A low-voltage distribution area single-phase user phase identification method and system

Abstract

The application discloses a low-voltage power distribution area single-phase user phase identification method and system, which is applied to the technical field of power distribution network topology identification, obtains single-phase user active power and distribution transformer active power of a low-voltage power distribution area in a preset time period, maps the single-phase user active power and a connection phase corresponding to the single-phase user active power to obtain an adjacency relation matrix, configures a prior probability according to the adjacency relation matrix to obtain an initial prior probability matrix, takes first-day power data as input, calculates likelihood probability and posterior probability of each single-phase user and all assumed phases, selects an assumed phase corresponding to the maximum posterior probability as the phase of each single-phase user, then takes power data of the next day as input, removes user power of the recorded phase from the side of the area, and again calculates the likelihood probability and the posterior probability of all assumptions, until no new date data appears, and the above method effectively improves the single-phase user phase identification precision.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power distribution network topology identification, and particularly relates to a low-voltage power distribution area single-phase user phase identification method and system. BACKGROUND

[0002] As the last mile of the power supply link, the low-voltage power distribution area is closely connected with the vast number of end users, and is the key area to carry massive distributed flexible resources and promote the high-quality development of the distribution network. However, due to the opaque operation state of equipment, imperfect data management mechanism and other front-line problems, the current identification of single-phase users and their actual belonging phase is still mainly based on manual line searching. When facing a large amount of data, the identification error rate is high, the efficiency is low, the labor cost is extremely high, and it is impossible to carry out comprehensively. Therefore, for a long time, the business contradictions of three-phase imbalance, reduced load carrying capacity, and difficult technical loss reduction frequently occur in the low-voltage power distribution area, making it difficult to accommodate the massive access of flexible resources such as household photovoltaic and charging piles, and hindering the construction of new power systems. SUMMARY

[0003] To solve the above technical problems, the embodiments of the present application provide a low-voltage power distribution area single-phase user phase identification method and system to solve the problem of low identification accuracy in the identification of single-phase user phase in the prior art.

[0004] The first aspect of the embodiments of the present application provides a low-voltage power distribution area single-phase user phase identification method, which comprises:

[0005] Obtaining the active power of the single-phase user of the low-voltage power distribution area in a preset time period;

[0006] Mapping the active power of the single-phase user and the connection phase corresponding to the active power of the single-phase user to obtain an adjacency relation matrix, configuring a prior probability according to the adjacency relation matrix to obtain an initial prior probability matrix;

[0007] Taking the active power of the single-phase user in the starting day in the preset time period as a to-be-calculated power, calculating the likelihood probability of the single-phase user corresponding to the to-be-calculated power and all hypothetical phases to obtain a plurality of likelihood probability values, and based on each likelihood probability value and the initial prior probability matrix, calculating the posterior probability of the to-be-calculated power and all hypothetical phases to obtain a plurality of posterior probability values;

[0008] The hypothesis phase corresponding to the maximum posterior probability value is selected as the phase of the single-phase user, and the initial prior probability matrix is updated according to each posterior probability value to obtain an updated prior probability matrix. After the active power of the single-phase user in the next day of the starting day is determined as new to-be-calculated power, the step of calculating the likelihood probability of the single-phase user and all hypothesis phases corresponding to the to-be-calculated power is performed, and phase identification is performed by using the updated prior probability matrix and the new to-be-calculated power until the phases of all single-phase user active powers in the preset time period are identified, the updating is stopped, and the final identification result is output.

[0009] In a possible implementation manner of the first aspect, the single-phase user active power of the low-voltage power distribution area in the preset time period is obtained, including:

[0010] The first split-phase active power and the second split-phase active power of the low-voltage power distribution area in the preset time period are obtained, wherein the second split-phase active power includes single-phase user split-phase active power and three-phase user split-phase active power.

[0011] The first split-phase active power is cleaned based on the three-phase user split-phase active power to obtain the single-phase user active power.

