Line loss monitoring method, system and platform, terminal and storage medium
By using a multivariate linear fitting model and principal component analysis, the line loss of power lines can be monitored in real time, which solves the problem of large line loss monitoring errors in existing technologies. It achieves efficient early warning and sorting of line loss anomalies, and improves the safety, stability and energy utilization of the power grid.
Patent Information
- Application Number
- CN202411043620.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-31
- Publication Date
- 2024-11-08
AI Technical Summary
Existing line loss monitoring methods, such as allocating losses based on electricity consumption ratios and using fixed coefficients, have large errors and cannot accurately reflect the relationship between the degree of line loss of each line. This makes it difficult to quickly locate the source of the problem, increasing energy waste and safety hazards.
By collecting the electricity consumption of the main power line and branch lines in real time, the predicted value of line loss is calculated using a multiple linear fitting model. A linear regression model is then constructed, and combined with principal component analysis, the percentage difference is calculated to provide early warning and ranking of abnormal line loss.
It reduces line loss monitoring errors, helps to quickly locate the source of problems, reduces investigation time, improves energy utilization, reduces the possibility of safety accidents, and ensures the safe and stable operation of the power grid.
Smart Images

Figure CN118914752A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of line loss monitoring, and in particular to a line loss monitoring method, system, platform, terminal and storage medium. Background Art
[0002] When electricity is transmitted through lines, it will cause energy consumption loss, i.e. line loss. Currently, line loss is often measured by two methods: apportionment according to the proportion of electricity consumption or apportionment according to a fixed coefficient. Correspondingly, existing line loss monitoring often obtains line loss data by apportionment according to the proportion of electricity consumption or apportionment according to a fixed coefficient for line loss monitoring. Through line loss monitoring, these losses can be discovered and handled in a timely manner, reducing unnecessary power waste and improving the overall utilization of energy.
[0003] However, the allocation based on the proportion of electricity consumption is to calculate the proportion of electricity consumption of each line and its branches in the total line, and multiply the line loss by the proportion of electricity consumption to obtain the corresponding line loss. This method may change the line loss itself even if the parameters of the line and its branches remain unchanged. Therefore, the line loss data obtained in this way for line loss monitoring often have large errors. The allocation based on fixed coefficients is to assign a fixed coefficient to each line based on production experience or line allocation relationship. Each line multiplies its own electricity consumption by the assigned fixed coefficient to obtain its own line loss. This method requires a fixed coefficient to be assigned to each line in advance, and the actual line loss cannot be calculated. The line loss data obtained in this way for line loss monitoring also has large errors, which is not conducive to energy consumption monitoring and production scheduling.
[0004] In addition, since the monitoring line loss errors based on the above two line loss monitoring methods are large, they cannot truly reflect the relationship between the line loss levels of each line, which is not conducive to finding out the causes of line loss in the lines from high to low after discovering abnormal line loss, especially when abnormal line loss occurs on multiple lines. On the one hand, it is not conducive to quickly locating the source of the problem, reducing the troubleshooting time, and improving the efficiency of problem solving. On the other hand, it is easy to cause electric energy to be consumed unnecessarily during transmission or distribution, resulting in energy waste, which is not conducive to improving energy utilization. On the other hand, for lines with high line loss problems, it may cause safety hazards such as equipment overheating and insulation aging. The existing technology is not conducive to troubleshooting the causes of line loss in the lines from high to low according to the relationship between line losses, which will increase the possibility of safety accidents in the lines to a certain extent, affecting the safe and stable operation of the entire power grid. Summary of the invention
[0005] The object of the present invention is to provide a line loss monitoring method, system, platform, terminal and storage medium to solve at least one of the above problems.
[0006] In a first aspect, the technical solution of the present invention provides a line loss monitoring method, which is applied to line loss monitoring of power lines, wherein the power lines include a main power line and a plurality of branch power lines for connecting user electrical equipment, and the input end of each branch power line is connected to the output end of the main power line; the method comprises:
[0007] Collect the current power consumption of the main power line and the current power consumption of each power branch line in real time, and calculate the current actual bus loss value of the power line based on the collected current power consumption of the main power line and the current power consumption of each power branch line;
[0008] Substitute the collected current power consumption of each power branch line into the pre-built multivariate linear fitting model for calculation to obtain the total power loss prediction value of the power line; the calculation formula of the multivariate linear fitting model is:
[0009] Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1)
[0010] In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept;
[0011] Calculating a percentage difference between the total bus loss prediction value and the actual total bus loss value;
[0012] If the percentage difference exceeds a preset percentage difference threshold range, an abnormal line loss warning is issued for the power line.
[0013] Furthermore, when the percentage difference exceeds a preset percentage difference threshold interval, the method further includes:
[0014] The power consumption on the power main line and the power consumption on each power branch line collected within a preset time period before the current time and the calculated actual bus loss value are obtained to construct a data set corresponding to the current time, and a new model is constructed using the data set corresponding to the current time and the same method as that of constructing the multivariate linear fitting model. The calculation formula of the new model is:
[0015] Pt′=a1′×P1+a2′×P2+…+ai′×Pi+…+an′×Pn+b1′, (2)
[0016] In formula (2), Pt′ represents the total power loss prediction value of the power line, i represents the i-th power branch line among the plurality of power branch lines, Pi represents the power consumption of the i-th power branch line, ai′ represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1′ represents the intercept;
[0017] Calculate the percentage difference between each pair of model coefficients ai and ai′ in formula (1) and formula (2) respectively;
[0018] The power branch lines of the power line are sorted in descending order according to the percentage differences between the calculated pairs of model coefficients;
[0019] Output the sorting results of each power branch line to show the user the sorting of line loss anomalies of each power branch line from high to low.
