A method for judging the cause of abnormal line loss of multi-dimensional characteristics of a distribution line

Abstract

The present invention relates to a method for judging the cause of abnormal line loss of multi-dimensional characteristics of a distribution line. First, the electric energy, current and voltage data of the line checkpoint and the subordinate distribution transformers for 30 consecutive days are obtained and preprocessed. Then, the detection of incorrect line-transformer relationship, the detection of asynchronous clock of meters, the detection of incorrect multiplier of distribution transformer meters, the detection of missing meter acquisition, the calculation of theoretical line loss, the detection of over-capacity operation of distribution transformers, the detection of light load of the line, etc. are carried out successively to realize the analysis of the cause of abnormal line loss of 10kV lines. The present invention applies big data technology to realize the value-added monetization of existing data and efficiently analyze the cause of abnormal line loss in 10kV lines.

Description

Technical Field

[0001] The present invention belongs to the technical field of power transmission, and particularly relates to a method for judging the reasons for abnormal line losses of multi-dimensional characteristics of distribution lines. Background Art

[0002] The line loss rate is a comprehensive technical and economic indicator reflecting the level of power production technology and operation management. On the one hand, due to the impedance of components such as lines and transformers, technical line losses will occur. On the other hand, due to management problems in various links of power grid operation and management, power losses will occur in each link of power production, transmission, and distribution, which is called management line loss. There are many reasons for the line losses of distribution transformers on the 10kV lines of the distribution network, including missing meter readings of distribution transformers, out-of-sync clock of meters, incorrect line-transformer relationship, metering failures, incorrect metering multiples, etc. At present, power supply enterprises mainly manage the operation of line-loss distribution transformers through large-scale on-site manual inspections, arranging professional personnel to manually export the power data of each distribution transformer from the electric energy management system for analysis, which requires a large number of personnel, consumes a lot of time, has low efficiency, and limited effects. Summary of the Invention

[0003] The purpose of the present invention is to solve the problem of unclear reasons for abnormal line losses in 10kV lines, and provide a method for judging the reasons for abnormal line losses of multi-dimensional characteristics of distribution lines. By making full use of the data center of power data, deeply exploring the data value, without adding new devices or modifying equipment, applying big data technology to realize value-added transformation of existing data, and achieving efficient analysis of the reasons for abnormal line losses in 10kV lines.

[0004] The present invention is realized through the following technical solutions. A method for judging the reasons for abnormal line losses of multi-dimensional characteristics of distribution lines includes the following steps:

[0005] Step S1: Obtain the electric energy, current, and voltage data of the line gateway and its subordinate distribution transformers and perform preprocessing;

[0006] Step S2: Use the Pearson correlation of the line voltage time series data to judge the line-transformer relationship, and determine whether the abnormal line loss is caused by an abnormal line-transformer relationship;

[0007] Step S3: Judge whether there is an abnormality in the clock of the meter;

[0008] Step S4: Use the principle of conservation of electric energy to establish and solve the line loss coefficient equation of the distribution transformer, and use the solved line loss coefficient to judge whether the abnormal line loss is caused by an abnormal multiple of the distribution transformer;

[0009] Step S5: By obtaining the data missing situation in the data set, judge whether the reason for the abnormal line loss is caused by missing meter readings;

[0010] Step S6, determine whether the reason for abnormal line loss is caused by excessive theoretical line loss through the calculation of theoretical line loss of the line;

[0011] In step S7, by judging whether the three-phase current I 实 of the distribution transformer exceeds the rated operating current I N of the distribution transformer, determine whether the reason for abnormal line loss is caused by over-capacity operation of the distribution transformer;

[0012] In step S8, calculate the percentage of the fixed loss L T of the distribution transformer in the theoretical line loss L L , and judge whether the reason for line loss is caused by light load of the line;

[0013] In step S9, extract the characteristics of each line through steps S2 - S8 to obtain the following characteristic matrix,

[0014]

