Transformer area user-transformer relation identification method based on voltage curve similarity and power-off and power-on events
Through the identification method based on voltage curve similarity and power-off events, combined with multiple algorithms, the household change relationship in the Taiwan area is automatically identified, which solves the problem of inaccurate identification of household change relationships in the Taiwan area distribution network, and improves the recognition accuracy and Reliability improves the operating safety and stability of the power grid.
Patent Information
- Application Number
- CN202411123769.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-15
- Publication Date
- 2024-11-29
AI Technical Summary
In the Taiwan distribution network, due to the complexity of equipment and lines and the lack of automation methods, the identification of household change relationships is inaccurate, resulting in inconsistent topology structure and main site display, frequent misjudgment and misoperation, affecting fault location and emergency repair efficiency .
The identification method based on voltage curve similarity and power-off events is adopted, through data cleaning, comprehensive evaluation of voltage similarity, household change relationship determination and power-off time comparison, combined with cosine distance, Pearson correlation coefficient and dynamic time regularization algorithm, Automatically identify the household change relationship in the station area to improve the recognition accuracy.
It realizes accurate identification of household change relationships in Taiwan, improves identification accuracy and reliability, reduces misjudgment and misoperation, improves the safety and stability of power grid operation, and provides support for troubleshooting and emergency response.
Smart Images

Figure CN119046697A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of distribution networks in substations, and specifically to a method for identifying the relationship between household transformers in substations based on voltage curve similarity and power outage and power-on events. Background Art
[0002] The distribution network in the substation area undertakes the key function of transmitting household electricity at the end of the power grid and directly serves users. Its reliable operation is an important part of the reliability chain of the entire power grid operation. However, with the continuous development and deepening of smart grids in the field of distribution networks, the scale of grid loads and equipment is getting larger and larger, resulting in numerous and complex equipment and lines in the substation area. Accurately identifying the relationship between households and transformers in the substation area is particularly important for the construction of smart grids. On the basis of accurately identifying the relationship between households and transformers in the substation area, lean management functions such as accurate calculation of line loss in the substation area, three-phase imbalance management, and anti-electricity theft can be carried out to adapt to the access of high-density renewable energy and promote the intelligent construction of distribution networks. However, due to different construction times and the lack of corresponding automation and information means, when there are problems such as changes in user information, replacement of meter failures, and upgrading and transformation of substation areas, manual input often results in input errors or untimely update of files, resulting in inconsistencies between the actual distribution network topology on site and the display of the master station, and the household-to-substation relationship is relatively chaotic. At the same time, with the continuous improvement of users' requirements for power quality and power supply service levels, distribution network fault location and emergency repair work are being carried out, and the relationship between households and transformers in the distribution network substation area has become an objective need. As basic data, the distribution network topology structure displays the topological relationship of the entire substation in real time, which helps operation and maintenance personnel to understand the operation status of the substation in time, realize fast and accurate fault location, and improve emergency repair efficiency. Summary of the invention
[0003] The purpose of the embodiments of the present application is to provide a method for identifying the relationship between household transformers in a substation area based on voltage curve similarity and power outage and power-on events, so as to improve the accuracy of identifying the relationship between household transformers in a substation area and prevent misjudgment and misoperation caused by topology errors by operation and maintenance personnel.
[0004] To achieve the above objectives, this application provides the following technical solutions:
[0005] The embodiment of the present application provides a method for identifying the relationship between transformers in a transformer substation area based on voltage curve similarity and power outage events, including the following specific steps:
[0006] Data acquisition and cleaning: obtain the information of the substation and electric energy meter archives from the marketing system, obtain the voltage, phase, and power-off data of the substation master meter and user meter from the electricity consumption information collection system, and establish screening rules for different types of scenarios to perform data cleaning;
[0007] Comprehensive evaluation of voltage similarity: the cosine distance is used to measure the similarity between two voltage vectors; the Pearson correlation coefficient is used to determine whether the change trends of two voltages are consistent; and the dynamic time warping (DWT) is used to obtain an optimal curved path by adjusting the relationship between the corresponding elements at different time points in the time series, so that it can measure the similarity between time series well.