[0020] Furthermore, the method for constructing the multivariate linear fitting model includes:
[0021] Constructing a training set: collecting historical power consumption on the main power line, and correspondingly collecting historical power consumption on each power branch line under the historical power consumption on the main power line; using each collected historical power consumption on the main power line and the historical power consumption on each power branch line under each collected historical power consumption on the main power line, correspondingly calculating the actual historical main power loss value of the power line; constructing the training set based on the collected historical power consumption on the main power line, the historical power consumption on each power branch line, and each actual historical main power loss value of the power line calculated;
[0022] A linear regression model is established between the total power line loss prediction value Pt and the total power consumption. The calculation formula of the model is:
[0023] Pt=a0×P0+b0, (3)
[0024] In formula (3), P0 represents the total power consumption of the power line, a0 is the regression coefficient, and b0 is the intercept; the power consumption on the power bus line is the total power consumption of the power line;
[0025] Substitute the actual historical bus loss value of the power line in the training set and the historical power consumption on the corresponding power bus into formula (3), and then determine the specific values of a0 and b0 in formula (3) by data fitting method, and substitute the specific values of a0 and b0 back into formula (3) to obtain the constructed linear regression model;
[0026] Substitute each historical power consumption on the power bus line in the training set into the constructed linear regression model, and calculate the predicted value of the historical power line loss corresponding to each historical power consumption on the power bus line;
[0027] The percentage errors between the predicted historical total bus loss values of each power line and the actual historical total bus loss values in the training sets corresponding to each power line are calculated respectively, and recorded as the first percentage error;
[0028] Calculate the mean μ and standard deviation σ of each first percentage error;
[0029] Based on the mean μ and the standard deviation σ, the percentage difference threshold interval is obtained, and the percentage difference threshold interval is [μ-3σ, μ+3σ];
[0030] The data corresponding to the first percentage error exceeding the percentage difference threshold interval [μ-3σ,μ+3σ] in the training set are deleted to obtain a new data set;
[0031] Normalize the data in the new data set to obtain a normalized data set;
[0032] The historical power consumption on each power branch line in the normalized data set is analyzed by principal component analysis to obtain a set of selected principal components Fj and the eigenvector dj corresponding to each selected principal component Fj; where dj = (d j1 ,d j2 ,…,d jk ,…,d jn ), j = 1, 2, ..., n, the expression of each selected principal component Fj is:
[0033] F1=d 11 P1+d 12 P2+…+d 1k Pk+…+d 1n Pn,
[0034] F2=d 21 P1+d 22 P2+…+d 2k Pk+…+d 2n Pn,
[0035] F3=d 31 P1+d 32 P2+…+d 3k Pk+…+d 3n Pn,
[0036] …
[0037] Fj=d j1 P1+d j2 P2+…+djk Pk+…+d jn Pn,
[0038] …
[0039] Fm=d m1 P1+d m2 P2+…+d mk Pk+…d mn Pn,
[0040] Wherein, m represents the total number of all selected principal components, Fj represents the jth principal component among all selected principal components, Pk represents the power consumption of the kth power branch line among the plurality of power branch lines, and d jk is the value of the kth element in the eigenvector dj corresponding to the jth principal component, k = 1, 2, ..., n;
[0041] Substitute each set of values of P1, P2, ..., Pn in the normalized data set into the expression of each selected principal component Fj, and calculate each set of values of F1, F2, F3, ..., Fj, ..., Fm;
[0042] All selected principal components are used to construct a multivariate linear fitting equation for calculating the total total power loss prediction value Pt of the power line. The calculation formula of the equation is as follows:
[0043] Pt=e 1 ×F1+e 2 ×F2+…+e j ×Fj+…+e m ×Fm+b, (4)
[0044] In formula (4), e j represents the coefficient of the term where Fj is located, and b represents the intercept;
[0045] Substitute each set of values of each selected principal component and its corresponding actual historical total bus loss value in the normalized data set into formula (4), and then determine e in formula (4) by data fitting method. 1 、e 2 ,…,e j ,…,e m , b specific value, and then e 1 、e 2 ,…,e j ,…,e m Substitute the specific values of , b into formula (4) to obtain the constructed multivariate linear fitting equation;
[0046] Substituting the expressions of F1, F2, F3, ..., Fj, ..., Fm into the constructed multivariate linear fitting equation, the following formula is obtained:
[0047] Pt=(e 1 d 11 +e 2 d 21 +…+e m d m1 )×P1+(e 1 d 12 +e 2 d 22 +…+e m d m2 )×P2+…+(e 1 d 1j +e 2 d 2j +…+e m d mj )×Pj+…+(e 1 d 1n +e 2 d 2n +…+e m d mn )×Pn+b; (5)
[0048] Formula (5) is the same as Formula (1), that is, the multivariate linear fitting model is obtained, wherein the coefficients of each item P1, P2, ..., Pn in Formula (1) are equal to the coefficients of the corresponding items in Formula (5), and the intercept b1 in Formula (1) is equal to the intercept b in Formula (5).
[0049] Furthermore, before substituting the actual historical bus loss value of the power line in the training set and the historical power consumption on the corresponding power bus line into formula (3), the method further includes:
[0050] Clean the training set to remove abnormal and empty data;
[0051] Standardize the cleaned data in the training set.
[0052] Furthermore, the least squares method is used for data fitting in the methods.
[0053] In a second aspect, the present invention provides a line loss monitoring system, which is applied to line loss monitoring of power lines, wherein the power lines include a main power line and a plurality of branch power lines for connecting user electrical equipment, and the input end of each branch power line is connected to the output end of the main power line; the system includes:
[0054] The first module is used to collect the current power consumption of the power main line and the current power consumption of each power branch line in real time, and calculate the current actual bus loss value of the power line based on the collected current power consumption of the power main line and the current power consumption of each power branch line;
[0055] The second module is used to substitute the collected current power consumption of each power branch line into a pre-built multivariate linear fitting model for calculation to obtain a total power loss prediction value of the power line; the calculation formula of the multivariate linear fitting model is:
[0056] Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1)
[0057] In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept;
[0058] The third module is used to calculate the percentage difference between the total line loss prediction value and the actual total line loss value; if the percentage difference exceeds a preset percentage difference threshold range, an abnormal line loss warning is issued for the power line.
[0059] In a third aspect, the present invention provides a line loss monitoring platform, which is applied to line loss monitoring of power lines. The power lines include a main power line and a plurality of branch power lines for connecting user electrical equipment, and the input end of each branch power line is connected to the output end of the main power line; smart meters are installed on the main power line and each branch power line, the smart meter installed on the main power line is the main meter, and the smart meters installed on each branch power line are branch meters;
[0060] The platform includes a power collection module and a first calculation module:
[0061] The power collection module is connected to the main power meter and each branch power meter to collect the current power consumption on the power main line and the current power consumption of each power branch line in real time;
[0062] The first calculation module is connected to the power collection module and is used to calculate the current actual bus loss value of the power line based on the current power consumption of the main power line and the current power consumption of each power branch line collected by the power collection module;
[0063] The platform also includes a second computing module, a third computing module and a decision module;
[0064] The second calculation module is connected to the power collection module, and is used to substitute the current power consumption of each power branch line collected by the power collection module into a pre-built multivariate linear fitting model for calculation to obtain a total power loss prediction value of the power line; the calculation formula of the multivariate linear fitting model is:
[0065] Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1)
[0066] In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept;
[0067] The third calculation module is connected to the second calculation module and the first calculation module, and is used to calculate the percentage difference between the bus loss prediction value and the actual bus loss value;
[0068] The decision module is connected to the third calculation module, and is used to issue an abnormal line loss warning to the power line when the percentage difference calculated by the third calculation module exceeds a preset percentage difference threshold range.
