Residential electricity meter–transformer relationship identification method based on spatio-temporal aggregated voltage curve

Abstract

The present invention relates to the technical field of power detection, and specifically relates to a residential electricity meter–transformer relationship identification method based on a spatio-temporal aggregated voltage curve. In the method, first, clock synchronization is performed on an electricity meter of a distribution transformer service area device, and a traveling-wave ranging characteristic signal with a timestamp is broadcast to a low-voltage line; second, on the basis of the traveling-wave ranging characteristic signal, a line length from a residential electricity meter device to a fusion terminal of the distribution transformer service area is calculated; the line length from the residential electricity meter device to the fusion terminal of the distribution transformer service area is then determined, and an aggregation method is used to acquire a theoretical voltage fluctuation curve of a residential electricity meter; and a similarity between the theoretical voltage fluctuation curve of the residential electricity meter and an actual voltage fluctuation curve of the residential electricity meter is finally calculated, so as to identify a residential electricity meter–transformer relationship, thereby improving the accuracy of identification.

Description

A method for identifying household-transformer relationships based on voltage spatiotemporal aggregation curves Technical Field

[0001] This invention relates to the field of power detection technology, and more specifically, to a method for identifying household-transformer relationships based on voltage spatiotemporal aggregation curves. Background Technology

[0002] The relationship between a residential electricity meter (REM) and a distribution transformer is fundamental for calculating line losses, fault location, and anti-theft identification in distribution transformer service areas (DAs). Low-voltage lines, often located inside buildings, are the access nodes for REMs and are characterized by complex network structures and frequent changes. During the processing of new REM installations, relocations, and name changes, on-site personnel at power supply companies may mistakenly connect REMs to incorrect low-voltage lines due to frequent changes in low-voltage wiring, leading to inconsistencies in the residential-transformer relationship in the marketing system. Therefore, effective research on identifying residential-transformer relationships is necessary.

[0003] Currently, the construction of smart meters based on high-power line communication (HPLC) technology has achieved the acquisition of electrical quantity data at 96 points per day on the REM (Real Estate Provider) using HPLC technology, providing a data foundation for identifying the relationship between the household and the transformer. The identification of the relationship between the household and the transformer based on smart meter HPLC technology currently includes two categories: characteristic pulse signal identification and electrical quantity similarity identification. The characteristic pulse signal identification method is based on the principle that HPLC cannot communicate across distribution transformers. It sends an HPLC characteristic pulse signal to the REM through the fusion terminal of distribution transformer service area (FTDA). After the REM receives the signal and feeds it back to the FTDA, it can be determined that the REM belongs to this FTDA. The characteristic pulse signal identification method suffers from problems such as large transmission attenuation of the HPLC characteristic pulse signal, crosstalk of the shared neutral line FTDA signal, and weak anti-interference ability, resulting in low accuracy in identifying the relationship between the household and the transformer. The electrical quantity similarity identification method is based on the principle that the changes in electrical quantities of the REM and the distribution transformer are similar. It uses algorithms such as the Pearson correlation coefficient method, discrete Fréchet distance, and outlier point comparison to compare the similarity of the voltage curve changes of the REM and the distribution transformer, thereby determining the relationship between the household and the transformer. However, with the continuous advancement of county-wide photovoltaic construction, the large-scale integration of distributed energy sources such as rooftop photovoltaics and energy storage into the distribution system has resulted in multi-directional power flow in the distribution system. The voltage variation pattern of the REM will be related to multiple power supply points and has a weak correlation with the voltage curve of the distribution transformer. Therefore, the electrical quantity similarity identification method cannot effectively identify the relationship between households and transformers.

[0004] Summary of the Invention

[0005] This invention addresses the problems of weak correlation between the low-voltage outgoing lines of distribution transformers and REM voltage curves, as well as low accuracy in identifying household-transformer relationships in distributed energy (DA) systems. It proposes a household-transformer relationship identification method based on voltage spatiotemporal aggregation curves. First, the meters in the distribution substation equipment are clocked, and a timestamped traveling wave ranging characteristic signal is broadcast to the low-voltage line. Second, the line length from the user's meter to the substation's fusion terminal is calculated based on the traveling wave ranging characteristic signal. Then, the line length is determined, and the theoretical voltage fluctuation curve of the user's meter is obtained using an aggregation method. Finally, the similarity between the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's meter is calculated, thereby achieving household-transformer relationship identification and improving the accuracy of the identification.

[0006] The specific implementation details of this invention are as follows:

[0007] A method for identifying the relationship between a user and a transformer based on a voltage spatiotemporal aggregation curve is proposed. First, the meters of the distribution transformer area equipment are clocked, and a timestamped traveling wave ranging characteristic signal is broadcast to the low-voltage line. Second, the line length from the user's meter to the transformer area's fusion terminal is calculated based on the traveling wave ranging characteristic signal. Then, it is determined whether the line length exceeds a first threshold. Electrical data of distribution transformer area equipment that does not exceed the first threshold is collected. The theoretical voltage fluctuation curve of the user's meter is obtained using an aggregation method. Finally, the similarity between the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's meter is calculated. If the similarity is greater than a third threshold, it is determined that the current user's meter is controlled by the distribution transformer area.

[0008] To better realize the present invention, the household-transformer relationship identification method based on voltage spatiotemporal aggregation curves further includes the following steps:

[0009] Step 1: Set the clock for the meters of the distribution area equipment and broadcast the timestamped traveling wave ranging characteristic signal to the low-voltage line;

[0010] Step 2: Based on the time of acquiring the traveling wave ranging characteristic signal and the propagation speed of the traveling wave ranging characteristic signal, calculate the line length d from the user's meter equipment to the substation convergence terminal. a ;

[0011] Step 3: Determine the line length d from the user's electricity meter to the substation convergence terminal. aIf the first threshold is exceeded, it is determined that the current user's meter device has an incorrect user-transformer relationship. If the first threshold is not exceeded, electrical data of the distribution transformer area device that does not exceed the first threshold is collected and repaired to obtain the repaired electrical data.