[0012] In a possible implementation manner of the first aspect, the likelihood probability of the single-phase user and all hypothesis phases corresponding to the to-be-calculated power is calculated to obtain a plurality of likelihood probability values, including:

[0013] The curve correlation between the to-be-calculated power and the distribution transformer phase power corresponding to all hypothesis phases is calculated to obtain a Pearson correlation coefficient.

[0014] The Pearson correlation coefficient is scaled by using a softplus function to obtain a likelihood function calculation formula, and the likelihood probability of the to-be-calculated power and all hypothesis phases is calculated according to the likelihood function calculation formula to obtain a plurality of likelihood probability values.

[0015] In a possible implementation manner of the first aspect, the calculation formula of the posterior probability value is:

[0016]

[0017] In the formula, is the single-phase user active power, is the prior probability matrix, is the likelihood probability value.

[0018] In a possible implementation manner of the first aspect, after the hypothesis phase corresponding to the maximum posterior probability value is selected as the phase of the single-phase user, the method further includes:

[0019] The phase of the to-be-calculated power is stored in a matrix to obtain a first matrix.

[0020] A new adjacency matrix is constructed, and the hypothetical phases in the first matrix are mapped to the new adjacency matrix.

[0021] In a possible implementation of the first aspect, determining the single-phase user active power in the next day of the starting day as the new to-be-calculated power comprises:

[0022] The distribution transformer active power and the single-phase user active power in the next day of the starting day are selected.

[0023] The new to-be-calculated power is obtained according to the distribution transformer active power, the single-phase user active power, and the new adjacency matrix.

[0024] To solve the same technical problem, a second aspect of the embodiment of the application provides a low-voltage power distribution area single-phase user phase identification system, comprising:

[0025] An acquisition module is configured to acquire single-phase user active power of a low-voltage power distribution area in a preset time period.

[0026] A mapping module is configured to map the single-phase user active power and a connection phase corresponding to the single-phase user active power to obtain an adjacency relationship matrix, and configure a prior probability according to the adjacency relationship matrix to obtain an initial prior probability matrix.

[0027] A first calculation module is configured to determine single-phase user active power in a starting day in the preset time period as to-be-calculated power, calculate likelihood probabilities of a single-phase user and all hypothetical phases corresponding to the to-be-calculated power to obtain a plurality of likelihood probability values, calculate posterior probabilities of the to-be-calculated power and all hypothetical phases based on the plurality of likelihood probability values and the initial prior probability matrix to obtain a plurality of posterior probability values, and select a hypothetical phase corresponding to a maximum posterior probability value as a phase of the single-phase user.

[0028] A second calculation module is configured to update the initial prior probability matrix according to the plurality of posterior probability values to obtain an updated prior probability matrix, determine single-phase user active power in a next day of the starting day as new to-be-calculated power, and then perform the step of calculating likelihood probabilities of a single-phase user and all hypothetical phases corresponding to the new to-be-calculated power, and perform phase identification by using the updated prior probability matrix and the new to-be-calculated power until phase identification of all single-phase user active powers in the preset time period is completed, and stop updating, and output a final identification result.

[0029] In a possible implementation of the second aspect, the acquisition module comprises a first acquisition unit and a second acquisition unit.

[0030] The first acquisition unit is used for acquiring first split-phase active power and second split-phase active power of a low-voltage distribution area in a preset time period, wherein the second split-phase active power comprises single-phase user split-phase active power and three-phase user split-phase active power;

[0031] The second acquisition unit is used for cleaning the first split-phase active power based on the three-phase user split-phase active power to obtain single-phase user active power.

[0032] The third aspect of the embodiment of the present application provides a computer device, comprising:

[0033] The memory is used for storing the computer program;

[0034] The processor is used for implementing the steps of the low-voltage distribution area single-phase user phase identification method of the first aspect when executing the computer program.

[0035] The fourth aspect of the embodiment of the present application provides a storage medium, and the storage medium stores a computer program, and the computer program is executed by the processor to implement the steps of the low-voltage distribution area single-phase user phase identification method of the first aspect.