[0015] where, P FTh is the abnormality of the line-transformer relationship of the h-th line, P Lh is the judgment result of whether there is an abnormality in the meter clock of the h-th line, S th is the abnormality of the distribution transformer magnification of the h-th line, S dh is the judgment result of whether there is meter acquisition missing for the h-th line, S Rh is the judgment result of whether the theoretical line loss is too high for the h-th line, S Sh is the judgment result of whether the distribution transformer is over-capacity operated for the h-th line, L Lh is the light load judgment result of the h-th line; if the eigenvalue in the characteristic matrix is 1, it means that there is a corresponding reason for abnormal line loss, and the conclusion of the reason for abnormal line loss is obtained through the characteristic matrix.

[0016] Further preferably, in step S2, the Pearson correlation coefficient matrix is calculated according to the line voltage time series:

[0017]

[0018] where P ij represents the Pearson correlation coefficient between the i-th distribution transformer and the j-th distribution transformer; the mean value P mean，i of the i-th column represents the correlation between the i-th distribution transformer and other distribution transformers. If P mean，i > the set threshold, it is considered that the line-transformer relationship is correct, and P FT is recorded as 0, otherwise the line-transformer relationship is incorrect, and P FT is recorded as 1.

[0019] In step S3, let E d0 be the input power on the d0-th day, ed0 Let \(E\) be the electricity sales volume on the \(d_0\)th day. d0 \(E\) d0+1 \(E\) d0+z \(\cdots\), \(E\) d0 Let \(E\) be the input electricity volume on the \((z + 1)\)th day, and \(e\) d0+1 \(e\) d0+z \(\cdots\), \(e\) d0 \(E\) d0+1 \(E\) d0+z \(\cdots\), \(E\) d0 and \(e\) d0+1 \(e\) d0+z \(\cdots\), \(e\) d0-1 \(E\) d0 \(E\) d0+z-1 \(\cdots\), \(E\) d0 and \(e\) d0+1 \(e\) d0+z \(\cdots\), \(e\) d0+1 \(E\) d0+2 \(E\) d0+z+1 \(\cdots\), \(E\) d0 and \(e\) d0+1 \(e\) d0+z \(\cdots\), \(e\) L Calculate the Pearson correlation coefficient \(P\) between \(E\) L and \(e\), the correlation coefficient \(P'\) between \(E\)

[0020] Furthermore, preferably, in step S4, the active electrical energy of the 10 kV line is conserved. Therefore, within any time period, there is the following relationship:

[0021] \(M\) 0 \((1 + \delta\) 0 ) = \(M\) 1 \((1 + \delta\) 1 ) + \(M\) 2 \((1 + \delta\) 2 ) + \(\cdots\) + \(M\) n \((1 + \delta\) n ) + \(M\) 损 (2)

[0022] In the formula, \(M\) 0 is the gateway watt-hour meter of the 10 kV line, and \(\delta\) 0 is the measurement error of the gateway watt-hour meter; \(M\) 1 , \(M\) 2 , \(M\) 3 , \(M\) 4 \(\cdots\), \(M\) n are the watt-hour meters of distribution transformers 1 to \(n\) on the line respectively, and \(\delta\) 1 , \(\delta\) 2 , \(\delta\) 3 , \(\delta\) 4…δ n are the line loss coefficients of distribution transformers 1 to distribution transformer n, respectively, and M 损 is the active power loss;

[0023] Assume that δ 0 , δ 1 , δ 2 , δ 3 , δ 4 …δ n is fixed within a certain load range. At the same time, assume that M 损 and the total energy M 0 at the gateway are in a proportional relationship. Assume the proportionality is a constant k. Therefore, substituting the readings of the watt-hour meters M1, M2, M3, M4…Mn in m time periods into Equation (2) respectively, the following system of equations is obtained:

[0024]

[0025] In the formula, respectively represent the readings of the watt-hour meters M 0 , M 1 , M 2 …M n in the first time period; respectively represent the readings of the watt-hour meters M 0 , M 1 , M 2 …M n in the second time period; and so on, respectively represent the readings of the watt-hour meters M 0 , M 1 , M 2 …M n in the m-th time period;