[0012] Step 4: Based on the repaired electrical data, obtain the theoretical voltage fluctuation curve of the user's electricity meter using an aggregation method;

[0013] Step 5: Obtain the actual voltage fluctuation curve of the user's electricity meter, calculate the similarity between the theoretical voltage fluctuation curve of the user's electricity meter and the actual voltage fluctuation curve of the user's electricity meter, and if the similarity is greater than the third threshold, then determine that the current user's electricity meter device is controlled by the distribution radio station area.

[0014] To better realize the present invention, step 1 further includes the following steps:

[0015] Step 11: Use the integrated distribution terminal to synchronize clocks with user meters and distributed power sources, and send the timestamp t... a The time synchronization signal is broadcast to the user's electricity meter device, and the timestamp t of the time synchronization signal arriving at the user's electricity meter device is recorded. b ;

[0016] Step 12: Add feedback of timestamp t to the converged terminal in the distribution area c The system records the timestamp t of the time synchronization feedback signal received by the integrated terminal of the distribution area from the user's electricity meter. d Calculate the clock synchronization error t of the user's electricity meter. h ;

[0017] Step 13: Calculate the clock synchronization error t of the user's electricity meter. h Add the clock to the user's electricity meter and calculate the user's electricity meter clock time T. e .

[0018] To better realize the present invention, step 2 further includes the following steps:

[0019] Step 21: Transmit the traveling wave ranging characteristic signal to the three phases of the low-voltage line. Based on the shortest time for the single-phase user meter to receive the traveling wave ranging characteristic signal from the distribution substation fusion terminal, the transmission time of the traveling wave ranging characteristic signal from the distribution substation fusion terminal, and the propagation speed of the traveling wave ranging characteristic signal in different medium lines, calculate the distance d from the single-phase user meter to the distribution substation fusion terminal. a ;

[0020] Step 22: Based on the distance d from the single-phase user meter to the substation convergence terminal a The distance matrix between the user's electricity meter and the distribution substation is established by determining the line lengths from different distributed power sources in the substation area.

[0021] To better realize the present invention, step 3 further includes the following steps:

[0022] Step 31: Determine the distance d from the user's electricity meter to the substation convergence terminal. a If the first threshold △f1 is exceeded, it is determined that the current user meter device has an incorrect user-transformer relationship. If the first threshold △f1 is not exceeded, electrical data of the distribution area device that does not exceed the first threshold △f1 is collected.

[0023] Step 32: Construct k using the Szawisky-Gray filter a A polynomial of order k, and according to the k... a A polynomial fits the user's electricity meter measurement window data f. j The range of data amplitude z is used to obtain the fitted user meter measurement window data h. a ;

[0024] Step 33: Based on the fitted user meter measurement window data h under different amplitudes a Raw data f from user meter measurement windows at different amplitudes jk Calculate the sum of squared residuals h between the fitted user meter data and the original user meter measurement window data. d ;

[0025] Step 34: Based on the sum of squared residuals h d The set least squares multinomial fitting degree e c Raw data f from user meter measurement windows with different amplitudes jk The smoothing coefficient g for fitting user meter measurement data with different amplitudes k The repaired electrical data h was calculated. e .

[0026] To better realize the present invention, step 4 further includes the following steps:

[0027] Step 41: Obtain the initial theoretical voltage u of the user's electricity meter before grid connection of the distributed energy source. z The initial theoretical voltage u after grid connection with distributed energy sources and user meters g Calculate the initial user meter voltage fluctuation sequence point R, and assign user meter voltage fluctuation curve parameters l to the sequence point R. i ;

[0028] Step 42: Based on the initial user meter voltage fluctuation sequence point R and the basis function A of the user meter theoretical voltage fluctuation curve for different control cycles... i( ), User meter voltage fluctuation curves with different control cycles, initial control control value s ai Construct the initial theoretical voltage fluctuation curve U(t) of the user's electricity meter;

[0029] Step 43: In the first iteration of weighted LSPIA, calculate the difference vector γ corresponding to the voltage data points based on the initial user meter theoretical voltage fluctuation curve U(t) and the voltage values ​​of the initial user meter voltage fluctuation sequence points;

[0030] Step 44: Based on the weighting of the influence w of different power distribution area power points on user meters... j The vector difference γ between the voltage fluctuation curves at different DA power supply points and the theoretical voltage fluctuation curve of the user's electricity meter. j Given the set weighted LSPIA constant λ and the difference vector γ corresponding to the voltage data points, calculate the adjustment vector Δu for the weighted LSPIA iteration. f ;

[0031] Step 45: Adjust the vector Δu according to the above. f Initial control point s a Calculate the set of control vertices s of the theoretical voltage fluctuation curve of the user's electricity meter. b The weighted LSPIA is iterated continuously until the fitted theoretical voltage fluctuation curve of the user's electricity meter reaches the second threshold △f2.

[0032] To better realize the present invention, step 5 further includes the following steps:

[0033] Step 51: The length d of the theoretical voltage fluctuation curve of the user's electricity meter x The length d of the actual voltage fluctuation curve of the user's electricity meter y The length d of the voltage fluctuation curve in the intersection region of the theoretical voltage fluctuation curve of the user's electricity meter. xa The length d of the voltage fluctuation curve in the region where it intersects with the actual voltage fluctuation curve of the user's electricity meter. ya Calculate the weights w of the different polygons formed by the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter. a ;

[0034] Step 52: Based on the weight w of the polygon a The number n of polygons formed by the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter. p The theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter form different polygon areas E. ai Calculate the similarity B between the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter. a ;

[0035] Step 53: Determine the similarity B a Whether it is greater than the third threshold △f3, if the similarity B a If the value is greater than the third threshold △f3, then it is determined that the current user's meter device is controlled by the distribution radio station area.

[0036] The present invention has the following beneficial effects:

[0037] (1) This invention takes into account the voltage fluctuation of multiple distributed power sources in the distribution area and the distance between the power source and the low-voltage line of the user's meter. It adopts a spatiotemporal aggregation method to obtain the theoretical voltage fluctuation curve of the user's meter location and compares the similarity with the actual voltage fluctuation curve of the user's meter to realize the identification of the relationship between the user and the transformer, thereby improving the accuracy of identification.

