maximum time points of the two phases, there is a maximum time point that occurs on the same slice; Condition C73: The absolute value of the active power at any maximum time point of any phase is greater than the absolute value of the reactive power; Condition C74: At any maximum time point slice of any phase, the ratio of the current at the maximum time point to the phase sequence current corresponding to the minimum average value is higher than the preset ratio P L .

8. A line loss analysis system based on carbon flow tracking, characterized in that: Used to execute the method according to any one of claims 1 to 7.

Images