[0008] To determine the household change relationship, based on the pilot data, we set the optimal K values corresponding to the cosine distance algorithm, Pearson correlation coefficient algorithm, and DWT distance algorithm. According to the offline verification results of the household change relationship, we continuously optimize the K value so that the household change relationship is determined when two of the three algorithms are most relevant.
[0009] The power-off and power-on times are compared, and the difference set users are clustered according to the comprehensive evaluation results of voltage similarity. The power outage information of branch meters and end users is collected by shutting down the branches of the substation in sequence to complete the analysis of the affiliation relationship between users and substations.
[0010] The data cleaning specifically includes data missing, data defect, no voltage fluctuation, large voltage fluctuation, low voltage, and high voltage.
[0011] The DWT distance algorithm reflects the fluctuation trend between voltage sequences and has a high sensitivity to the fluctuation trend between different voltage sequences. The DWT distance is used to measure the similarity of voltage sequences. If the DWT distance between two voltage sequences is smaller, the voltage sequences of the two users are more similar, the two users are closer, and are in the same substation.
[0012] The power-off time comparison specifically includes:
[0013] Orderly power outage and restoration: when an active power outage is required in a low-voltage area, each user's meter will generate a power outage event with a timestamp and save it in the meter. When the area is restored, the restoration of the main branches of the area will be carried out in an orderly manner. When the main branch is restored, the user meters under the branch will generate restoration information with a timestamp and save it in the user's meter.
[0014] Read the user's meter information. The smart gateway of the substation area is connected to the concentrator. When all areas are fully powered on, the smart gateway reads the latest power outage and restoration information of the user's meter in the substation area through the concentrator and records it in the smart gateway database.
[0015] For household transformer analysis, the intelligent gateway of the substation classifies and stores all the user meters according to the start and end time of power outage and restoration of each user's meter, and the user meters with similar power outage and restoration time are classified and stored to form household transformer relationships.
[0016] Compared with the prior art, the beneficial effects of the present invention are: effective integration of voltage, event and other data, combined with the voltage similarity comprehensive analysis algorithm and power outage and power-on time comparison, to achieve automatic identification of substation and household transformer relationships, which is helpful to promote the intelligent transformation and upgrading of the power industry, and use multi-source data for mutual verification to improve the accuracy and reliability of identification, so as to timely update the topological relationship between substations and household transformers, provide support for troubleshooting and responding to emergencies, and improve the safety and stability of power grid operation. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without paying creative work.
[0018] Figure 1 A flow chart of the method of this application;
[0019] Figure 2 This is a schematic diagram of the cosine similarity algorithm of this application;
[0020] Figure 3 This is the dynamic time warping path diagram of this application. DETAILED DESCRIPTION
[0021] The technical solutions in the embodiments of the present application will be described below in conjunction with the drawings in the embodiments of the present application. It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings.
[0022] The terms "comprises," "comprising," or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or apparatus that includes a list of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes the element.
[0023] The terms "first", "second", etc. are only used to distinguish one entity or operation from another entity or operation, and should not be understood as indicating or implying relative importance, nor should they be understood as requiring or implying any such actual relationship or order between these entities or operations.
[0024] like Figure 1As shown, the present application proposes a method for identifying the relationship between transformers in the transformer substation area based on voltage curve similarity and power outage events, including the following specific steps:
[0025] Data acquisition and cleaning: obtain the information of the substation and electric energy meter archives from the marketing system, obtain the voltage, phase, and power-off data of the substation master meter and user meter from the electricity consumption information collection system, and establish screening rules for different types of scenarios to perform data cleaning;
[0026] Comprehensive evaluation of voltage similarity: the cosine distance is used to measure the similarity between two voltage vectors; the Pearson correlation coefficient is used to determine whether the change trends of two voltages are consistent; and the dynamic time warping (DWT) is used to obtain an optimal curved path by adjusting the relationship between the corresponding elements at different time points in the time series, so that it can measure the similarity between time series well.
[0027] To determine the household change relationship, based on the pilot data, we set the optimal K values corresponding to the cosine distance algorithm, Pearson correlation coefficient algorithm, and DWT distance algorithm. According to the offline verification results of the household change relationship, we continuously optimize the K value so that the household change relationship is determined when two of the three algorithms are most relevant.