[0026] Therefore, Equation (3) is an n-variable system of equations. From the knowledge of linear algebra, when the meter readings in each time period are non-linearly correlated, the system of equations has a unique solution when m is equal to n. At the same time, it is known that given any error in δ 0 , δ 1 , δ 2 , …δ n , the errors of other meters can be solved according to the readings; if δ i >>1, then the distribution transformer is the cause of abnormal line loss, denoted as S t , and S t belongs to 0 or 1, where 0 represents no abnormal distribution transformer and 1 represents an abnormal distribution transformer.

[0027] Further preferably, in step S5, if the cause of abnormal line loss is due to data loss, then record S d as 1. If the cause of abnormal line loss is not due to data loss, then record Sd is 0.

[0028] Further preferably, in step S6, calculate the theoretical line loss rate L R = ΔA / E, where E is the line input power. If the difference between the actual line loss rate and the theoretical loss rate is greater than 2%, the reason for the line loss is excessive theoretical line loss, and record S R as 1. Otherwise, the reason for the line loss is not excessive theoretical line loss, and record S R as 0.

[0029] Further preferably, in step S7, if I 实 > I N , there is a situation of over-capacity operation of the distribution transformer, and record S S as 1. Otherwise, the distribution transformer is not over-capacity, and record S S as 0.

[0030] Further preferably, in step S8, if L T / L L > the set ratio, the line is lightly loaded, and record L L as 1. Otherwise, the line load is normal, and record L L as 0.

[0031] The present invention classifies the common reasons for abnormal line loss and gradually analyzes the reasons for abnormal line loss by using the power, current, and voltage data of the line gateway and its subordinate distribution transformers for 30 consecutive days. That is, first obtain the power, current, and voltage data of the line gateway and its subordinate distribution transformers for 30 consecutive days and perform preprocessing, and then successively perform line-transformer relationship error detection, meter clock out-of-synchronization detection, distribution transformer meter multiplier error detection, meter acquisition missing detection, theoretical line loss calculation, distribution transformer over-capacity operation detection, line light load detection, etc., to realize the analysis of the reasons for abnormal line loss of 10kV lines. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 is the flowchart of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0033] The present invention will be further described in detail below.

[0034] Referring to Figure 1 , a method for judging the reasons for abnormal line loss of multi-dimensional characteristics of a distribution line includes the following steps:

[0035] Step S1, obtain the power, current, and voltage data of the line gateway and its subordinate distribution transformers and perform preprocessing. The required data is shown in Table 1.

[0036] Table 1. Data required for analysis of reasons for abnormal line loss

[0037] Data type Data length Function Electric energy Daily power consumption for a consecutive month Detection of abnormal distribution transformers Voltage Voltage at 96 points for 7 consecutive days Detection of the relationship between line and transformer Current Current at 96 consecutive points in a day Detection of over-capacity operation of distribution transformers

[0038] Step S2, using the Pearson correlation of the line voltage time series data to identify the line-to-line relationship, to determine whether the abnormal line loss is caused by the abnormal line-to-line relationship.

[0039] Pearson correlation coefficient P x,y It is obtained by dividing the covariance by the standard deviation of the two variables. The calculation formula is as follows:

[0040]

[0041] Where cov(x,y) is the covariance of the two line voltage time series; σ x ,σ y are the standard deviations of the two line voltage time series. The Pearson correlation coefficient is a value between -1 and 1. When the Pearson correlation coefficient is 0, there is no relationship between the two variables x and y. When the value of x increases (decreases), the value of y increases (decreases), the two variables are positively correlated, and the Pearson correlation coefficient is between 0 and 1. When the value of x increases (decreases), the value of y decreases (increases), the two variables are negatively correlated, and the Pearson correlation coefficient is between -1 and 0. Generally, the strength of the correlation is shown in Table 2.