[0036] The technical scheme of the present application has the following advantages:

[0037] The low-voltage distribution area single-phase user phase identification method provided by the embodiment of the present application acquires single-phase user active power and distribution transformer active power of a low-voltage distribution area in a preset time period; maps the single-phase user active power and a connection phase corresponding to the single-phase user active power to obtain an adjacency matrix, configures a prior probability according to the adjacency matrix to obtain an initial prior probability matrix, takes single-phase user active power in a starting day in the preset time period as to-be-calculated power, calculates likelihood probability and posterior probability of the to-be-calculated power and all assumed phases, and selects an assumed phase corresponding to the maximum posterior probability as a phase of the to-be-calculated power. Then, the next day power data is taken as input, the user power of the recorded phase is removed from the distribution area side, and the likelihood probability and the posterior probability of all assumptions are calculated again until no new date data appears, and the single-phase user phase identification result is output. The above method effectively improves the single-phase user phase identification accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical scheme in the specific embodiments or prior art of the present application, the drawings needed in the description of the specific embodiments or prior art will be briefly introduced. Obviously, the drawings in the following description are some embodiments of the present application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0039] Figure 1A recognition flowchart of the low-voltage power distribution area single-phase user phase recognition method in the embodiment of the present application;

[0040] Figure 2 A low-voltage power distribution area topology diagram of the low-voltage power distribution area single-phase user phase recognition method in the embodiment of the present application;

[0041] Figure 3 A phase recognition method flowchart of the low-voltage power distribution area single-phase user phase recognition method in the embodiment of the present application;

[0042] Figure 4 A likelihood function curve diagram of the low-voltage power distribution area single-phase user phase recognition method in the embodiment of the present application;

[0043] Figure 5 A block diagram of the low-voltage power distribution area single-phase user phase recognition system in the embodiment of the present application;

[0044] The reference signs are: 500, a low-voltage power distribution area single-phase user phase recognition system; 501, an acquisition module; 502, a mapping module; 503, a first calculation module; 504, a second calculation module. DETAILED DESCRIPTION

[0045] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0046] The low-voltage power distribution area single-phase user phase recognition provided by the embodiments of the present application is as shown in Figure 1 Figure 1 A low-voltage power distribution area single-phase user phase recognition flowchart, including steps S101 to S104, each step is specifically as follows:

[0047] S101, acquiring the single-phase user active power of the low-voltage power distribution area in a preset time period.

[0048] In the embodiment, Figure 2 A low-voltage power distribution area topology diagram, mainly containing a power distribution transformer, an overhead line, a low-voltage single-phase / three-phase user, etc., as shown in Figure 2 In the embodiment, the node ABC represents a three-phase user, and the node A, B or C is a single-phase user measurement configuration. Only the power distribution transformer outlet and the terminal user are deployed with acquisition devices, and the power, voltage, current and other load measurement information can be acquired by using a smart meter and uploaded to a cloud service system at regular intervals. The remaining nodes are usually in an unobservable state due to the lack of acquisition and monitoring equipment. ​

[0049] Since the low-voltage distribution area only configures the acquisition device for the first and last terminal equipment, as shown in the prior art, Figure 2 As shown in the prior art, only two months of distribution outlet and low-voltage user split-phase active power can be obtained from the business system database. Since the phase of the three-phase user is known, and the reverse sequence access or wrong connection is relatively small, the split-phase power of the three-phase user can be removed from the distribution outlet, that is, the power of all three-phase users in the phase is subtracted from the split-phase active power of the distribution outlet, so that the three-phase user no longer participates in the phase identification calculation, and finally the single-phase user power matrix is obtained, so as to reduce the subsequent model calculation dimension, wherein the split-phase power matrix of the distribution outlet is:

[0050]

[0051] The single-phase user power matrix is:

[0052]

[0053] In the formula, , g respectively represent the split-phase power matrix of the distribution outlet and the single-phase user active time sequence power matrix; A, B, and C are phase symbols, N is the total number of single-phase users under a certain area, and T is the last data collection time every day, usually 96.

[0054] In an embodiment, the single-phase user active power of the low-voltage distribution area in a preset time period is obtained, including:

[0055] The first split-phase active power and the second split-phase active power of the low-voltage distribution area in the preset time period are obtained, wherein the second split-phase active power includes single-phase user split-phase active power and three-phase user split-phase active power;

[0056] Based on the three-phase user split-phase active power, the first split-phase active power is cleaned to obtain the single-phase user active power.