[0028] The power-off and power-on times are compared, and the difference set users are clustered according to the comprehensive evaluation results of voltage similarity. The power outage information of branch meters and end users is collected by shutting down the branches of the substation in sequence to complete the analysis of the affiliation relationship between users and substations.
[0029] The specific steps are as follows:
[0030] Data acquisition and cleaning
[0031] Obtain the archive information of the substation and electric energy meter from the marketing system, obtain the voltage, phase, and power-off data of the substation master meter and user meter from the electricity consumption information collection system, and establish screening rules for different types of scenarios to ensure the reliability and effectiveness of the data, thereby improving the performance and prediction accuracy of the model. The following are some examples of cleaning data:
[0032] Data missing: If the user's voltage curve is missing on a certain day or a phase is missing, the voltage information of that phase on that day will no longer be analyzed.
[0033] Data shortcomings: If the missing voltage curve (24 points) of a user on a certain day exceeds k%, it is recommended that k = 16, and no analysis will be performed.
[0034] No voltage fluctuation: The voltage value is 0, and the voltage curve remains at the same value without any fluctuation, so no analysis is performed.
[0035] Large voltage fluctuations: The recommended voltage fluctuation range for single-phase 220V residential power supply is -10% to +7%, that is, the highest voltage is not higher than 236V, and the lowest voltage is not lower than 198V. If the peak-to-valley difference of the voltage curve within a day is greater than kV, k=38 is recommended, and no analysis is performed.
[0036] Low voltage: The number of points in the voltage curve within one day where the voltage is lower than kV, k = k1*R, k1 is the average voltage of the total meter, and R is (1-10%). If k exceeds N points in a row, it is recommended that N = 3, then no analysis will be performed.
[0037] High voltage: The number of points in the voltage curve within one day where the voltage is lower than kV, k = k1*R, k1 is the average voltage of the total meter, and R is (1+7%). If k exceeds N points in a row, it is recommended that N = 3, and no analysis will be performed.
[0038] Voltage similarity comprehensive evaluation
[0039] 2.1 Algorithm Introduction
[0040] Cosine distance
[0041] Cosine similarity s evaluates the similarity between vectors by calculating the cosine value of the angle between them, such as Figure 1 shown.
[0042]
[0043] In the formula,
[0044] The cosine similarity s has nothing to do with the size of the vector, but only with the angle between the vectors. ab for:
[0045] d ab =1-s ab (2)
[0046] A cosine distance d value of 0 indicates that the two vectors are in the same direction, a d value of 1 indicates that the two vectors are orthogonal, and a d value of 2 indicates that the two vectors are in opposite directions.
[0047] Pearson correlation coefficient
[0048] The Pearson correlation coefficient is also called the Pearson product-distance correlation coefficient. It takes the deviation between two sets of data and their respective average values as the base, and multiplies the deviations to measure the linear correlation between the two variables. Its definition is as follows:
[0049]
[0050] Where: is the mean of the sequence X, is the mean of sequence Y, P is the Pearson correlation coefficient, and the value of P is between [-1, 1]. The larger the absolute value of P, the higher the correlation between sequences X and Y; the smaller the absolute value of P, the lower the correlation between sequences X and Y.
[0051] Dynamic Time Warping (DWT)
[0052] The DWT algorithm can reflect the fluctuation trend between voltage sequences and has a high sensitivity to the fluctuation trend between different voltage sequences. The following algorithm uses the DWT distance to measure the similarity of voltage sequences. If the DWT distance between two voltage sequences is smaller, the voltage sequences of the two users are more similar, the two users are closer, and they are in the same substation.
[0053] The DTW algorithm uses the idea of dynamic programming to adjust the correspondence between the elements of the two voltage sequences at different time points, and finds an optimal curved path so that the distance between the voltage sequences of the two smart meters along the path is minimized.
[0054] Suppose there are two voltage sequences R{r1, r2, ..., r m} and T{t1, t2, ..., t n}, m and n represent the number of data in the two voltage sequences respectively. The dynamic time bending paths of the voltage sequences R and T are shown in Figure 2 As shown in the figure, the path composed of black diagonal squares represents the corresponding relationship between the voltage sequences R and T of the smart meter.