[0038] (2) The present invention uses the traveling wave ranging method to calculate the line length from the user's electricity meter to the distribution transformer and the distributed power source, and compares it with the power supply radius of the distribution transformer area, thereby realizing the identification of user electricity meters with abnormal line lengths. Attached Figure Description

[0039] Figure 1 is a schematic diagram of the household change relationship identification process provided by the present invention.

[0040] Figure 2 is a schematic diagram of the radiation-type distribution station area structure provided by the present invention. Detailed Implementation

[0041] To more clearly illustrate the technical solutions of the embodiments of the present invention, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. It should be understood that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments, and therefore should not be regarded as a limitation on the scope of protection. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "set up," "connected," and "linked" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0043] Example 1:

[0044] This embodiment proposes a method for identifying the relationship between a user and a transformer based on a voltage spatiotemporal aggregation curve. First, the meters of the distribution transformer area equipment are clocked, and a timestamped traveling wave ranging characteristic signal is broadcast to the low-voltage line. Second, the line length from the user's meter equipment to the transformer area fusion terminal is calculated based on the traveling wave ranging characteristic signal. Then, it is determined whether the line length from the user's meter equipment to the transformer area fusion terminal exceeds a first threshold. Electrical data of the distribution transformer area equipment that does not exceed the first threshold are collected, and the theoretical voltage fluctuation curve of the user's meter is obtained by aggregation. Finally, the similarity between the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's meter is calculated. If the similarity is greater than a third threshold, it is determined that the current user's meter equipment is controlled by the distribution transformer area.

[0045] Working principle: In this embodiment, the electricity meters of the distribution substation equipment are first clocked, and the traveling wave ranging characteristic signal with timestamp is broadcast to the low-voltage line. Then, the line length from the user's electricity meter to the substation fusion terminal is calculated based on the traveling wave ranging characteristic signal. Next, the line length from the user's electricity meter to the substation fusion terminal is determined, and the theoretical voltage fluctuation curve of the user's electricity meter is obtained by aggregation. Finally, the similarity between the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter is calculated to realize the identification of the household-transformer relationship and improve the accuracy of household-transformer relationship identification.

[0046] Example 2:

[0047] This embodiment is based on Embodiment 1 above and is described in the form of steps.

[0048] Step 1: Set the clock for the meters of the distribution area equipment and broadcast the timestamped traveling wave ranging characteristic signal to the low-voltage line.

[0049] Step 1 specifically includes the following steps:

[0050] Step 11: Use the integrated distribution terminal to synchronize clocks with user meters and distributed power sources, and send the timestamp t... a The time synchronization signal is broadcast to the user's electricity meter device, and the timestamp t of the time synchronization signal arriving at the user's electricity meter device is recorded. b ;

[0051] Step 12: Add feedback of timestamp t to the converged terminal in the distribution area c The system records the timestamp t of the time synchronization feedback signal received by the integrated terminal of the distribution area from the user's electricity meter. d Calculate the clock synchronization error t of the user's electricity meter. h ;

[0052] Step 13: Calculate the clock synchronization error t of the user's electricity meter. h Add the clock to the user's electricity meter and calculate the user's electricity meter clock time T.e .

[0053] Step 2: Based on the time of acquiring the traveling wave ranging characteristic signal and the propagation speed of the traveling wave ranging characteristic signal, calculate the line length d from the user's meter equipment to the substation convergence terminal. a .

[0054] Step 2 specifically includes the following steps:

[0055] Step 21: Transmit the traveling wave ranging characteristic signal to the three phases of the low-voltage line. Based on the shortest time for the single-phase user meter to receive the traveling wave ranging characteristic signal from the distribution substation fusion terminal, the transmission time of the traveling wave ranging characteristic signal from the distribution substation fusion terminal, and the propagation speed of the traveling wave ranging characteristic signal in different medium lines, calculate the distance d from the single-phase user meter to the distribution substation fusion terminal. a ;

[0056] Step 22: Based on the distance d from the single-phase user meter to the substation convergence terminal a The distance matrix between the user's electricity meter and the distribution substation is established by determining the line lengths from different distributed power sources in the substation area.

[0057] Step 3: Determine the line length d from the user's electricity meter to the substation convergence terminal. a If the threshold is exceeded, it is determined that the current user's meter device has an incorrect user-transformer relationship. If the threshold is not exceeded, electrical data of the distribution substation equipment that does not exceed the threshold is collected and repaired to obtain the repaired electrical data.

[0058] Furthermore, step 3 specifically includes the following steps:

[0059] Step 31: Determine the distance d from the user's electricity meter to the substation convergence terminal. a If the first threshold △f1 is exceeded, it is determined that the current user meter device has an incorrect user-transformer relationship. If the first threshold △f1 is not exceeded, electrical data of the distribution area device that does not exceed the first threshold △f1 is collected.

[0060] Step 32: Construct k using the Szawisky-Gray filter a A polynomial of order k, and according to the k... a A polynomial fits the user's electricity meter measurement window data f. j The range of data amplitude z is used to obtain the fitted user meter measurement window data h. a ;

[0061] Step 33: Based on the fitted user meter measurement window data h under different amplitudes a Raw data f from user meter measurement windows at different amplitudesjk Calculate the sum of squared residuals h between the fitted user meter data and the original user meter measurement window data. d ;

[0062] Step 34: Based on the sum of squared residuals h d The set least squares multinomial fitting degree e c Raw data f from user meter measurement windows with different amplitudes jk The smoothing coefficient g for fitting user meter measurement data with different amplitudes k The repaired electrical data h was calculated. e .

[0063] Step 4: Based on the repaired electrical data, obtain the theoretical voltage fluctuation curve of the user's electricity meter using an aggregation method.