[0069] Furthermore, the platform also includes a display module and a storage module;
[0070] The power collection module is connected to the storage module and is used to store the collected power consumption on the power main line and the power consumption of each power branch line into the storage module in real time;
[0071] The first calculation module is connected to the storage module and is used to store the actual bus loss value of the power line calculated based on the power consumption of the power main line and the power consumption of each power branch line collected by the power collection module into the storage module;
[0072] The decision module is connected to the storage module, and is used to perform the following steps when the percentage difference calculated by the third calculation module exceeds a preset percentage difference threshold interval:
[0073] The power consumption on the power bus line and the power consumption on each power branch line collected within a preset time period before the current time and the calculated actual bus loss value stored in the storage module are obtained to construct a data set corresponding to the current time, and the data set corresponding to the current time is used to construct a new model using the same construction method as the multivariate linear fitting model. The calculation formula of the new model is:
[0074] Pt′=a1′×P1+a2′×P2+…+ai′×Pi+…+an′×Pn+b1′, (2)
[0075] In formula (2), Pt′ represents the total power loss prediction value of the power line, i represents the i-th power branch line among the plurality of power branch lines, Pi represents the power consumption of the i-th power branch line, ai′ represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1′ represents the intercept;
[0076] Calculate the percentage difference between each pair of model coefficients ai and ai′ in formula (1) and formula (2) respectively;
[0077] The power branch lines of the power line are sorted in descending order according to the percentage differences between the calculated pairs of model coefficients;
[0078] Output the ranking results of each power branch line to show the user the ranking of abnormal line loss of each power branch line from high to low;
[0079] The display module is connected to the decision module and is used to display the ranking results to the user.
[0080] In a fourth aspect, the present invention provides a terminal, comprising:
[0081] processor;
[0082] A memory for storing execution instructions of the processor;
[0083] The processor is configured to execute the methods described in the above aspects.
[0084] In a fifth aspect, the present invention provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the methods described in the above aspects.
[0085] Compared with the prior art, the present invention has the following beneficial effects:
[0086] The present invention realizes line loss monitoring of circuit lines through the collected power consumption and a pre-trained multivariate linear fitting model, which helps to avoid obtaining line loss data for line loss monitoring by two methods of measuring line loss: allocating according to power consumption ratio and allocating according to fixed coefficient, thereby helping to reduce monitoring errors.
[0087] The present invention can help relevant staff members understand the ranking of line losses by comparing the circuit influence coefficients in the model, and then help relevant staff members to sort the line losses, and each time check the line with the highest line loss among the remaining unchecked lines, thereby helping to quickly locate the source of the problem, reduce the troubleshooting time, and improve the efficiency of problem solving. At the same time, it can help to reduce to a certain extent the energy consumed by high line loss lines during transmission or distribution, causing energy waste, and then improve energy utilization, reduce the possibility of safety accidents occurring in the lines, and improve the safe and stable operation of the entire power grid.
[0088] The invention has reliable design principle, simple structure and very broad application prospect. BRIEF DESCRIPTION OF THE DRAWINGS
[0089] In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0090] Figure 1 is a schematic flow chart of a method according to an embodiment of the present invention;
[0091] Figure 2 is a schematic diagram of an embodiment of the power circuit described in the present invention;
[0092] Figure 3 is a schematic block diagram of a system according to an embodiment of the present invention;
[0093] Figure 4 is a schematic block diagram of a platform according to an embodiment of the present invention;
[0094] Figure 5 is a schematic block diagram of a platform according to another embodiment of the present invention;
[0095] Figure 6 A schematic diagram of the structure of a terminal provided by an embodiment of the present invention;
[0096] Figure 7 It is a structural schematic diagram of a computer storage medium provided by an embodiment of the present invention.
[0097] Among them: 200, system, 201, first module, 202, second module, 203, third module; 300, terminal, 310, processor, 320, memory, 330, communication unit; 500, platform, 501, power collection module, 502, first calculation module, 503, second calculation module, 504, third calculation module, 505, decision module, 506, display module, 507, storage module. DETAILED DESCRIPTION
[0098] To make the technical solution and advantages of the present invention more clear, the technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.
[0099] The following description of at least one exemplary embodiment is actually only illustrative and is by no means intended to limit the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0100] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.
[0101] It should be noted that the present invention uses words such as "first" and "second" to limit the corresponding parts only to facilitate the distinction of the corresponding parts. If not otherwise stated, the above words have no special meaning and therefore cannot be understood as limiting the scope of protection of the present invention.
[0102] The technical solution of the present invention will be introduced and explained below through several embodiments.
[0103] In one embodiment, the present invention provides a line loss monitoring method. The method is applied to line loss monitoring of power lines, where the power lines include a main power line and a plurality of branch power lines for connecting user electrical equipment, and the input end of each branch power line is connected to the output end of the main power line.
[0104] Please refer to Figure 1 , the method 100 comprises:
[0105] Step 110, collecting the current power consumption of the main power line and the current power consumption of each power branch line in real time, and calculating the current actual main power loss value of the power line based on the collected current power consumption of the main power line and the current power consumption of each power branch line;
[0106] Step 120, the collected current power consumption of each power branch line is substituted into a pre-built multivariate linear fitting model for calculation to obtain a total power loss prediction value of the power line; the calculation formula of the multivariate linear fitting model is:
[0107] Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1)
[0108] In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept;
[0109] Step 130, calculating the percentage difference between the bus loss prediction value and the actual bus loss value;
[0110] Step 140: If the percentage difference exceeds a preset percentage difference threshold range, an abnormal line loss warning is issued to the power line.
[0111] In step 130, the calculation formula of the percentage difference between the total bus loss prediction value and the actual total bus loss value is:
[0112]
[0113] Optionally, when the percentage difference exceeds a preset percentage difference threshold interval, the method 100 further includes:
[0114] The power consumption on the power main line and the power consumption on each power branch line collected within a preset time period before the current time and the calculated actual bus loss value are obtained to construct a data set corresponding to the current time, and a new model is constructed using the data set corresponding to the current time and the same method as that of constructing the multivariate linear fitting model. The calculation formula of the new model is:
[0115] Pt′=a1′×P1+a2′×P2+…+ai′×Pi+…+an′×Pn+b1′, (2)
[0116] In formula (2), Pt′ represents the total power loss prediction value of the power line, i represents the i-th power branch line among the plurality of power branch lines, pi represents the power consumption of the i-th power branch line, ai′ represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1′ represents the intercept;
[0117] Calculate the percentage difference between each pair of model coefficients ai and ai′ in formula (1) and formula (2) respectively;
[0118] The power branch lines of the power line are sorted in descending order according to the percentage differences between the calculated pairs of model coefficients;
[0119] Output the sorting results of each power branch line to show the user the sorting of line loss anomalies of each power branch line from high to low.