[0042] Table 2. Pearson correlation coefficient strength comparison table

[0043] Correlation Negative Positive None -0.09 to 0.0 0.0 to 0.09 Weak -0.3 to -0.1 0.1 to 0.3 Medium -0.5 to -0.3 0.3 to 0.5 Strong -1.0 to -0.5 0.5 to 1.0

[0044] The Pearson correlation coefficient is used to check whether the line-to-line relationship is correct. The correlation threshold can be used as a reference, but the interpretation of the Pearson correlation coefficient depends on the specific application background and purpose. Therefore, assuming the threshold δ of the correlation between the two line voltage time series, if the obtained Pearson correlation coefficient P x,y >δ, it means that there is a strong correlation between the two, and the line-to-line relationship of the two distribution transformers is consistent. x,y <δ, it means that the correlation between the two is poor and the line-to-line relationship of the two distribution transformers is inconsistent.

[0045] The Pearson correlation coefficient matrix is ​​calculated:

[0046]

[0047] Where P ij represents the Pearson correlation coefficient between the i-th distribution transformer and the j-th distribution transformer. The mean value P of the i-th column mean，i represents the correlation between the ith distribution transformer and other distribution transformers. If P mean，i >0.8, the linear relationship is considered correct, and P FT is 0, otherwise the linear relationship is wrong, record P FT is 1.

[0048] Step S3, determine whether there is an abnormality in the meter clock. Let E d0 be the input power on the d0th day, and e d0 be the sold power on the d0th day. Then set E d0 , E d0+1 , …, E d0+z as the input power for z + 1 days, and e d0 , e d0+1 , …, e d0+z as the sold power for z + 1 days. Calculate the Pearson correlation P between E d0 , E d0+1 , …, E d0+z and e d0 , e d0+1 , …, e d0+z . Also calculate the correlation P′ between E d0-1 , E d0 , …, E d0+z-1 and e d0 , e d0+1 , …, e d0+z , and the correlation P″ between E d0+1 , E d0+2 , …, E d0+z+1 and e d0 , e d0+1 , …, e d0+z . If P < MAX(P′, P″), then the meter clock has an abnormality, denoted as P L , where P L belongs to 0 or 1, 0 representing that the clock is normal and 1 representing that the clock is abnormal.

[0049] Step S4, utilize the principle of conservation of electrical energy to establish and solve the line loss coefficient equation of the distribution transformer, and use the obtained line loss coefficient to determine whether the line loss abnormality is caused by the abnormal ratio of the distribution transformer.

[0050] The active electrical energy of the 10kV line is conserved. Therefore, within any time period, there is the following relationship:

[0051] M 0 (1 + δ 0 ) = M 1 (1 + δ 1 ) + M 2 (1 + δ 2 ) + … + M n (1 + δ n ) + M 损 (2)

[0052] In the formula, M 0 is the 10kV line gateway watt-hour meter, and δ 0 is the measurement error of the gateway watt-hour meter; M1 , M 2 , M 3 , M 4 …M n are the watt-hour meters of distribution transformers 1 to n on the line respectively, δ 1 , δ 2 , δ 3 , δ 4 …δ n are the line loss coefficients of distribution transformers 1 to n respectively, M 损 is the active power loss.

[0053] Assume that δ 0 , δ 1 , δ 2 , δ 3 , δ 4 …δ n are fixed within a certain load range. At the same time, assume that M 损 is in a proportional relationship with the total energy M 0 at the gateway. Assume the proportionality is a constant k. Therefore, substituting the readings of the watt-hour meters M1, M2, M3, M4…Mn in m time periods into Equation (2) respectively, the following system of equations can be obtained:

[0054]

[0055] In the formula, respectively represent the readings of the watt-hour meters M 0 , M 1 , M 2 …M n in the first time period; respectively represent the readings of the watt-hour meters M 0 , M 1 , M 2 …M n in the second time period; and so on, respectively represent the readings of the watt-hour meters M 0 , M 1 , M 2 …M n in the mth time period.