[0064] Furthermore, step 4 specifically includes the following steps:

[0065] Step 41: Obtain the initial theoretical voltage u of the user's electricity meter before grid connection of the distributed energy source. z The initial theoretical voltage u after grid connection with distributed energy sources and user meters g Calculate the initial user meter voltage fluctuation sequence point R, and assign user meter voltage fluctuation curve parameters l to the sequence point R. i ;

[0066] Step 42: Based on the initial user meter voltage fluctuation sequence point R and the basis function A of the user meter theoretical voltage fluctuation curve for different control cycles... i ( ), User meter voltage fluctuation curves with different control cycles, initial control control value s ai Construct the initial theoretical voltage fluctuation curve U(t) of the user's electricity meter;

[0067] Step 43: In the first iteration of weighted LSPIA, calculate the difference vector γ corresponding to the voltage data points based on the initial user meter theoretical voltage fluctuation curve U(t) and the voltage values ​​of the initial user meter voltage fluctuation sequence points;

[0068] Step 44: Based on the weighting of the influence w of different power distribution area power points on user meters... j The vector difference γ between the voltage fluctuation curves at different DA power supply points and the theoretical voltage fluctuation curves of REM theory. j Given the set weighted LSPIA constant λ and the difference vector γ corresponding to the voltage data points, calculate the adjustment vector Δu for the weighted LSPIA iteration. f ;

[0069] Step 45: Adjust the vector Δu according to the above. f Initial control point s aCalculate the set of control vertices s of the REM theoretical voltage fluctuation curve. b The weighted LSPIA is iterated continuously until the fitted REM theoretical voltage fluctuation curve reaches the second threshold △f2.

[0070] Step 5: Obtain the actual voltage fluctuation curve of the user's electricity meter, calculate the similarity between the theoretical voltage fluctuation curve of the user's electricity meter and the actual voltage fluctuation curve of the user's electricity meter, and if the similarity is greater than the third threshold, then determine that the current user's electricity meter device is controlled by the distribution radio station area.

[0071] Step 5 specifically includes the following steps:

[0072] Step 51: The length d of the theoretical voltage fluctuation curve of the user's electricity meter x The length d of the actual voltage fluctuation curve of the user's electricity meter y The length d of the voltage fluctuation curve in the intersection region of the theoretical voltage fluctuation curve of the user's electricity meter. xa The length d of the voltage fluctuation curve in the region where it intersects with the actual voltage fluctuation curve of the user's electricity meter. ya Calculate the weights w of the different polygons formed by the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter. a ;

[0073] Step 52: Based on the weight w of the polygon a The number n of polygons formed by the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter. p The theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter form different polygon areas E. ai Calculate the similarity B between the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter. a ;

[0074] Step 53: Determine the similarity B a Whether it is greater than the third threshold △f3, if the similarity B a If the value is greater than the third threshold △f3, then it is determined that the current user's meter device is controlled by the distribution radio station area.

[0075] The other parts of this embodiment are the same as those in Embodiment 1 above, so they will not be described again.

[0076] Example 3:

[0077] This embodiment is based on any one of Embodiments 1-2 above, and is described in detail with reference to a specific embodiment as shown in Figures 1 and 2. In this embodiment, REM is short for User Meter, FTDA is short for Distribution Utility Terminal, and DA is short for Distribution Area.

[0078] This embodiment is explained in two stages: the first stage is DA device topology ranging, and the second stage is household-transformer relationship identification.

[0079] 1. Household change relationship identification process

[0080] (1) First stage: DA device topology ranging.

[0081] The purpose of DA (Distributed Energy Management) device topology ranging is to measure the low-voltage line length between REM (Resource Energy Management), distributed photovoltaic (PV), and energy storage devices and the distribution transformer, providing a data foundation for DA voltage spatiotemporal curve aggregation. This method measures the length by calculating the traveling wave ranging characteristic time difference between the distribution transformer and the DA device. First, the FTDA (Fiber Optic Distribution Adapter) provides clock synchronization to the DA's meters and distributed energy sources to ensure clock consistency across all DA devices. Second, the FTDA broadcasts a timestamped traveling wave ranging characteristic signal to the low-voltage line. After receiving the traveling wave ranging characteristic signal, the DA device calculates the distance matrix between each DA device based on the received time and the traveling wave propagation speed. Finally, meters with distances exceeding a threshold are identified as having incorrect user-transformer relationships.

[0082] (2) Second stage: Identification of household change relationship.

[0083] First, the quality of the collected electrical data from the DA device was repaired. Then, considering factors such as the DA power supply and the length of the power supply line to the household meter, a spatiotemporal aggregation method was used to obtain the theoretical voltage fluctuation curve at the location of the user's meter. Based on this, the similarity of this curve with the actual voltage fluctuation curve of the REM was compared to identify the relationship between the household and the transformer.

[0084] 2. Household-to-household relationship identification and modeling

[0085] 2.1 DA Device Topology Ranging

[0086] 2.1.1 FTDA clock synchronization

[0087] Traveling wave ranging requires both ends of the ranging device to have high-precision clock synchronization capabilities. The accuracy of GPS clock synchronization to REM (Power Grid Distributed Electricity) is 1 microsecond, while the propagation speed of the characteristic signal in power lines is 300 meters per microsecond, meaning a traveling wave ranging error of 300 meters. Considering the DA (Power Grid Distributed Electricity) power supply radius in China is 500 meters, this error will severely affect the accuracy of identifying the relationship between the user and the transformer. The accuracy of BeiDou satellite clock synchronization in the Asia-Pacific region is 100 nanoseconds, meaning a traveling wave ranging error of 30 meters. Therefore, this paper uses BeiDou satellite to synchronize time with FTDA, distributed power supply, and REM to improve synchronization accuracy and reduce traveling wave ranging error.

[0088] FTDA is an IoT management device installed on the distribution transformer side, enabling data acquisition and edge analysis of devices such as distribution transformers, REMs, distributed energy sources, and residual current devices. Therefore, this paper uses FTDA to provide time synchronization to distributed power sources and REMs.