[0055] The path can be represented as w = {w1, w2, ..., w s, ... k}, where k represents the total number of elements in the path (the black upper diagonal squares in the figure), and element w s is the coordinate of the sth point on the path, that is, w s =(i, j), which represents the r in the smart meter voltage sequence R i and the voltage sequence T of the smart meter j Correspondingly, r i and t j The distance between them is d(w s )=d(i,j)=|r i -t j |.
[0056] There are multiple paths w, and all possibilities of w form a path space W. There is an optimal path in W that makes Minimum. The DTW distance between voltage sequences R and T is From (r1, t1) to (r m , t n )'s shortest path needs to satisfy: when the measurement reaches the first (r i, t j ) is the shortest path, we need to find the (r i , t j-1 )、(r i-1 , t j-1 )、(r i-1 , t j )3 o'clock to (r j , t j ). To solve Using dynamic programming and the distance d between every two points in the two sequences, the cumulative distance matrix D is defined, that is, Where: i = 1, 2, ..., m; j = 1, 2, ..., n; D (0, 0) = 0;
[0057] D(i,0)=D(0,j)=+∞.
[0058] The last element D(m, n) of the cumulative distance matrix D is the DTW distance calculated according to the fluctuation trend of the sequence.
[0059] D(m,n)=DWT(R,T) (6)
[0060] 2.2 Comprehensive evaluation of voltage similarity based on the above three algorithms
[0061] The cosine distance is used to measure the similarity between two voltage vectors, the Pearson correlation coefficient determines whether the change trends of two voltages are consistent, and the dynamic time warping DWT obtains an optimal curved path by adjusting the relationship between the corresponding elements at different time points in the time series, so that it can measure the similarity between time series well. Based on the pilot data, the optimal K values corresponding to the cosine distance, Pearson correlation coefficient, and DWT distance are set respectively. According to the offline verification results of the household change relationship, the K value is continuously optimized so that the household change relationship is determined when two of the three algorithms are most relevant.
[0062] 2.3 Case Analysis
[0063] A case analysis is conducted using a distribution network community as the data source. The community has two substations. According to the transformer location information, two adjacent substations are selected, 15 users in the two substations are investigated, and the daily average voltage data from December 1 to 10, 2024 are selected as samples. The cosine distance, Pearson correlation coefficient, and DTW calculation results between the user voltage and the distribution transformer voltage of the two substations are calculated respectively, as shown in Table 1.
[0064]
[0065]
[0066] It can be seen from Table 1 that: 1) 7 users, including users 1, 4, 6, 8, 10, 14, and 15, have smaller Euclidean distance and DDTW of the voltage sequence with distribution transformer 1, and the absolute value of the Pearson correlation coefficient is larger. These 5 users belong to the No. 1 station area; 2) 8 users, including users 2, 3, 5, 7, 9, 11, 12, and 13, have smaller Euclidean distance and DDTW of the voltage sequence with distribution transformer 2, and the absolute value of the Pearson correlation coefficient is larger. These 8 users belong to the No. 2 station area. After manual verification, the recognition results using the above scheme are accurate.
[0067] (1) Comparison of power-off and power-on time
[0068] Based on the comprehensive evaluation results of voltage similarity, the difference set users are clustered. By shutting down the branches of the substation in turn, the power outage information of branch meters and end users is collected to complete the analysis of the affiliation relationship between users and substations.
[0069] ① Orderly power outage and restoration
[0070] When an active power outage is required in the low-voltage area, such as transformer maintenance or circuit breaker maintenance, each user's meter generates a power outage event with a timestamp and saves it in the meter. After the maintenance or repair is completed and the area is powered on, the power restoration of the large branches in the area can be carried out in an orderly manner, such as closing the molded case circuit breaker of the trunk branch 1 first, and then closing the molded case circuit breaker of the trunk branch 2 after 30 seconds, and so on. The power restoration of each trunk branch is completed in an orderly manner at a certain time. When the trunk branch is powered on, the user meter under the branch will generate power restoration information with a timestamp and save it in the user meter.
[0071] ②Read the user's meter information
[0072] The smart gateway of the substation is connected to the concentrator. When all districts are fully powered on, the smart gateway reads the latest power outage and restoration information of the user's meter in the substation through the concentrator and records it in the smart gateway database. Since the timeliness requirement of topology identification is not high, the reading behavior can be staggered during the peak period of meter reading.