[0056] Therefore, Equation (3) is an n - variable system of equations. From the knowledge of linear algebra, when the meter readings in each time period are non - linearly correlated, when m equals n, the system of equations has a unique solution. At the same time, it can be known that when any error in δ 0 , δ 1 , δ 2 , …δ n is known, the errors of other meters can be solved according to the readings. If δ i >>1, then the reason for the abnormal line loss of this distribution transformer is recorded as St , S t belongs to 0 or 1, where 0 represents no abnormality in the distribution transformer, and 1 represents the existence of an abnormality in the distribution transformer.

[0057] Step S5: By obtaining the data missing situation in the dataset, determine whether the reason for the abnormal line loss is caused by missing meter readings. The data missing situation of the distribution transformer is shown in Table 3. If the abnormal line loss is caused by data missing, record S d as 1, if the reason for the abnormal line loss is not caused by data missing, record S d as 0.

[0058] Table 3. Data Missing Situation of Distribution Transformer

[0059]

[0060] Step S6: Determine whether the reason for the abnormal line loss is caused by excessive theoretical line loss through the calculation of the theoretical line loss of the line.

[0061] The average current loss time method is a simple method for estimating the theoretical loss of the line. First, assume that the line supplies power to a concentrated load. During the calculation time T = NΔt, N is the number of current acquisitions, Δt is the current acquisition time interval, and let I i be the i-th current value, I av be the algebraic average of the currents, I i,dif be I i deviating from I av by the difference, and we can get:

[0062]

[0063] I i,dif (t) = I i - I av (5)

[0064] Let R be the constant impedance, and the theoretical line loss ΔA can be expressed as:

[0065]

[0066] Since the algebraic sum of I i,dif is 0, we can get:

[0067]

[0068] Further simplification gives:

[0069]

[0070]

[0071]

[0072] In the formula, T av is the average current effect required to generate equivalent loss, which is called the average current loss time.

[0073] Calculate the theoretical line loss rate L R = ΔA / E, where E is the input power of the line. If the actual line loss rate minus the theoretical loss rate is greater than 2%, the reason for the line loss is that the theoretical line loss is too large, and record S R as 1. Otherwise, the reason for the line loss is not that the theoretical line loss is too large, and record S R as 0.

[0074] Step S7, by judging whether the three-phase current I 实 of the distribution transformer exceeds the rated operating current I N of the distribution transformer, judge whether the reason for the abnormal line loss is caused by the over-capacity operation of the distribution transformer. If I 实 > I N then there is an over-capacity situation of the distribution transformer, and record S S as 1. Otherwise, the distribution transformer is not over-capacity, and record S S as 0.

[0075] Step S8, calculate the percentage of the fixed loss L T of the distribution transformer in the theoretical line loss L L to judge whether the reason for the line loss is caused by light load of the line. If L T / L L > 0.5, the line is lightly loaded, and record L L as 1. Otherwise, the line load is normal, and record L L as 0.

[0076] Step S9, through Steps S2 - S8, extract the characteristics of each line to obtain the following characteristic matrix,

[0077]

[0078] where P FTh is the abnormality of the line-transformer relationship of the h-th line, P Lh is the judgment result of whether there is an abnormality in the meter clock of the h-th line, S th is the abnormality of the transformer ratio of the distribution transformer of the h-th line, S dh is the judgment result of whether there is a missing meter reading of the h-th line, S Rh is the judgment result of whether the theoretical line loss of the h-th line is too high, S Sh is the judgment result of whether the distribution transformer of the h-th line is over-capacity operated, L Lh is the judgment result of the light load of the h-th line; if the eigenvalue in the characteristic matrix is 1, it means that there is a corresponding reason for the abnormal line loss, and there may be multiple abnormal reasons at the same time. The conclusion of the abnormal line loss reason is obtained through the characteristic matrix.

[0079] For example, the eigenvalue of a certain line in the feature matrix is as follows: 0 0 1 0 0 0 0

[0081] It is concluded from the feature matrix that the abnormal line loss of this line is caused by the abnormal ratio of the distribution transformer.