[0057] In this embodiment, as shown in the prior art, Figure 3 Then, the two-month split-phase active power of the distribution outlet and the low-voltage user is obtained, and the active power of each phase of the three-phase user is removed from the split-phase power of the distribution outlet to construct a time sequence power matrix with a daily interval. The power of all three-phase users in the phase is subtracted from the split-phase active power of the distribution outlet, and finally the time sequence power matrix of the single-phase user is obtained, which can also be called single-phase user split-phase active power. The specific method of removing the split-phase power from the distribution outlet is to subtract the power of all three-phase users in the phase from the split-phase active power of the distribution outlet, and finally obtain the time sequence power matrix of the single-phase user. The time sequence power matrix represents a matrix of power data collected and stored in time sequence, and the single-phase user power matrix and the split-phase power matrix of the distribution outlet are both time sequence power matrices.

[0058] The time sequence power matrix represents a matrix of power data collected and stored in time sequence, and the single-phase user power matrix and the distribution transformer outlet phase power matrix are time sequence power matrices.

[0059] It should be noted that the first split-phase active power refers to the distribution transformer outlet split-phase active power, which is the active power output by each phase of the low-voltage side outlet end of the distribution transformer. The second split-phase active power is the low-voltage user split-phase active power, which refers to the active power consumed by each phase of the low-voltage user side load.

[0060] S102, map the single-phase user active power and the connection phase corresponding to the single-phase user active power to obtain an adjacency relationship matrix, configure a prior probability according to the adjacency relationship matrix to obtain an initial prior probability matrix.

[0061] In this embodiment, since the single-phase user and the connected phase have a binary unique mapping relationship, the adjacency relationship matrix of the single-phase user is as follows:

[0062]

[0063] In the formula, , , respectively represent the connection relationship between the user numbered n and phases A, B and C. If it belongs to a certain phase, the corresponding value is 1, otherwise it is 0.

[0064] Then the iterative update solution of the matrix is solved. And for any n∈{1,2,...,N}, it has:

[0065]

[0066] In the formula, represents the connection relationship between the user numbered n and phases A, B and C.

[0067] Further, the prior probability is configured according to the adjacency relationship matrix to obtain an initial prior probability matrix. Specifically, the Bayesian inference is a method of updating the confidence degree related to a certain hypothesis according to new evidence by using the Bayesian theorem, and the expression is:

[0068]

[0069] wherein, and are the independent probabilities of event a and event b, and are the conditional probabilities that a assumes b is true and b assumes a is true, respectively.

[0070] In combination with the application scenario, the above expression can be rewritten as:

[0071]

[0072] wherein, represents the hypothesis , is new evidence, is the likelihood probability, representing the compatibility of the new evidence supporting the hypothesis, is called the prior probability, which represents the current probability that is true before seeing the new evidence , is the marginal probability, i.e. and the sum of the joint probabilities of all hypotheses, is the posterior probability, which is the probability of the hypothesis under the new evidence.

[0073] Through Bayesian inference, the certainty of a user related to a certain phase can be updated. Let represent the hypothesis of each single-phase user and the phase thereof, then the prior probability thereof is represented as ), and the integration is in a matrix structure:

[0074]

[0075] In the formula, is the prior probability representing the hypothesis of each single-phase user and the phase A, wherein, is the prior probability representing the hypothesis of each single-phase user and the phase B, wherein, is the prior probability representing the hypothesis of each single-phase user and the phase C.

[0076] The prior probability matrix is initialized and valued according to the active power of the single-phase user. Since the prior probability allows the existing knowledge to be added before calculation, the calculation result is more reliable, therefore, for the single-phase user with recorded phase, a higher prior probability of the corresponding hypothesis can be given, and the experience value 0.7 is usually taken, and the probabilities of the remaining two phases are both 0.15, and the prior probability of the single-phase user with other unrecorded phase = 1 / 3.