[0120] In specific implementation, the present invention can immediately issue an abnormal line loss warning to the power line after outputting the sorting results of each power branch line, or can also issue an abnormal line loss warning to the power line while outputting the sorting results of each power branch line.
[0121] It is understandable that the abnormal line loss warning for the power line can send a warning message to a preset mobile phone number. In specific implementation, other warning methods can also be used for warning, such as arranging an alarm to warn the power line of abnormal line loss sound.
[0122] In this embodiment, the percentage difference between the model coefficients ai and ai′ is the percentage difference of the model coefficient ai relative to ai′, and the calculation formula is:
[0123]
[0124] It can be understood that the model coefficients ai and ai′ correspond to the i-th power branch line among the plurality of power branch lines, i=1, 2, …, n.
[0125] Optionally, the power consumption on the power main line and the power consumption on each power branch line collected within a preset time period before the current time and the calculated actual bus loss value are obtained to construct a data set corresponding to the current time, and a new model is constructed by using the data set corresponding to the current time and adopting the same method as that of constructing the multivariate linear fitting model, including:
[0126] Acquire the power consumption on the power bus line and the power consumption on each power branch line collected within a preset time period from the current time and the calculated actual bus loss value to construct a data set corresponding to the current time;
[0127] Determine whether the amount of data in the data set corresponding to the current time is greater than or equal to the number of the power branch lines (ie, n):
[0128] If yes, then using the data set corresponding to the current time, the new model is constructed using the same method as that used to construct the multivariate linear fitting model;
[0129] If not, an abnormal line loss warning will be issued for the power line.
[0130] Optionally, the method for constructing the multivariate linear fitting model includes the following steps S1 to S10.
[0131] Step S1: Construct a training set.
[0132] The training set construction method includes:
[0133] Step S11, collecting historical power consumption on the power main line, and correspondingly collecting historical power consumption on each power branch line under the historical power consumption on the power main line;
[0134] Step S12, using each collected historical power consumption on the power main line and the historical power consumption on each power branch line corresponding to each collected historical power consumption on the power main line, correspondingly calculate the actual historical bus loss value of the power line;
[0135] Step S13: construct the training set based on the collected historical power consumption on the main power line and the historical power consumption on each power branch line and the calculated actual historical main power loss value of each power line.
[0136] Step S2: Establish a linear regression model between the total power loss prediction value Pt of the power line and the total power consumption.
[0137] It can be understood that the present invention collects historical power consumption on each power branch line at the same time as collecting historical power consumption on the power main line.
[0138] The calculation formula of the linear regression model established in step S2 is:
[0139] Pt=a0×P0+b0, (3)
[0140] In formula (3), P0 represents the total power consumption of the power line, a0 is the regression coefficient, and b0 is the intercept; the power consumption on the power bus line is the total power consumption of the power line;
[0141] Step S3: Substitute the actual historical bus loss value of the power line in the training set and the historical power consumption on the corresponding power bus into formula (3), then determine the specific values of a0 and b0 in formula (3) by data fitting method, and substitute the specific values of a0 and b0 back into formula (3) to obtain the constructed linear regression model.
[0142] Step S4: Substitute each historical power consumption on the power bus line in the training set into the constructed linear regression model to calculate the power line historical bus loss prediction value corresponding to each historical power consumption on the power bus line.
[0143] Step S4: respectively calculating the percentage error between the obtained historical total bus loss prediction value of each power line and the actual historical total bus loss value in the training set corresponding to each power line, which is recorded as the first percentage error;
[0144] Calculate the mean μ and standard deviation σ of each first percentage error;
[0145] Based on the mean μ and the standard deviation σ, the percentage difference threshold interval is obtained, and the percentage difference threshold interval is [μ-3σ, μ+3σ];
[0146] The data corresponding to the first percentage error exceeding the percentage difference threshold interval [μ-3σ,μ+3σ] in the training set are deleted to obtain a new data set.
[0147] Step S5: normalize the data in the new data set to obtain a normalized data set.
[0148] Step S6: Perform principal component analysis on the historical power consumption of each power branch line in the normalized data set to obtain a set of selected principal components Fj and a eigenvector dj corresponding to each selected principal component Fj; wherein dj=(d j1 ,d j2 ,…,d jk ,…,d jn ), j = 1, 2, ..., n, the expression of each selected principal component Fj is:
[0149] F1=d 11 P1+d 12 P2+…+d 1k Pk+…+d 1n Pn,
[0150] F2=d 21 P1+d 22 P2+…+d 2k Pk+…+d 2n Pn,
[0151] F3=d 31 P1+d 32 P2+…+d 3k Pk+…+d 3n Pn,
[0152] …
[0153] Fj=d j1 P1+d j2 P2+…+d jk Pk+…+d jn Pn,
[0154] …
[0155] Fm=d m1 P1+d m2 P2+…+d mk Pk+…d mn Pn,
[0156] Wherein, m represents the total number of all selected principal components, Fj represents the jth principal component among all selected principal components, Pk represents the power consumption of the kth power branch line among the plurality of power branch lines, and d jk is the value of the kth element in the eigenvector dj corresponding to the jth principal component, k = 1, 2, ..., n;
[0157] Substitute each set of values of P1, P2, ..., Pn in the normalized data set into the expression of each selected principal component Fj, and calculate each set of values of F1, F2, F3, ..., Fj, ..., Fm;
[0158] In this embodiment, each set of values of (P1, P2, ..., Pn) in the normalized data set corresponds to an actual historical bus loss value in the normalized data set, and corresponds to a set of values of F1, F2, F3, ..., Fj, ..., Fm. It can be understood that each set of values of F1, F2, F3, ..., Fj, ..., Fm corresponds to an actual historical bus loss value in the normalized data set.
[0159] The historical power consumption on each power branch line in the normalized data set is subjected to principal component analysis to obtain a set of selected principal components and the eigenvectors corresponding to each selected principal component, including:
[0160] Calculate the covariance matrix of the data matrix corresponding to the historical power consumption on the n power branch lines in the normalized data set;
[0161] Calculate the eigenvalues of the covariance matrix and the eigenvectors corresponding to each eigenvalue;
[0162] Arrange the eigenvalues from largest to smallest, and arrange the eigenvectors accordingly;
[0163] Calculate the contribution rate of each principal component; the contribution rate of each principal component is equal to the eigenvalue corresponding to the principal component divided by the sum of the eigenvalues corresponding to all principal components;
[0164] According to a preset contribution rate threshold (the contribution rate threshold can be set to 85% or above, for example, it can be set to 90%), the principal component whose cumulative contribution rate reaches the contribution rate threshold is selected as the principal component, and then the eigenvector corresponding to each selected principal component is obtained.