[0089] After receiving the time synchronization command from the BeiDou satellite, FTDA broadcasts a timestamp (t) to devices such as REM. a The FTDA time synchronization signal. REM and other devices record the timestamp t of the arrival of the FTDA time synchronization signal. b And add sending timestamp t to FTDA feedback c The timestamp t of the REM time feedback signal received by the FTDA. d REM clock synchronization error t h for:

[0090] Where: n a The FTDA sends the same number of time synchronization commands as the number of distributed power sources, corresponding to the number of distributed power sources; t bi The arrival time of the REM record time synchronization information is recorded for different numbers of time synchronization groups; t ci The timestamps for the time synchronization signals fed back to the FTDA by REM with different numbers of time synchronization groups; t di The timestamps for the REM feedback commands received by the FTDA for different time groups.

[0091] After obtaining the REM time synchronization error, the clock time error is added to the clock of each meter, and the REM clock time T is... e For: T e =t f +t h (2)

[0092] In the formula: t f This is the time for FTDA synchronization.

[0093] 2.1.2 REM line length distance measurement

[0094] Traveling wave distance measurement (TWD) is a method for measuring the distance between devices along a power line. This method estimates the distance by measuring the travel time of a traveling wave pulse signal along the power line. Specifically, TWD involves adding TWD functionality to the FTDA (Power Distribution Analyzer) and HPLC (Power Utility Model) modules of a distribution transformer. When the FTDA transmits a TWD characteristic signal, the signal propagates along the low-voltage line at the speed of light. When the REM (Receiving Meter) receives the TWD characteristic signal, the distance between the FTDA and the REM can be calculated based on the time difference between transmission and reception.

[0095] In the embodiments described above, traveling wave ranging modules are installed at the power sources such as FTDA, distributed photovoltaic, and energy storage. By calculating the low-voltage line length between the power source and REM, basic data is provided for the line length influence factor in the aggregation of DA voltage spatiotemporal curves.

[0096] The FTDA transmits traveling wave ranging characteristic signals on each of the three phases of the low-voltage line. These signals first propagate along the same phase to the end of the low-voltage line and then couple to the other two phases through the neutral wire. The single-phase REM receives the same-phase traveling wave ranging characteristic signal in the shortest time, while receiving the other two phases takes longer. Therefore, the single-phase REM line length measurement and phase identification can be performed by measuring the time it takes for the REM to receive the shortest traveling wave ranging characteristic signal. Since the three-phase REM receives the traveling wave ranging characteristic signals of the three phases in equal times, the low-voltage line length between the three-phase REM and the FTDA or distributed power source can be calculated by averaging the time it takes for the three-phase REM to receive the traveling wave ranging characteristic signals.

[0097] The distance d from single-phase REM to FTDA a For: d a =(t kmin -t1)v a (3)

[0098] In the formula: t kmin The shortest traveling wave ranging characteristic signal time for a single-phase REM to receive the FTDA; t l v is the transmission time of the traveling wave ranging characteristic signal of FTDA; a The propagation speed of the traveling wave ranging characteristic signal in different medium lines. The distance d from the three-phase REM to the FTDA. b for:

[0099] Where: n b Three-phase, 0.4kV; t oi The time t represents the time it takes for the traveling wave ranging characteristic signal received by the three-phase REM at different phases from the FTDA. pi The transmission time of the characteristic traveling wave ranging signal of FTDA under different phases; v b This represents the propagation speed of the traveling wave ranging characteristic signal in different medium lines.

[0100] When performing REM ranging, the distributed power source transmits traveling wave ranging characteristic signals to the three phases of the low-voltage line respectively. The line ranging method is the same as FTDA, and will not be described again in this embodiment.

[0101] 2.1.3 Establish the distance matrix from REM to multiple power sources

[0102] The distance matrix from REM to multiple power sources is established to analyze the significance of line length on the fluctuation of the REM voltage curve. Under the same output current from distributed power sources, shorter line lengths result in lower impedance and a greater impact on the fluctuation of the REM voltage curve, and vice versa.

[0103] Distance matrix L from REM to FTDA and distributed generation d for:

[0104] In the formula: d al The line lengths from REM to FTDA are d1, d2, ..., d. m denoted as REM, representing the line length from REM to the different distributed power sources in DA; m represents the number of distributed power sources in DA.

[0105] When the line length from REM to FTDA is less than the threshold Δf1, it indicates that REM belongs to this DA.

[0106] 2.2 Identification of Household Change Relationships

[0107] 2.2.1 DA Electrical Data Recovery

[0108] REM collects electrical measurement data at a frequency of 15 minutes per iteration. The data volume is enormous, and it is susceptible to noise during data acquisition and transmission, leading to data errors, missing data, and other anomalies. Therefore, the collected DA electrical data needs to be repaired before identifying the relationship between the household and the transformer.

[0109] The Savitzky-Gray (SG) filter is a noise data processing method that uses convolution to process noisy data. It employs linear least squares to fit polynomials to consecutive data points adjacent to the noise, thus achieving balanced data denoising. This filter can improve data accuracy without altering the trend of REM electrical measurement data. Compared to other filters, the SG filter can adaptively adjust its filtering parameters according to the REM measurement data curve, thus preserving the overall variation of the original data. It features fast filtering speed and high accuracy; therefore, the SG filter is used for data restoration of DA measurement data.

[0110] REM measurement window data is f j If the range of data amplitude is z, then the REM measurement window data is j = -z,…,0,…,z. By constructing k... a A polynomial of order h is used to fit the above data, and the fitted REM measurement window data is h. a for:

[0111] In the formula: c iFor REM measurement window data of different orders, k for measuring REM time a Power of 1.

[0112] The sum of squared residuals h between the fitted REM data and the original values d for:

[0113] Where: h ai For fitted REM measurement window data of different amplitudes; f ji The raw data for REM measurement windows of different amplitudes.

[0114] The purpose of using the least squares method to calculate the REM measurement window data is to find the optimal fit, minimizing the sum of squared residuals, i.e., the partial derivatives of the residuals with respect to the polynomial coefficients are zero. Once the number of sides, order, and data to be fitted for the REM measurement data are clear, data repair is completed by estimating the center point value within the window and continuously moving the window. This process is equivalent to convolving the REM unit impulse response data input to the SG filter to obtain the filtered data h. e for:

[0115] In the formula: e c f is the least squares multinomial fitting order; jk The raw data for different amplitudes of the input SG filter; g k A smoothing coefficient is fitted to REM measurement data of different amplitudes.