[0082] To analyze the reason for the abnormal line loss of a certain 10kV line in June, the gateway power of the line and the power data of each distribution transformer under the line are obtained from the data Zhongtai, as shown in Table 4.

[0083] Table 4. Gateway power data

[0084]

[0085]

[0086] Table 5. Power data of distribution transformers (partial)

[0087]

[0088] According to steps S2 - S8, the possible reasons for the abnormal line loss of this line are analyzed step by step. Successively, the detection of incorrect line-transformer relationship, the detection of out-of-phase clock of meters, the detection of incorrect ratio of distribution transformer meters, the detection of missing meter acquisition, the calculation of theoretical line loss, the detection of over-capacity operation of distribution transformers, and the detection of light load of the line are carried out.

[0089] Table 6. Line loss coefficients of each distribution transformer

[0090]

[0091] It can be seen from the calculation results in Table 6 that the line loss coefficients of distribution transformer 9 and distribution transformer 14 are much greater than 1. The operation mode of this line in June has not changed, the line-transformer relationship is correct, and the gateway metering is accurate. However, through power verification, it is found that the monthly power consumption of distribution transformer 9 is small, and its missing power is only 15768.34 kW·h. The theoretical line loss of the line is 32205.6 kW·h, and the actual power loss of the line is 98966.69 kW·h, with an abnormal loss of 66761.09 kW·h. Therefore, distribution transformer 9 is excluded. The missing power of distribution transformer 14 is 70612 kW·h, which is close to the abnormal loss of the line. Therefore, there is an abnormal line loss situation in distribution transformer 14. The above embodiments can prove that this method can effectively analyze the reasons for the abnormal line loss of 10kV lines.

Claims

1. A method for judging the cause of abnormal line loss of multi-dimensional characteristics of a distribution line, Characterized in that, It includes the following steps: Step S1, obtain the electric energy, current, and voltage data of the line gateway and affiliated distribution transformers and perform preprocessing; Step S2, use the Pearson correlation of the line voltage time series data to judge the line-transformer relationship, and judge whether the abnormal line loss is caused by the abnormal line-transformer relationship; Step S3, judge whether there is an abnormality in the meter clock; In step S3, let be the input power on the d0th day, be the sold power on the d0th day, then let be the input power on the (z + 1)th day, be the sold power on the (z + 1)th day, calculate the Pearson correlation P between and , the correlation between and as well as the correlation between and . If , then the timing clock is abnormal, denoted as P L , where P L belongs to 0 or 1, 0 represents that the clock is normal, and 1 represents that the clock is abnormal; Step S4, use the principle of conservation of electrical energy to establish and solve the line loss coefficient equation of the distribution transformer, and use the obtained line loss coefficient to judge whether the abnormal line loss is caused by the abnormal ratio of the distribution transformer; In step S4, the active electrical energy of the 10kV line is conserved. Therefore, in any time period, there is the following relationship: （2）； where M 0 is the total energy at the grid connection point, and δ 0 is the metering error of the watt-hour meter at the grid connection point; M 1 , M 2 , M 3 , M 4 , …, M n are the readings of the watt-hour meters of distribution transformers 1 to n on the line respectively, and δ 1 , δ 2 , δ 3 , δ 4 …, δ n are the line loss coefficients of distribution transformers 1 to n respectively, and M 损 is the active power loss; Let δ 0， δ 1 ,δ 2 ,δ 3 ,δ 4 …,δ n is fixed within a certain load range. At the same time, let M 损 be proportional to the total energy M 0 at the gateway, with a proportionality constant k. Therefore, substituting the counts of the electricity meters in m time periods into Equation (2) respectively, the following system of equations is obtained: （3）； Wherein, respectively represent the total energy at the gateway and the meter readings of the electricity meters of distribution transformers 1 to n on the line during the first period of time; respectively represent the total energy at the gateway and the meter readings of the electricity meters of distribution transformers 1 to n on the line during the second period of time; and so on, respectively represent the total energy at the gateway and the meter readings of the electricity meters of distribution transformers 1 to n on the line during the mth period of time; Therefore, Equation (3) is an n - element system of equations. According to the knowledge of linear algebra, when the meter readings within each period are non - linearly correlated, the system of equations has a unique solution when m equals n. At the same time, it is known that given δ 0， δ 1 , δ 2 , …, δ n , when any one of the errors is known, the errors of other meters can be solved according to the readings; if δ i >> 1, then the distribution transformer is the cause of abnormal line loss, denoted as S t , S t belongs to 0 or 1, where 0 represents no abnormality of the distribution transformer and 1 represents the existence of abnormality of the distribution transformer; Step S5, by obtaining the data missing situation of the data set, judge whether the cause of the abnormal line loss is caused by the missing meter collection; Step S6, judge whether the cause of the abnormal line loss is caused by the too high theoretical line loss through the calculation of the theoretical line loss of the line; Step S7, by judging whether the three-phase current I of the distribution transformer 实 exceeds the rated operating current I of the distribution transformer N , to judge whether the abnormal line loss is caused by the over-capacity operation of the distribution transformer; Step S8, calculate the fixed loss L of the distribution transformer T accounting for the theoretical line loss L L as a percentage, and determine whether the cause of the line loss is due to light loading of the line; Step S9, through steps S2 - S8, perform feature extraction on each line to obtain the following feature matrix, ； Among them, P FTh is the abnormal judgment result of the line transformation relationship of the h-th line, and P Lh is the abnormal judgment result of whether the meter clock of the h-th line exists. S th is the abnormal judgment result of the distribution transformer magnification of the h-th line. S dh is the judgment result of whether the meter reading acquisition of the h-th line is missing. S Rh is the judgment result of whether the theoretical line loss of the h-th line is too high. S Sh is the judgment result of whether the distribution transformer of the h-th line is operating with overcapacity. L Lh is the light load judgment result of the h-th line; if the eigenvalue in the feature matrix is 1, it means that there is a corresponding line loss abnormal reason, and the conclusion of the abnormal line loss reason is obtained through the feature matrix.