[0165] Step S7: construct a multivariate linear fitting equation for calculating the total line loss prediction value Pt of the power line using all selected principal components. The calculation formula of the equation is as follows:
[0166] Pt=e 1 ×F1+e 2 ×F2+…+e j ×Fj+…+e m ×Fm+b, (4)
[0167] In formula (4), e j represents the coefficient of the term where Fj is located, and b represents the intercept.
[0168] Step S8: Substitute each set of values of the selected principal component and the actual historical total bus loss value in the normalized data set corresponding thereto into formula (4) (i.e., train formula (4) using each set of values of the selected principal component and the actual historical total bus loss value in the normalized data set corresponding thereto), and then determine e in formula (4) by a data fitting method. 1 、e 2 ,…,e j ,…,e m , b specific value, and then e 1 、e 2 ,…,e j ,…,e m Substitute the specific values of , b into formula (4) to obtain the constructed multivariate linear fitting equation.
[0169] Step S9: Substitute the expressions of F1, F2, F3, ..., Fj, ..., Fm into the constructed multivariate linear fitting equation to obtain the following formula:
[0170] Pt=(e 1 d 11 +e 2 d 21 +…+e m d m1 )×P1+(e 1 d 12 +e 2 d 22 +…+e m d m2 )×P2+…+(e 1 d 1j +e 2 d 2j +…+e m d mj )×Pj+…+(e 1 d 1n +e 2 d2n +…+e m d mn )×Pn+b; (5)
[0171] Formula (5) is the same as Formula (1), that is, the multivariate linear fitting model is obtained, wherein the coefficients of each item P1, P2, ..., Pn in Formula (1) are equal to the coefficients of the corresponding items in Formula (5), and the intercept b1 in Formula (1) is equal to the intercept n in Formula (5).
[0172] It can be understood that the coefficients of each term P1, P2, ..., Pn in formula (1) are correspondingly equal to the coefficients of each term P1, P2, ..., Pn in formula (5).
[0173] Specifically:
[0174]
[0175] The present invention determines the specific value of the percentage difference threshold interval ([μ-3σ, μ+3σ]) by constructing a linear regression model, so as to further optimize the data in the training set for creating a multivariate linear fitting model for subsequent line loss monitoring. This can ensure the creation of a better multivariate linear fitting model to a certain extent, so that the residual of the overall model will not be large due to a single linear equation, resulting in low accuracy.
[0176] In actual production, in order to balance the equation, the fitting process sometimes has to set some coefficients to negative values. For the present invention, if some model coefficients are set to negative values, the model loses its practical significance. The present invention uses principal component analysis to convert high-dimensional data into low-dimensional data, retaining important information in the data as much as possible, ensuring that the coefficients of each power consumption are positive values in the process of linear fitting, simplifying the data structure, filtering out some redundant influencing information, and being more conducive to subsequent analysis and research.
[0177] In the training set constructed in the above manner in the present invention, the data in the training set may contain abnormal data and empty data, and actual production is often affected by such data, for example, affecting the accuracy of the trained or constructed model.
[0178] To this end, as a preferred embodiment, before substituting the actual historical bus loss value of the power line in the training set and the historical power consumption on the corresponding power bus line into formula (3), method 100 further includes:
[0179] Clean the training set to remove abnormal and empty data;
[0180] Standardize the cleaned data in the training set.
[0181] Each piece of data in the training set corresponds to a sample, and each sample contains the historical power consumption on the power main line and the historical power consumption on each power branch line collected by simultaneous sampling, as well as the actual historical main line loss value of the power line calculated using the historical power consumption collected by simultaneous sampling.
[0182] The abnormal data are samples containing excessively large or insufficient power consumption, that is, samples containing data exceeding the data collection range of the power consumption collection device.
[0183] The empty data is a sample containing empty data caused by a data transmission problem of the power consumption collection device.
[0184] For example, Figure 2 As shown, this is a simplified schematic diagram of the power circuit of a certain equipment circuit system. Figure 2 The power line in China includes one main power line and four branch power lines. Smart meters are installed on the main power line and the four branch power lines of the power line. The smart meters installed on the main power line are main meters, and the smart meters installed on each branch power line are branch meters.
[0185] The main electricity meter and four branch electricity meters make five electricity meters with no difference at all. Figure 2 In the figure, the historical power consumption on the main power line is represented by P0, and the historical power consumption on the four branch power lines are represented by P1, P2, P3, and P4 respectively.
[0186] During the operation of the power line, when collecting the historical power consumption on the main power line and the historical power consumption on each power branch line, the power consumption collected by the main meter and the four branch meters is recorded simultaneously each time.
[0187] Because there is loss in the power line, there is a difference between the total meter count and the sum of the branch meter counts recorded at the same time. Assuming this part of the loss is Pt0, the total meter count is subtracted from the sum of the branch meter counts recorded at the same time, and the actual historical total line loss value corresponding to the total meter count and the branch meter count recorded at the same time is obtained. Figure 2 The historical power consumption and its corresponding actual historical bus loss value in the power line diagram shown in FIG. 1 are shown in Table 1.
[0188] Table 1 Figure 2 The historical power consumption of the power line shown in the figure and its corresponding actual historical bus loss value
[0189]
[0190]
[0191] In Table 1, the data recorded in the column where P0 is located is the historical power consumption of the power main line, and the columns where P1, P2, P3, and P4 are located each record the historical power consumption of their corresponding power branch lines. The data in Table 1 constitute a training set of this embodiment. Each data in Table 1 is a sample. Pt01 = P01-(P11+P21+P31+P41), and other actual historical bus loss values Pt0 in Table 1 can be calculated by referring to the calculation method of Pt01.
[0192] Perform data cleaning on Table 1. For example, if P11 in the first row of data in Table 1 is abnormal data, delete the entire row of data where P11 is located. For example, if P12 is empty data, delete the entire row of data where P12 is located.
[0193] In this embodiment, standardizing the data is to convert the original data into data with a mean of 0 and a variance of 1, so as to eliminate the influence of different dimensions on the analysis results.
[0194] In this embodiment, the data is standardized, that is, each set of values of P = (P0, P1, P2, P3, P4, Pt0) is standardized, and the standardized formula is:
[0195]
[0196] Where P is the original data before standardization, is the mean of the original data, σP is the standard deviation of the original data, P norm is the standardized data.