[0116] 2.2.2 Aggregation of DA voltage curves

[0117] With a high proportion of distributed energy sources integrated into the distribution network (DA), the power flow of DA changes from unidirectional to multidirectional. Changes in the REM voltage curve will be related to factors such as the output of multiple DA power sources, current flow direction, and low-voltage line length. Therefore, the purpose of DA voltage curve aggregation is to consider various influencing factors of DA power sources, calculate the theoretical voltage fluctuation curve at the current location of REM, and use this as the basis for identifying the household-transformer topology. Based on this, by comparing the similarity between the theoretical voltage fluctuation curve at the current location of REM and the actual voltage fluctuation curve of REM, the household-transformer relationship can be identified.

[0118] Voltage curve aggregation analysis was performed on a radial distributed energy source (DA) containing distributed energy resources, and its typical structure is shown in Figure 2. The DA has m distributed energy sources and n REMs (Resources, Meters, and Modules); the voltage at the beginning of the low-voltage line of the DA measured by FTDA is u0; the voltage of the m-th distributed energy source is u... am The power output of the power supply is p amIn distributed energy, distributed photovoltaics only generates electricity; distributed energy storage is equivalent to REM (Remote Energy Storage) when storing electrical energy, and equivalent to distributed photovoltaics when discharging; the voltage of the nth REM is u. n The power consumption is p n The line length from the nth REM to the distribution transformer and distributed energy source is calculated by equation (5).

[0119] Before the distributed energy source (DA) is connected to the grid, the voltage drop of REM is directly proportional to the length of the low-voltage line; that is, as the line length increases, the voltage of REM gradually decreases, and the voltage fluctuation curve is similar to the voltage u0 at the beginning of the 0.4kV outgoing line of DA. After the distributed energy source is connected to the grid, assuming that the photovoltaic and energy storage of the distributed energy source only have active power, the voltage of REM will increase. Its voltage is related to the power of multiple distributed energy sources in DA, the line length of REM from multiple power sources, etc. Therefore, the voltage fluctuation curve of REM is related to the power of multiple power sources in DA and the line length from the power source.

[0120] The DA power supply radius is 500 meters, and the resistance value of wires of different diameters and materials varies little. Assuming the resistance variation between different wires in the low-voltage line is negligible, the resistance per meter of the low-voltage line is r (Ω). a The low-voltage line length from the power supply point to the nth REM is o. an For example, before distributed energy is connected to the grid, the line loss L from the power source to the nth REM line is... ln for:

[0121] In the formula: p n q represents the active power of the nth REM; n The reactive power of the nth REM; u n This is the measured voltage of the nth REM.

[0122] After distributed energy is connected to the grid, the line loss L from the m power sources to the nth REM line is... s for:

[0123] In the formula: p an The active power supplied to the grid for the nth REM; p fi The active power provided by different distributed energy sources to the nth REM; q an To provide reactive power to the nth REM in the power grid; u an The voltage of the nth REM; o ani Let be the length of the low-voltage line from different distributed energy sources in DA to the nth REM. When the grid-connected capacity of DA's distributed energy sources is larger, the more distributed energy is absorbed by the DA load, and the smaller the line loss is; when the grid-connected capacity of distributed energy sources is greater than the DA load, the power of the distribution transformer will be reversed.

[0124] Before the DA distributed energy source is connected to the grid, the voltage drop Δu of the REM voltage is... z It is positive and proportional to the length of the low-voltage line. The initial theoretical voltage u of the nth REM. zn for:

[0125] In the formula: u n-1 Let u0 be the voltage of the (n-1)th REM. If u0 is the voltage of the first REM of DA, then u0 is the voltage at the beginning of the low-voltage line of DA. The initial theoretical voltage of all REMs in DA can be obtained by calculating the voltage of each REM in turn according to their distance from the beginning of the low-voltage line of DA.

[0126] After the distributed energy source is connected to the grid, assuming that the distributed energy source only provides active power, the initial theoretical voltage u of the nth REM is... gn for:

[0127] As can be seen from the above formula, when the DA distributed energy is connected to the grid, the voltage of REM will increase, and the increase is related to the power of multiple power sources in DA and the line length from the power source point.

[0128] Assuming the power of REM remains constant before and after grid connection of distributed energy sources, the impact of the m-th distributed energy source on the voltage fluctuation curve of the n-th REM includes two parts: the line loss from the m power sources to the n-th REM and the impact of REM itself. The weight w of this impact on the REM voltage curve fluctuation is... m for:

[0129] In the formula: p bi The active power provided by different distributed energy sources to the nth REM; p fm The active power provided by the m-th distributed energy source to the n-th REM.

[0130] Voltage spatiotemporal aggregation is a method for fitting REM voltage curves that considers multiple factors. This method uses equations to aggregate REM voltage spatiotemporal data into a smooth voltage fluctuation curve. Specifically, in the time dimension, it aggregates the voltage fluctuation curves of multiple power sources within the DA (Data Source) affecting REM; in the spatial dimension, it aggregates the line lengths from multiple power sources within the DA to REM. This method can calculate the theoretical values ​​of the impact of multiple power sources within the DA on REM voltage fluctuations, allowing observation of the intrinsic correlation of voltage spatiotemporal data and understanding of the changing relationships within the DA data.

[0131] The Progressive and Iterative Approximation for Least Squares (LSPIA) method is a curve fitting technique that uses iterative control points to construct a curve that approximates the true value and converges it to the least squares fitting result. LSPIA is computationally efficient and suitable for handling large-scale distribution substation (DA) REM datasets. However, in this paper, the voltage fluctuations of multiple power sources within the DA and the length of low-voltage lines from the REM have different impacts on the REM voltage fluctuations. Therefore, this paper adds weights to the fitting data of different DA power sources based on LSPIA to calculate the theoretical voltage fluctuation curve of the REM.