[0077] S103, the active power of the single-phase user in the starting day in the preset time period is taken as the to-be-calculated power, the likelihood probability of the single-phase user corresponding to the to-be-calculated power and all assumed phases is calculated, a plurality of likelihood probability values are obtained, based on each likelihood probability value and the initial prior probability matrix, the posterior probability of the to-be-calculated power and all assumed phases is calculated, and a plurality of posterior probability values are obtained.

[0078] In the embodiment, the prior probability of the correlation between each household in the area and the phase is initialized, the first-day power data is taken as the input, the Bayesian inference principle is used to calculate the likelihood probability and the posterior probability of each single-phase user and all the assumed phases, and then the assumed phase corresponding to the maximum posterior probability is taken as the phase of the user, so that the first iteration is achieved.

[0079] In an embodiment, the likelihood probability of the single-phase user corresponding to the to-be-calculated power and all the assumed phases is calculated, and a plurality of likelihood probability values are obtained, including:

[0080] The curve correlation between the to-be-calculated power and the phase power of the distribution transformer corresponding to all the assumed phases is calculated, and a Pearson correlation coefficient is obtained.

[0081] The Pearson correlation coefficient is scaled by using a softplus function, and a likelihood function calculation formula is obtained. According to the likelihood function calculation formula, the likelihood probability of the to-be-calculated power and all the assumed phases is calculated, and a plurality of likelihood probability values are obtained.

[0082] In the embodiment, the Pearson correlation coefficient is a commonly used index for measuring the similarity of two curves, and the curve correlation between the power of each single-phase user and the phase power of the distribution transformer corresponding to the assumed phase can be calculated based on the method, as shown in FIG. 2. Figure 4 Figure 4 The corresponding relationship between the results of different correlation coefficients of the user and the phase power of the distribution transformer and the likelihood probability is shown. As can be seen from the figure, when the function is taken as the likelihood probability, a lower likelihood probability can be given to the result with smaller or negative correlation, so as to avoid 0. For medium and high correlation, the likelihood probability changes nearly linearly, which helps to enhance the Bayesian inference.

[0083] Since the coefficient takes a value in [-1, 1], the value range is scaled to [0, 1] by using the softplus function, and a likelihood function calculation formula is established:

[0084]

[0085] wherein, is the likelihood probability, corr is , The mapping curve of the Pearson correlation coefficient of the two groups of power vectors and the likelihood probability is shown in FIG. 3. Figure 3

[0086] In an embodiment, the calculation formula of the posterior probability value is:

[0087]

[0088] wherein, is the active power of the single-phase user, is the prior probability matrix, ​​is the likelihood probability.

[0089] In this embodiment, after the likelihood probability of each hypothesis is calculated by taking the first day power data as input in the first iteration, the posterior probability of the hypothesis is calculated according to the Bayesian inference and the prior probability result, and the calculation formula is:

[0090]

[0091] In the formula, is the single-phase user active power, is the prior probability matrix, is the likelihood probability.

[0092] Let the maximum posterior probability of each user correspond to the hypothesis phase as the actual phase of the user-related adjacency matrix, and complete a round of iteration, and the calculation process is:

[0093]

[0094] In the formula, is the single-phase user active power, is the prior probability matrix, is the , , is the likelihood probability,

[0095] S104, select the hypothesis phase corresponding to the maximum posterior probability value as the phase of the single-phase user, and update the initial prior probability matrix according to each posterior probability value to obtain an updated prior probability matrix. After determining the single-phase user active power in the next day of the starting day as the new to-be-calculated power, go to the step of calculating the likelihood probability of the single-phase user corresponding to the to-be-calculated power and all hypothesis phases, and use the updated prior probability matrix and the new to-be-calculated power to identify the phase until the phase of all single-phase user active powers in a preset time period is identified and the updating is stopped. Output the final identification result.

[0096] In this embodiment, the hypothesis corresponding to the maximum posterior probability is taken as the phase of the user to achieve the first round of iteration; the posterior probability of the current iteration is taken as the prior probability of the next iteration, and the user hypothesis phase corresponding to the highest probability meeting the threshold in the current round is recorded.