2. The method for judging the cause of abnormal line loss of multi-dimensional characteristics of a distribution line according to claim 1, Characterized in that, In step S2, the Pearson correlation coefficient matrix is calculated according to the line voltage time series: （1）； where P ij represents the Pearson correlation coefficient between the i-th distribution transformer and the j-th distribution transformer; the mean value of the i-th column P mean，i represents the correlation between the i-th distribution transformer and other distribution transformers. If P mean，i > the set threshold, it is considered that the line-transformer relationship is correct, and P FT is recorded as 0. Otherwise, the line-transformer relationship is incorrect, and P FT is recorded as 1.

3. The method for judging the cause of abnormal line loss of multi-dimensional characteristics of a distribution line according to claim 1, Characterized in that, In step S5, if the reason for the abnormal line loss is data missing, record S d as 1; if the reason for the abnormal line loss is not data missing, record S d as 0.

4. The method for judging the cause of abnormal line loss of multi-dimensional characteristics of a distribution line according to claim 1, Characterized in that, In step S6, calculate the theoretical line loss rate L R =ΔA / E, where E is the input power of the line and ΔA is the theoretical line loss. If the difference between the actual line loss rate and the theoretical loss rate is greater than 2%, the reason for the line loss is excessive theoretical line loss, and record S R as 1. Otherwise, the reason for the line loss is not excessive theoretical line loss, and record S R as 0.

5. The method for judging the cause of abnormal line loss of multi-dimensional characteristics of a distribution line according to claim 1, Characterized in that, In step S7, if I 实 >I N , there is an over-capacity situation of the distribution transformer, and record S S as 1. Otherwise, the distribution transformer is not over-capacity, and record S S as 0.

6. The method for judging the cause of abnormal line loss of multi-dimensional characteristics of a distribution line according to claim 1, Characterized in that, In step S8, if L T / L L > the set ratio, the line is lightly loaded, and record L L as 1, otherwise the line load is normal, and record L L as 0.

Images