[0132] Based on the grid connection and disconnection of DA distributed energy, an initial REM voltage fluctuation sequence point is calculated according to equations (11) and (12). Where, n h Specify the number of points on the REM voltage curve and assign a parameter l to each point on the REM voltage curve. i (i = 1, 2, ..., n) h The REM voltage curve parameters satisfy:

[0133] Select the initial control set for the REM voltage curve. n l To control the number of times the REM voltage curve is generated; construct an initial theoretical REM voltage curve U(t):

[0134] In the formula: A i (t) represents the basis functions of the REM theoretical voltage curves for different control orders; s ai These are the control values ​​for different control cycles.

[0135] In the first iteration of weighted LSPIA, the difference vector γ corresponding to the voltage data point at point i is calculated: γ = R i -U i (t) (16)

[0136] In the formula: U i (t) represents the voltage value at point i on the initial REM theoretical voltage curve; R i Let be the voltage value at point i in the initial REM voltage fluctuation sequence.

[0137] Adjustment vector Δu in weighted LSPIA iteration f for:

[0138] Where: n o The number of DA power points; wj The weights of the impact of different DA power supply points on REM; γ j λ is the vector difference between the voltage curves at different DA power supply points and the theoretical voltage curves of REM; λ is a constant for weighted LSPIA.

[0139] Based on this, define the set of control vertices s of the REM theoretical voltage curve. b : s b =s a +Δu f (18)

[0140] After the first iteration, the control points of the REM theoretical voltage curve are determined by the initial control point and the adjustment vector Δu. f The second iteration control point is calculated using equation (18). The weighted LSPIA is iterated continuously until the fitted REM theoretical voltage fluctuation curve meets the accuracy Δf2.

[0141] 2.2.3 Comparison of REM voltage curve similarity

[0142] Within a DA, the voltage variation of REM is related to the variation of multiple power supply points. Therefore, the theoretical voltage fluctuation curve of REM can be compared with the actual voltage fluctuation curve. If the variation of the two is similar, then REM belongs to this DA; otherwise, it does not belong to this DA.

[0143] Locality in between polylines (LIP) is a curve similarity metric that measures the similarity between two REM voltage fluctuation curves by calculating the area between them. A zero area indicates that the two REM voltage fluctuation regions are completely identical; a larger area indicates a greater difference in similarity between the two REM voltage fluctuation curves. The LIP method is highly robust to interference. Therefore, LIP is used to calculate the similarity between theoretical and actual REM voltage fluctuation curves.

[0144] The similarity between the REM theoretical voltage fluctuation curve and the actual voltage fluctuation curve (B) a for:

[0145] Where: n p E represents the number of polygons formed between the REM theoretical voltage fluctuation curve and the actual voltage fluctuation curve; ai The two voltage curves REM form different polygon areas; w ai Different polygon weights are used to construct the two voltage curves of REM.

[0146] The weight is determined by the ratio of the perimeter of the region formed by the two voltage curves REM to the total length of the curve, and the weight wa for:

[0147] In the formula: d x and d y These represent the lengths of the theoretical and actual voltage fluctuation curves, respectively; d xa and d ya These represent the lengths of the voltage fluctuation curves in the intersection region of the REM theoretical and actual voltage fluctuation curves, respectively.

[0148] When the similarity between the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of REM is greater than Δf3, it indicates that REM belongs to this DA.

[0149] This embodiment addresses the problem of weak correlation between the low-voltage outgoing lines of distribution transformers and the REM voltage curves in distributed energy (DA) systems, resulting in low accuracy in identifying the relationship between households and transformers. It proposes a method for identifying household-transformer relationships based on voltage spatiotemporal aggregation curves. This method has the following characteristics:

[0150] 1) The traveling wave ranging method was used to calculate the line length from REM to the distribution transformer and distributed power source, and the length was compared with the power supply radius of DA to identify REM with abnormal line length.

[0151] 2) The impact of distributed energy grid connection and REM line length on REM voltage was analyzed. The theoretical REM voltage fluctuation curve was obtained by voltage spatiotemporal curve aggregation, which has the characteristic of small error.

[0152] 3) The accuracy of identifying the relationship between households and transformers is improved by comparing REM theory with the actual REM voltage fluctuation curve.

[0153] The other parts of this embodiment are the same as any one of the above embodiments 1-2, so they will not be described again.

[0154] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Any simple modifications or equivalent changes made to the above embodiments based on the technical essence of the present invention shall fall within the protection scope of the present invention.

Claims

1. A method for identifying the relationship between household and variable based on voltage spatiotemporal aggregation curve, characterized in that, First, the electricity meters of the distribution substation equipment are clocked, and a timestamped traveling wave ranging characteristic signal is broadcast to the low-voltage line. Second, the line length from the user's electricity meter to the substation fusion terminal is calculated based on the traveling wave ranging characteristic signal. Then, it is determined whether the line length from the user's electricity meter to the substation fusion terminal exceeds a first threshold. Electrical data of the distribution substation equipment that does not exceed the first threshold is collected, and the theoretical voltage fluctuation curve of the user's electricity meter is obtained by aggregation. Finally, the similarity between the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter is calculated. If the similarity is greater than a third threshold, it is determined that the current user's electricity meter is controlled by the distribution substation.

2. The method according to claim 1, wherein, The household-transformer relationship identification method based on voltage spatiotemporal aggregation curves specifically includes the following steps: Step 1: Set the clock for the meters of the distribution area equipment and broadcast the timestamped traveling wave ranging characteristic signal to the low-voltage line; Step 2: calculating the line length d from the user's electric meter device to the transformer area fusion terminal according to the time of acquiring the traveling wave ranging characteristic signal and the propagation speed of the traveling wave ranging characteristic signal a ; Step 3: Determine the line length d from the user's electricity meter to the substation convergence terminal. a If the first threshold is exceeded, it is determined that the current user's meter device has an incorrect user-transformer relationship. If the first threshold is not exceeded, electrical data of the distribution transformer area device that does not exceed the first threshold is collected and repaired to obtain the repaired electrical data. Step 4: Based on the repaired electrical data, obtain the theoretical voltage fluctuation curve of the user's electricity meter using an aggregation method; Step 5: Obtain the actual voltage fluctuation curve of the user's electricity meter, calculate the similarity between the theoretical voltage fluctuation curve of the user's electricity meter and the actual voltage fluctuation curve of the user's electricity meter, and if the similarity is greater than the third threshold, determine that the current user's electricity meter device is controlled by the distribution radio station area.