[0097] Then, taking the power data of the next day as input, the power of the user with the recorded phase is removed from the transformer area side, and the likelihood probability and the posterior probability of each single-phase user and all hypothesis phases are calculated again, and the hypothesis corresponding to the maximum probability is taken as the phase of the user to complete this round of iteration. Finally, repeat this iteration process until the power data of the latest date, and output the final identification result, i.e., the corrected phase account.

[0098] In an embodiment, after selecting the hypothesis phase corresponding to the maximum posterior probability value as the phase of the single-phase user, the method further comprises:

[0099] Storing the phases of the to-be-calculated power into a matrix to obtain a first matrix;

[0100] Constructing a new adjacency matrix, and mapping the hypothesis phases in the first matrix in the adjacency relationship matrix to the new adjacency matrix.

[0101] In the embodiment, the hypothesis phase of each single-phase user and the posterior probability value thereof in the current iteration result are taken as the prior probability in the next iteration, that is:

[0102]

[0103] Then, the maximum posterior probability value of each user hypothesis after the current iteration is extracted:

[0104]

[0105] If the maximum posterior probability value exceeds the experience threshold , the hypothesis corresponding to the maximum posterior probability value is stored in the matrix which is automatically emptied in each iteration, and it is considered that the hypothesis has good credibility in the current iteration, that is:

[0106]

[0107] It should be noted that the first matrix refers to the matrix , and the threshold is 0.25 which is more appropriate through experiments.

[0108] In an embodiment, the single-phase user active power in the next day of the starting day is determined as the new to-be-calculated power, comprising:

[0109] Selecting the distribution transformer active power and the single-phase user active power in the next day of the starting day;

[0110] Obtaining the new to-be-calculated power according to the distribution transformer active power, the single-phase user active power, and the new adjacency matrix.

[0111] In the embodiment, since there are some single-phase users with relatively insignificant power fluctuations, the curve characteristics thereof are difficult to reflect from the distribution transformer outlet split-phase power, resulting in a small correlation coefficient when calculating the power curve fluctuation consistency, and the likelihood probability of the hypothesis thereof cannot be effectively calculated and accurately classified to the actual membership phase. Therefore, a new adjacency matrix is constructed first, so that the user hypothesis in the adjacency matrix is also stored in the element value in the matrix is mapped to the matrix :

[0112]

[0113] In the formula, is the element value in the new adjacency matrix .

[0114] After the power data of the next day is input, the matrix is removed from the corresponding phase on the side of the distribution transformer, so that the power fluctuation characteristics of the remaining users are more easily captured by the similarity calculation. In the next round of subsequent calculation of the likelihood probability, the input is , and the time sequence matrix composed of the new to-be-calculated power is:

[0115]

[0116] In the formula, g represents a single-phase user active time sequence power matrix composed of the single-phase user active power in a preset time period, is a new adjacency matrix, represents a distribution transformer outlet split-phase power matrix, represents a single-phase user active time sequence power matrix input in the next round.

[0117] Then, the likelihood probability and the posterior probability of all single-phase users in the next round are calculated. The corresponding phase of the single-phase user is obtained from the posterior probability, and the iteration process is repeated until no new date data appears. Finally, the adjacency matrix representing the relationship between the user and the actual phase is output. The result is compared with the original account, and the incorrect phase relationship data is corrected and the missing phase data is filled.

[0118] The low-voltage power distribution area single-phase user phase identification system provided by the embodiment of the application is as shown in Figure 5 . Figure 5 is a system block diagram of the low-voltage power distribution area single-phase user phase identification system 500, comprising:

[0119] The acquisition module 501 is configured to acquire single-phase user active power of a low-voltage power distribution area in a preset time period.

[0120] The mapping module 502 is configured to map the single-phase user active power and the connection phase corresponding to the single-phase user active power to obtain an adjacency relationship matrix, configure a prior probability according to the adjacency relationship matrix, and obtain an initial prior probability matrix.