[0197] The present invention standardizes the data, which can speed up the model training speed (i.e., fitting speed) and improve the accuracy of the model. In addition, the standardization process can also eliminate the dimensional differences of the data, making the model training more stable and efficient, ensuring that the collected different power consumption data have a consistent scale, and then improving the model construction efficiency and prediction ability.
[0198] Optionally, the data fitting involved in method 100 adopts the least square method.
[0199] The present invention realizes line loss monitoring of circuit lines through the collected power consumption, which helps to avoid obtaining line loss data for line loss monitoring by two methods of measuring line loss, namely, allocating according to power consumption ratio and allocating according to fixed coefficient, and helps to reduce monitoring errors to a certain extent.
[0200] The present invention can help relevant personnel clearly understand the ranking of line losses by comparing the circuit influence coefficients in the model, and then help to sort by line losses. Each time, the line with the highest line loss among the remaining unchecked lines is checked, which then helps to quickly locate the source of the problem, reduce the troubleshooting time, and improve the efficiency of problem solving. At the same time, it helps to reduce to a certain extent the energy consumed by high line loss lines during transmission or distribution, causing energy waste, and then improve energy utilization, reduce the possibility of safety accidents in the line, and improve the safe and stable operation of the entire power grid.
[0201] like Figure 3 As shown, the present invention provides a line loss monitoring system 200 , and the system 200 includes a first module 201 , a second module 202 and a third module 203 .
[0202] In this embodiment, the system 200 is applied to power line line loss monitoring. The power line includes a main power line and several branch power lines for connecting user electrical equipment. The input end of each branch power line is connected to the output end of the main power line.
[0203] The first module 201 is used to collect the current power consumption of the main power line and the current power consumption of each power branch line in real time, and calculate the current actual bus loss value of the power line based on the collected current power consumption of the main power line and the current power consumption of each power branch line.
[0204] The second module 202 is used to substitute the collected current power consumption of each power branch line into a pre-built multivariate linear fitting model for calculation to obtain a total power loss prediction value of the power line.
[0205] The calculation formula of the multivariate linear fitting model is:
[0206] Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1)
[0207] In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept.
[0208] The third module 203 is used to calculate the percentage difference between the total line loss prediction value and the actual total line loss value; if the percentage difference exceeds a preset percentage difference threshold interval, an abnormal line loss warning is issued for the power line.
[0209] It can be understood that the percentage difference exceeds the preset percentage difference threshold interval, that is, the percentage difference is not within the preset percentage difference threshold interval.
[0210] like Figure 4 As shown, the present invention also provides a line loss monitoring platform 500, which is applied to power line line loss monitoring. The power line includes a main power line and a plurality of branch power lines for connecting user electrical equipment. The input end of each branch power line is connected to the output end of the main power line.
[0211] In this embodiment, smart meters are installed on the main power line and each power branch line. The smart meter installed on the main power line is the main meter, and the smart meter installed on each power branch line is the branch meter.
[0212] In this embodiment, the platform 500 includes a power collection module 501 and a first calculation module 502 .
[0213] The power collection module 501 is connected to the main power meter and each branch power meter, and is used to collect the current power consumption of the main power line and the current power consumption of each branch power line in real time.
[0214] In specific implementation, the power collection module 501 can be connected to the main electric meter and each branch electric meter by wireless connection. The wireless connection can be achieved by any method such as GPRS, 4G, 5G, etc.
[0215] The first calculation module 502 is connected to the power collection module 501 and is used to calculate the current actual bus loss value of the power line based on the current power consumption of the main power line and the current power consumption of each power branch line collected by the power collection module 501.
[0216] The platform 500 further includes a second calculation module 503 , a third calculation module 504 and a decision module 505 .
[0217] The second calculation module 503 is connected to the power collection module 501 and is used to substitute the current power consumption of each power branch line collected by the power collection module 501 into a pre-built multivariate linear fitting model for calculation to obtain a total power loss prediction value of the power line.
[0218] The calculation formula of the multivariate linear fitting model is:
[0219] Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1)
[0220] In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept.
[0221] The third calculation module 504 is connected to the second calculation module 503 and the first calculation module 502, and is used to calculate the percentage difference between the bus loss prediction value and the actual bus loss value.
[0222] The decision module 505 is connected to the third calculation module 504, and is used to issue an abnormal line loss warning to the power line when the percentage difference calculated by the third calculation module 504 exceeds a preset percentage difference threshold range.
[0223] Alternatively, if Figure 5 As shown, the platform 500 further includes a display module 506 and a storage module 507 .
[0224] The power collection module 501 is connected to the storage module 507 and is used to store the collected power consumption on the power main line and the power consumption of each power branch line into the storage module 507 in real time.
[0225] The first calculation module 502 is connected to the storage module 507 and is used to store the actual bus loss value of the power line calculated based on the power consumption of the main power line and the power consumption of each power branch line collected by the power collection module 501 into the storage module 507 .
[0226] The decision module 505 is connected to the storage module 507, and is used to perform the following steps when the percentage difference calculated by the third calculation module 504 exceeds a preset percentage difference threshold interval:
[0227] The power consumption on the power bus line and the power consumption on each power branch line collected within a preset time period before the current time and the calculated actual bus loss value stored in the storage module 507 are obtained to construct a data set corresponding to the current time, and a new model is constructed using the data set corresponding to the current time by adopting the same construction method as the multivariate linear fitting model. The calculation formula of the new model is:
[0228] Pt′=a1′×P1+a2′×P2+…+ai′×Pi+…+an′×Pn+b1′, (2)
[0229] In formula (2), Pt′ represents the total power loss prediction value of the power line, i represents the i-th power branch line among the plurality of power branch lines, Pi represents the power consumption of the i-th power branch line, ai′ represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1′ represents the intercept;
[0230] Calculate the percentage difference between each pair of model coefficients ai and ai′ in formula (1) and formula (2) respectively;
[0231] The power branch lines of the power line are sorted in descending order according to the percentage differences between the calculated pairs of model coefficients;
[0232] Output the ranking results of each power branch line to show the user the ranking of abnormal line loss of each power branch line from high to low;
[0233] The display module 506 is connected to the decision module 505 and is used to display the ranking results to the user.
[0234] Figure 6 The present invention provides a schematic diagram of the structure of a terminal 300 provided in an embodiment of the present invention. The terminal 300 can be used to execute the method 100 provided in an embodiment of the present invention.