[0073] ③Analysis of household changes
[0074] The intelligent gateway of the substation classifies and stores all user meters according to the start and end time of power outage and restoration of each user's meter (in the low-voltage substation based on carrier communication, user meters are synchronized by the concentrator, and the time consistency of the meters is good). User meters with similar power outage and restoration times are assigned to the same trunk branch, forming a "user-substation" relationship.
[0075] This application effectively integrates voltage, event and other data, combines the voltage similarity comprehensive analysis algorithm and power outage and power-on time comparison, and realizes automatic identification of the relationship between substations and households, which helps to promote the intelligent transformation and upgrading of the power industry.
[0076] This application uses multi-source data for mutual verification to improve the accuracy and reliability of identification, so as to timely update the topological relationship between substations and household transformers, provide support for troubleshooting and responding to emergencies, and improve the safety and stability of power grid operation.
[0077] The above description is only an embodiment of the present application and is not intended to limit the protection scope of the present application. For those skilled in the art, the present application may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.
Claims
1. A method for identifying the relationship between transformers in a substation area based on voltage curve similarity and power outage events, characterized in that: The specific steps include: Data acquisition and cleaning: obtain the information of the substation and electric energy meter archives from the marketing system, obtain the voltage, phase, and power-off data of the substation master meter and user meter from the electricity consumption information collection system, and establish screening rules for different types of scenarios to perform data cleaning; Comprehensive evaluation of voltage similarity: the cosine distance is used to measure the similarity between two voltage vectors; the Pearson correlation coefficient is used to determine whether the change trends of two voltages are consistent; and the dynamic time warping (DWT) is used to obtain an optimal curved path by adjusting the relationship between the corresponding elements at different time points in the time series, so that it can measure the similarity between time series well. To determine the household change relationship, based on the pilot data, we set the optimal K values corresponding to the cosine distance algorithm, Pearson correlation coefficient algorithm, and DWT distance algorithm. According to the offline verification results of the household change relationship, we continuously optimize the K value so that the household change relationship is determined when two of the three algorithms are most relevant. The power-off and power-on times are compared, and the difference set users are clustered according to the comprehensive evaluation results of voltage similarity. The power outage information of branch meters and end users is collected by shutting down the branches of the substation in sequence to complete the analysis of the affiliation relationship between users and substations.
2. According to claim 1, a method for identifying the relationship between transformers in substations and households based on voltage curve similarity and power outage events, characterized in that: The data cleaning specifically includes data missing, data defect, no voltage fluctuation, large voltage fluctuation, low voltage, and high voltage.
3. The method for identifying the relationship between transformers in substations and households based on voltage curve similarity and power outage events according to claim 1 is characterized in that: The DWT distance algorithm reflects the fluctuation trend between voltage sequences and has a high sensitivity to the fluctuation trend between different voltage sequences. The DWT distance is used to measure the similarity of voltage sequences. If the DWT distance between two voltage sequences is smaller, the voltage sequences of the two users are more similar, the two users are closer, and are in the same substation.
4. The method for identifying the relationship between transformers in substations and households based on voltage curve similarity and power outage events according to claim 1 is characterized in that: The power-off time comparison specifically includes: Orderly power outage and restoration: when an active power outage is required in a low-voltage area, each user's meter will generate a power outage event with a timestamp and save it in the meter. When the area is restored, the restoration of the main branches of the area will be carried out in an orderly manner. When the main branch is restored, the user meters under the branch will generate restoration information with a timestamp and save it in the user's meter. Read the user's meter information. The smart gateway of the substation area is connected to the concentrator. When all areas are fully powered on, the smart gateway reads the latest power outage and restoration information of the user's meter in the substation area through the concentrator and records it in the smart gateway database. For household transformer analysis, the intelligent gateway of the substation classifies and stores all the user meters according to the start and end time of power outage and restoration of each user's meter, and the user meters with similar power outage and restoration time are classified and stored to form household transformer relationships.
Citation Information
Cited By
Transformer area user address consistency evaluation method and system, electronic equipment and storage medium
CN120218391A
Fault positioning method, system and equipment based on transformer area topology identification, and medium
CN120630090A
Transformer area identification and user binding method based on special transformer acquisition terminal
CN120891267A
A method for identifying a transformer area and binding a user based on a special transformer acquisition terminal
CN120891267B