[0121] The first calculation module 503 is configured to take the single-phase user active power in the starting day in the preset time period as to-be-calculated power, calculate the likelihood probability of the to-be-calculated power and all assumed phases, obtain a plurality of likelihood probability values, calculate the posterior probability of the to-be-calculated power and all assumed phases based on the plurality of likelihood probability values and the initial prior probability matrix, and obtain a plurality of posterior probability values.

[0122] The second calculation module 504 is configured to select the hypothesis phase corresponding to the maximum posterior probability value as the phase of the single-phase user, and update the initial prior probability matrix according to each posterior probability value to obtain an updated prior probability matrix. After the active power of the single-phase user in the next day of the starting day is determined as the new to-be-calculated power, the step of calculating the likelihood probability of the single-phase user corresponding to the to-be-calculated power and all hypothesis phases is performed, and the updated prior probability matrix and the new to-be-calculated power are used for phase identification until the phases of all single-phase users in the preset time period are identified, the updating is stopped, and the final identification result is output.

[0123] In an embodiment, the acquisition module 501 includes a first acquisition unit and a second acquisition unit.

[0124] The first acquisition unit is configured to acquire the first split-phase active power and the second split-phase active power of the low-voltage power distribution area in the preset time period, wherein the second split-phase active power includes single-phase user split-phase active power and three-phase user split-phase active power.

[0125] The second acquisition unit is configured to clean the first split-phase active power based on the three-phase user split-phase active power to obtain the single-phase user active power.

[0126] The specific embodiments of the low-voltage power distribution area single-phase user phase identification system are basically the same as those of the above-mentioned low-voltage power distribution area single-phase user phase identification method, and will not be described here.

[0127] The technical features of the above embodiments can be combined in any manner. To make the description concise, all possible combinations of the technical features in the above embodiments are not described, however, as long as the combinations of the technical features do not exist contradictions, they should be considered as the scope of the present disclosure.

[0128] The above embodiments further illustrate the objectives, technical solutions, and advantages of the present disclosure. It should be understood that the above embodiments are merely specific embodiments of the present disclosure, and are not used to limit the protection scope of the present disclosure. It is particularly pointed out that any modification, equivalent replacement, improvement, etc. made by those skilled in the art within the spirit and principles of the present disclosure should be included in the protection scope of the present disclosure.

Claims

1. A low-voltage power distribution network single-phase user phase identification method, characterized in that, The method comprises the following steps: obtaining single-phase user active power of a low-voltage distribution area in a preset time period; mapping the single-phase user active power and the connection phase corresponding to the single-phase user active power to obtain an adjacency relationship matrix, and configuring a prior probability according to the adjacency relationship matrix to obtain an initial prior probability matrix; taking single-phase user active power in a starting day in the preset time period as to-be-calculated power, calculating likelihood probabilities of a single-phase user corresponding to the to-be-calculated power and all assumed phases to obtain a plurality of likelihood probability values, and calculating posterior probabilities of the to-be-calculated power and all the assumed phases based on each of the likelihood probability values and the initial prior probability matrix to obtain a plurality of posterior probability values; selecting the assumed phase corresponding to the maximum posterior probability value as the phase of the single-phase user, and updating the initial prior probability matrix according to each of the posterior probability values to obtain an updated prior probability matrix; after determining single-phase user active power in the next day of the starting day as new to-be-calculated power, proceeding to the step of calculating the likelihood probabilities of the single-phase user corresponding to the to-be-calculated power and all the assumed phases, and performing phase identification by using the updated prior probability matrix and the new to-be-calculated power until the phases of all single-phase user active powers in the preset time period are identified and the updating is stopped, and outputting a final identification result, wherein the obtaining of the updated prior probability matrix comprises taking the posterior probability value obtained in this iteration as the updated prior probability matrix; after the assumed phase corresponding to the maximum posterior probability value is selected as the phase of the single-phase user, the method further comprises the following steps: storing the phase of the to-be-calculated power in a matrix to obtain a first matrix; constructing a new adjacency matrix, and mapping the assumed phases in the first matrix in the adjacency relationship matrix to the new adjacency matrix.