[0235] The terminal 300 may include: a processor 310, a memory 320 and a communication unit 330. These components communicate via one or more buses. Those skilled in the art will appreciate that the server structure shown in the figure does not limit the present invention, and it may be a bus structure or a star structure, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.
[0236] The memory 320 can be used to store the execution instructions of the processor 310, and the memory 320 can be implemented by any type of volatile or non-volatile storage terminal or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. When the execution instructions in the memory 320 are executed by the processor 310, the terminal 300 can perform some or all of the steps in the following method embodiments.
[0237] The processor 310 is the control center of the storage terminal, and uses various interfaces and lines to connect various parts of the entire electronic terminal. It runs or executes software programs and / or modules stored in the memory 320, and calls data stored in the memory to perform various functions of the electronic terminal and / or process data. The processor can be composed of an integrated circuit (IC), for example, it can be composed of a single packaged IC, or it can be composed of a plurality of packaged ICs with the same or different functions. For example, the processor 310 can include only a central processing unit (CPU). In the embodiment of the present invention, the CPU can be a single computing core or multiple computing cores.
[0238] The communication unit 330 is used to establish a communication channel so that the storage terminal can communicate with other terminals, receive user data sent by other terminals or send user data to other terminals.
[0239] The present invention also provides a computer storage medium 400, such as Figure 7 As shown, the computer storage medium 400 stores a program 410, which may include some or all of the steps in each embodiment of the present invention when executed. The storage medium may be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM).
[0240] The same and similar parts between the various embodiments in this specification can be referenced to each other.
[0241] Although the present invention has been described in detail by referring to the accompanying drawings and in combination with the preferred embodiments, the present invention is not limited thereto. Without departing from the spirit and essence of the present invention, a person of ordinary skill in the art may make various equivalent modifications or substitutions to the embodiments of the present invention, and these modifications or substitutions shall be within the scope of the present invention. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, and all of these shall be within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.
Claims
1. A line loss monitoring method is applied to line loss monitoring of power lines, wherein the power lines include a main power line and a plurality of branch power lines for connecting user electrical equipment, and the input end of each branch power line is connected to the output end of the main power line; characterized in that: Methods include: Collect the current power consumption of the main power line and the current power consumption of each power branch line in real time, and calculate the current actual bus loss value of the power line based on the collected current power consumption of the main power line and the current power consumption of each power branch line; Substitute the collected current power consumption of each power branch line into the pre-built multivariate linear fitting model for calculation to obtain the total power loss prediction value of the power line; the calculation formula of the multivariate linear fitting model is: Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1) In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept; Calculating a percentage difference between the total bus loss prediction value and the actual total bus loss value; If the percentage difference exceeds a preset percentage difference threshold range, an abnormal line loss warning is issued for the power line.
2. The method according to claim 1, characterized in that When the percentage difference exceeds a preset percentage difference threshold interval, the method further includes: The power consumption on the power main line and the power consumption on each power branch line collected within a preset time period before the current time and the calculated actual bus loss value are obtained to construct a data set corresponding to the current time, and a new model is constructed using the data set corresponding to the current time and the same method as that of constructing the multivariate linear fitting model. The calculation formula of the new model is: Pt′=a1′×P1+a2′×P2+…+ai′×Pi+…+an′×Pn+b1′, (2) In formula (2), Pt′ represents the total power loss prediction value of the power line, i represents the i-th power branch line among the plurality of power branch lines, Pi represents the power consumption of the i-th power branch line, ai′ represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1′ represents the intercept; Calculate the percentage difference between each pair of model coefficients ai and ai′ in formula (1) and formula (2) respectively; The power branch lines of the power line are sorted in descending order according to the percentage differences between the calculated pairs of model coefficients; Output the sorting results of each power branch line to show the user the sorting of line loss anomalies of each power branch line from high to low.
3. The method according to claim 1, characterized in that: The method for constructing the multivariate linear fitting model comprises: Constructing a training set: collecting historical power consumption on the main power line, and correspondingly collecting historical power consumption on each power branch line under the historical power consumption on the main power line; using each collected historical power consumption on the main power line and the historical power consumption on each power branch line under each collected historical power consumption on the main power line, correspondingly calculating the actual historical main power loss value of the power line; constructing the training set based on the collected historical power consumption on the main power line, the historical power consumption on each power branch line, and each actual historical main power loss value of the power line calculated; A linear regression model is established between the total power line loss prediction value Pt and the total power consumption. The calculation formula of the model is: Pt=a0×P0+b0, (3) In formula (3), P0 represents the total power consumption of the power line, a0 is the regression coefficient, and b0 is the intercept; the power consumption on the power bus line is the total power consumption of the power line; Substitute the actual historical bus loss value of the power line in the training set and the historical power consumption on the corresponding power bus into formula (3), and then determine the specific values of a0 and b0 in formula (3) by data fitting method, and substitute the specific values of a0 and b0 back into formula (3) to obtain the constructed linear regression model; Substitute each historical power consumption on the power bus line in the training set into the constructed linear regression model, and calculate the predicted value of the historical power line loss corresponding to each historical power consumption on the power bus line; The percentage errors between the predicted historical total bus loss values of each power line and the actual historical total bus loss values in the training sets corresponding to each power line are calculated respectively, and recorded as the first percentage error; Calculate the mean μ and standard deviation σ of each first percentage error; Based on the mean μ and the standard deviation σ, the percentage difference threshold interval is obtained, and the percentage difference threshold interval is [μ-3σ, μ+3σ]; The data corresponding to the first percentage error exceeding the percentage difference threshold interval [μ-3σ,μ+3σ] in the training set are deleted to obtain a new data set; Normalize the data in the new data set to obtain a normalized data set; The historical power consumption on each power branch line in the normalized data set is analyzed by principal component analysis to obtain a set of selected principal components Fj and the eigenvector dj corresponding to each selected principal component Fj; where dj = (d j1 ,d j2 ,…,d jk ,…,d jn ), j = 1, 2, ..., n, the expression of each selected principal component Fj is: F1=d 11 P1+d 12 P2+…+d 1k Pk+…+d 1n Pn, F2=d 21 P1+d 22 P2+…+d 2k Pk+…+d 2n Pn, F3=d 31 P1+d 32 P2+…+d 3k Pk+…+d 3n Pn, … Fj=d j1 P1+d j2 P2+…+d jk Pk+…+d jn Pn, … Fm=d m1 P1+d m2 P2+…+d mk Pk+…d mn Pn, Wherein, m represents the total number of all selected principal components, Fj represents the jth principal component among all selected principal components, Pk represents the power consumption of the kth power branch line among the plurality of power branch lines, and d jk is the value of the kth element in the eigenvector dj corresponding to the jth principal component, k = 1, 2, ..., n; Substitute each set of values of P1, P2, ..., Pn in the normalized data set into the expression of each selected principal component Fj, and calculate each set of values of F1, F2, F3, ..., Fj, ..., Fm; All selected principal components are used to construct a multivariate linear fitting equation for calculating the total total power loss prediction value Pt of the power line. The calculation formula of the equation is as follows: Pt=e1×F1+e2×F2+…+e j ×Fj+…+e m ×Fm+b, (4) In formula (4), e j represents the coefficient of the term where Fj is located, and b represents the intercept; Substitute each set of values of each selected principal component and its corresponding actual historical total bus loss value in the normalized data set into formula (4), and then determine e1, e2, ..., e in formula (4) by data fitting method. j ,…,e m , b, and then e1, e2, …, e j ,…,e m Substitute the specific values of , b into formula (4) to obtain the constructed multivariate linear fitting equation; Substituting the expressions of F1, F2, F3, ..., Fj, ..., Fm into the constructed multivariate linear fitting equation, the following formula is obtained: Pt=(e1d 11 +e2d 21 +…+e m d m1 )×P1+(e1d 12 +e2d 22 +…+e m d m2 )×P2+…+(e1d 1j +e2d 2j +…+e m d mj )×Pj+…+(e1d 1n +e2d 2n +…+e m d mn )×Pn+b; (5) Formula (5) is the same as Formula (1), that is, the multivariate linear fitting model is obtained, wherein the coefficients of each item P1, P2, ..., Pn in Formula (1) are equal to the coefficients of the corresponding items in Formula (5), and the intercept b1 in Formula (1) is equal to the intercept b in Formula (5).