3. The method of claim 2, wherein the method further comprises: Step 1 specifically includes the following steps: Step 11: Adopting the district fusion terminal to clock the user electric meter and the distributed power, broadcast the time signal with time stamp t a to the user electric meter equipment, and record the time stamp t b of the time signal arriving at the user electric meter equipment; Step 12: increase the feedback sending time stamp t to the station area fusion terminal c the time of the signal, and record the time stamp t of the user electric meter receiving the time feedback signal of the user electric meter d , calculate the clock time error t of the user electric meter h ; Step 13: The clock of the user's meter is synchronized with the error t h to the clock of the user's meter, the time T of the clock of the user's meter is calculated e .

4. The method of claim 3, wherein the method further comprises: Step 2 specifically includes the following steps: Step 21: transmit the traveling wave distance measurement characteristic signal to the low-voltage line three-phase, according to the shortest traveling wave distance measurement characteristic signal time received by the single-phase user meter to the substation fusion terminal, the traveling wave distance measurement characteristic signal transmission time of the substation fusion terminal, the propagation speed of the traveling wave distance measurement characteristic signal in different medium lines, calculate the distance d from the single-phase user meter to the substation fusion terminal a ; Step 22: Distance d of the single phase consumer meter to the transformer fusion terminal is determined according to the line length of the single phase consumer meter to the transformer fusion terminal a The distance matrix of the single phase consumer meter to the transformer fusion terminal is established according to the line length of the single phase consumer meter to the transformer fusion terminal.

5. The method of claim 4, wherein the method further comprises: Step 3 specifically includes the following steps: Step 31: judging the distance d of the user electric meter to the distribution area fusion terminal a whether the first threshold value Δf1 is exceeded, if the first threshold value Δf1 is exceeded, judging that the user electric meter device is in error, if the first threshold value Δf1 is not exceeded, collecting the electrical data of the distribution area device which does not exceed the first threshold value Δf1; Step 32: Construct k a order polynomial using Savitsky-Golay filter, and fit the user's electric meter measurement window data f a j , the range of data amplitude z, to obtain the fitted user's electric meter measurement window data h a ;​ Step 33: Calculate the residual sum of squares of the fitted user meter data and the user meter measurement window raw data h a for different amplitudes jk Step 34: Calculate the residual sum of squares of the fitted user meter data and the user meter measurement window raw data h d for different amplitudes Step 34: According to the residual sum of squares h d , the set least square polynomial fitting order e c , the original data of the user electricity meter measurement window of different amplitudes f jk , the fitting and smoothing coefficient of the user electricity meter measurement data of different amplitudes g k , the calculated repaired electrical data h e .

6. The method of claim 5, wherein the method further comprises: Step 4 specifically includes the following steps: Step 41: Obtain the initial theoretical voltage u of the user's electricity meter distributed energy before grid connection z And the initial theoretical voltage u of the user's electricity meter distributed energy after grid connection g , calculate the initial user's electricity meter voltage fluctuation sequence point R, and distribute the user's electricity meter voltage fluctuation curve parameter l at the sequence point R i ; Step 42: According to the initial user meter voltage fluctuation sequence point R, the user meter theoretical voltage fluctuation curve base function A of different control times i (), the initial control control value s of the user meter voltage fluctuation curve of different control times ai , construct the initial user meter theoretical voltage fluctuation curve U(t); Step 43: In the first iteration of the weighted least squares algorithm's asymptotic iterative approximation algorithm, calculate the difference vector γ corresponding to the voltage data points based on the initial user meter's theoretical voltage fluctuation curve U(t) and the voltage values ​​of the initial user meter's voltage fluctuation sequence points. Step 44: Weights w of the influence of different distribution area power supply points on user electricity meters j Vector difference γ of the voltage fluctuation curve of different DA power supply points and the theoretical voltage fluctuation curve of user electricity meters j Progressive iteration of the set weighted least squares algorithm a constant λ of the proximal algorithm and a difference vector γ corresponding to the voltage data points, to calculate an adjustment vector Δu of an iterative approximation algorithm of the weighted least squares algorithm f ; Step 45: Adjusting the control point s according to the adjustment vector Δu f , the initial control point s a , the control vertex set s of the theoretical voltage fluctuation curve of the user's electric meter b , the progressive iterative approximation algorithm of the weighted least squares algorithm is iterated constantly until the fitted theoretical voltage fluctuation curve of the user's electric meter reaches the second threshold value Δf2.

7. The method of claim 6, wherein the method further comprises: Step 5 specifically includes the following steps: Step 51: The length d of the theoretical voltage fluctuation curve of the user's electricity meter x The length d of the actual voltage fluctuation curve of the user's electricity meter y The length d of the voltage fluctuation curve in the intersection region of the theoretical voltage fluctuation curve of the user's electricity meter. xa The length d of the voltage fluctuation curve in the region where it intersects with the actual voltage fluctuation curve of the user's electricity meter. ya Calculate the weights w of the different polygons formed by the theoretical voltage fluctuation curve and the actual voltage fluctuation curve of the user's electricity meter. a ; Step 52: According to the weight w of the polygon a The number of polygons n formed between the theoretical voltage fluctuation curve of the user's electric meter and the actual voltage fluctuation curve p The different polygon areas E formed between the theoretical voltage fluctuation curve of the user's electric meter and the actual voltage fluctuation curve ai Calculate the similarity B between the theoretical voltage fluctuation curve of the user's electric meter and the actual voltage fluctuation curve a ; Step 53: judging whether the similarity B a is greater than a third threshold value Af3, if the similarity B a is greater than the third threshold value Af3, judging that the current user meter device is controlled by the power distribution station.

Images