2. The low voltage distribution area single phase consumer phase identification method as claimed in claim 1, wherein, The obtaining of the single-phase user active power of the low-voltage distribution area in the preset time period comprises the following steps: obtaining first split-phase active power and second split-phase active power of a low-voltage distribution area in a preset time period, wherein the second split-phase active power comprises single-phase user split-phase active power and three-phase user split-phase active power; based on the three-phase user split-phase active power, cleaning the first split-phase active power to obtain single-phase user active power.

3. The low voltage distribution area single phase consumer phase identification method as claimed in claim 1, wherein, The calculating of the likelihood probabilities of the single-phase user corresponding to the to-be-calculated power and all the assumed phases to obtain a plurality of likelihood probability values comprises the following steps: calculating a curve correlation between the to-be-calculated power and the phase power of all the assumed phases to obtain a Pearson correlation coefficient; scaling the Pearson correlation coefficient by using a softplus function to obtain a likelihood function calculation formula, and calculating the likelihood probabilities of the to-be-calculated power and all the assumed phases according to the likelihood function calculation formula to obtain a plurality of likelihood probability values.

4. The low voltage distribution area single phase consumer phase identification method as claimed in claim 1, wherein, The calculation formula of the posterior probability value is: wherein P is the single-phase user active power, P is the prior probability matrix, P is the likelihood probability value.

5. The low voltage distribution area single phase consumer phase identification method as claimed in claim 4, wherein, The determination of the single-phase user active power in the next day of the starting day as new to-be-calculated power comprises the following steps: selecting distribution transformer active power and single-phase user active power in the next day of the starting day; According to the active power of the distribution, the active power of the single-phase user and the new adjacency matrix, a new to-be-calculated power is obtained.

6. A low voltage distribution area single phase consumer phase identification system characterized in that, The method for identifying the phase of a single-phase user in a low-voltage distribution area according to any one of claims 1-5 comprises: An acquisition module is configured to acquire active power of single-phase users in a low-voltage distribution area within a preset time period; A mapping module is configured to map the active power of the single-phase users and connection phases corresponding to the active power of the single-phase users to obtain an adjacency relationship matrix, configure a prior probability according to the adjacency relationship matrix, and obtain an initial prior probability matrix; A first calculation module is configured to take the active power of single-phase users on a starting day in the preset time period as to-be-calculated power, calculate likelihood probabilities of the to-be-calculated power corresponding to single-phase users and all hypothetical phases to obtain a plurality of likelihood probability values, calculate posterior probabilities of the to-be-calculated power and all the hypothetical phases based on each of the likelihood probability values and the initial prior probability matrix, and obtain a plurality of posterior probability values; A second calculation module is configured to select a hypothetical phase corresponding to a maximum posterior probability value as a phase of the single-phase user, update the initial prior probability matrix according to each of the posterior probability values to obtain an updated prior probability matrix, determine active power of single-phase users on a next day of the starting day as new to-be-calculated power, and then proceed to the step of calculating likelihood probabilities of the new to-be-calculated power corresponding to the single-phase users and all the hypothetical phases, and perform phase identification using the updated prior probability matrix and the new to-be-calculated power until phase identification of all the single-phase user active power within the preset time period is completed, the updating is stopped, and a final identification result is output.

7. The low voltage distribution area single phase consumer phase identification system as claimed in claim 6, wherein, The acquisition module comprises a first acquisition unit and a second acquisition unit: The first acquisition unit is configured to acquire first and second split-phase active power within a preset time period in a low-voltage distribution area, wherein the second split-phase active power comprises single-phase user split-phase active power and three-phase user split-phase active power; The second acquisition unit is configured to clean the first split-phase active power based on the three-phase user split-phase active power to obtain single-phase user active power.

8. A computer device, comprising: It comprises: A memory is configured to store a computer program; A processor is configured to execute the computer program to implement the method for identifying the phase of a single-phase user in a low-voltage distribution area according to any one of claims 1-5.

9. A storage medium, characterized by The storage medium has a computer program stored thereon, and the computer program is executed by the processor to implement the steps of the method for identifying the phase of a single-phase user in a low-voltage distribution area according to any one of claims 1-5.

Images