4. The method according to claim 3, characterized in that Before substituting the actual historical bus loss value of the power line in the training set and the historical power consumption on the corresponding power bus line into formula (3), the method further includes: Clean the training set to remove abnormal and empty data; Standardize the cleaned data in the training set.
5. The method according to claim 3, characterized in that: The least squares method was used for data fitting in the method.
6. A line loss monitoring system is applied to line loss monitoring of power lines, wherein the power lines include a main power line and a plurality of branch power lines for connecting user electrical equipment, and the input end of each branch power line is connected to the output end of the main power line; characterized in that: The system includes: The first module is used to collect the current power consumption of the power main line and the current power consumption of each power branch line in real time, and calculate the current actual bus loss value of the power line based on the collected current power consumption of the power main line and the current power consumption of each power branch line; The second module is used to substitute the collected current power consumption of each power branch line into a pre-built multivariate linear fitting model for calculation to obtain a total power loss prediction value of the power line; the calculation formula of the multivariate linear fitting model is: Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1) In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept; The third module is used to calculate the percentage difference between the total line loss prediction value and the actual total line loss value; if the percentage difference exceeds a preset percentage difference threshold range, an abnormal line loss warning is issued for the power line.
7. A line loss monitoring platform, applied to line loss monitoring of power lines, wherein the power lines include a main power line and a plurality of branch power lines for connecting user electrical equipment, wherein the input end of each branch power line is connected to the output end of the main power line; characterized in that: Smart meters are installed on the main power line and each power branch line. The smart meter installed on the main power line is the main meter, and the smart meter installed on each power branch line is the branch meter; The platform includes a power collection module and a first calculation module: The power collection module is connected to the main power meter and each branch power meter to collect the current power consumption on the power main line and the current power consumption of each power branch line in real time; The first calculation module is connected to the power collection module and is used to calculate the current actual bus loss value of the power line based on the current power consumption of the main power line and the current power consumption of each power branch line collected by the power collection module; The platform also includes a second computing module, a third computing module and a decision module; The second calculation module is connected to the power collection module, and is used to substitute the current power consumption of each power branch line collected by the power collection module into a pre-built multivariate linear fitting model for calculation to obtain a total power loss prediction value of the power line; the calculation formula of the multivariate linear fitting model is: Pt=a1×P1+a2×P2+…+ai×Pi+…+an×Pn+b1, (1) In formula (1), Pt is the total power loss prediction value of the power line, i is the ith power branch line among the plurality of power branch lines, Pi is the power consumption of the ith power branch line, ai represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1 represents the intercept; The third calculation module is connected to the second calculation module and the first calculation module, and is used to calculate the percentage difference between the bus loss prediction value and the actual bus loss value; The decision module is connected to the third calculation module, and is used to issue an abnormal line loss warning to the power line when the percentage difference calculated by the third calculation module exceeds a preset percentage difference threshold range.
8. The platform according to claim 7, characterized in that The platform also includes a display module and a storage module; The power collection module is connected to the storage module and is used to store the collected power consumption on the power main line and the power consumption of each power branch line into the storage module in real time; The first calculation module is connected to the storage module and is used to store the actual bus loss value of the power line calculated based on the power consumption of the power main line and the power consumption of each power branch line collected by the power collection module into the storage module; The decision module is connected to the storage module, and is used to perform the following steps when the percentage difference calculated by the third calculation module exceeds a preset percentage difference threshold interval: The power consumption on the power bus line and the power consumption on each power branch line collected within a preset time period before the current time and the calculated actual bus loss value stored in the storage module are obtained to construct a data set corresponding to the current time, and the data set corresponding to the current time is used to construct a new model using the same construction method as the multivariate linear fitting model. The calculation formula of the new model is: Pt′=a1′×P1+a2′×P2+…+ai′×Pi+…+an′×Pn+b1′, (2) In formula (2), Pt′ represents the total power loss prediction value of the power line, i represents the i-th power branch line among the plurality of power branch lines, Pi represents the power consumption of the i-th power branch line, ai′ represents the model coefficient, i=1,2,…,n, n represents the number of power branch lines contained in the power line, and b1′ represents the intercept; Calculate the percentage difference between each pair of model coefficients ai and ai′ in formula (1) and formula (2) respectively; The power branch lines of the power line are sorted in descending order according to the percentage differences between the calculated pairs of model coefficients; Output the ranking results of each power branch line to show the user the ranking of abnormal line loss of each power branch line from high to low; The display module is connected to the decision module and is used to display the ranking results to the user.
9. A terminal, characterized in that: include: processor; A memory for storing execution instructions of the processor; The processor is configured to execute the method according to any one of claims 1 to 5.
10. A computer-readable storage medium storing a computer program, characterized in that: When the program is executed by a processor, the method described in any one of claims 1 to 5 is implemented.
Citation Information
Cited By
Low-voltage transformer area line loss abnormity diagnosis method, equipment, medium and product
CN121164791A
A method, equipment, medium, and product for diagnosing abnormal line losses in low-voltage distribution areas.
CN121164791B
