CN116454856A - SC-DTW algorithm-based low-voltage distribution network user variable topology identification method - Google Patents

SC-DTW algorithm-based low-voltage distribution network user variable topology identification method Download PDF

Info

Publication number
CN116454856A
CN116454856A CN202211226698.XA CN202211226698A CN116454856A CN 116454856 A CN116454856 A CN 116454856A CN 202211226698 A CN202211226698 A CN 202211226698A CN 116454856 A CN116454856 A CN 116454856A
Authority
CN
China
Prior art keywords
daily
voltage
low
distribution network
voltage time
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211226698.XA
Other languages
Chinese (zh)
Inventor
谈竹奎
刘斌
程庆
殷子皓
许文强
张秋雁
朱勇
徐玉韬
欧家祥
肖小兵
王冕
唐赛秋
聂沧禹
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Guizhou Power Grid Co Ltd
Original Assignee
Guizhou Power Grid Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Guizhou Power Grid Co Ltd filed Critical Guizhou Power Grid Co Ltd
Priority to CN202211226698.XA priority Critical patent/CN116454856A/en
Publication of CN116454856A publication Critical patent/CN116454856A/en
Pending legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
    • H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
    • H02J3/00—Circuit arrangements for ac mains or ac distribution networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Electricity, gas or water supply

Abstract

The invention discloses a low-voltage distribution network household transformer topology identification method based on an SC-DTW algorithm, which comprises the steps of obtaining a secondary side three-phase of a low-voltage distribution transformer of a to-be-identified platform area and a platform area daily voltage time sequence cluster formed by all user sides measured by a platform area AMI system; performing spatial polar log-coordinate histogram calculation to obtain a shape context representation at each point; calculating a shape context cost matrix among the daily voltage time sequences and replacing an Euclidean distance matrix used in a Dynamic Time Warping (DTW) algorithm; accumulating Euclidean distance between each pair of points according to the generated regular paths to obtain an accumulated distance matrix between each daily voltage time sequence; and clustering by using a k-means algorithm to obtain labels of the phases of all users by taking an accumulated distance matrix among the daily voltage time sequences as a measure of the similarity among the sequences, so as to identify the user transformer topology of the low-voltage distribution transformer area. The installation and maintenance cost is saved, and the feasibility is better. The alignment effect is better, and the accuracy of the user clustering labels is improved.

Description

SC-DTW algorithm-based low-voltage distribution network user variable topology identification method
Technical Field
The invention relates to the technical field of low-voltage distribution networks, in particular to a low-voltage distribution network household transformer topology identification method based on an SC-DTW algorithm.
Background
With the development of power grid state estimation technology, the power system topology structure analysis method is widely paid attention to by experts and scholars, the traditional power system topology structure analysis method generally expresses the topology structure as a linked list relationship, and the connectivity of nodes is analyzed by using search technologies in graph theory, such as a depth-first search method and a breadth-first search method. This approach generally requires the creation of a linked list that reflects the topology, and the topology analysis is implemented by processing the linked list. In a modern power distribution network system, the structure and the topological relation of a medium-high voltage power distribution network are quite clear, but in a low-voltage power distribution network (LVDN) link, as the asset distribution of a plurality of cells is unclear, even a plurality of rural power or urban villages do not have structural record information, the topological structure is ambiguous, so that the state estimation of a power grid is difficult to carry out, meanwhile, the topology is also the basis of line loss analysis and fault positioning, and the accuracy of the line loss analysis and the effectiveness of fault treatment are restricted by the wrong topological information. With the continuous promotion of smart grid construction, advanced measurement systems (AdvancedMeterInfrastructure, AMI) are increasingly configured in the power distribution network, and effective data information is provided for verification and correction of the topology relationship of the power distribution network.
The current LVDN topological relation identification method is mainly divided into 3 types, namely an injection signal method, a data tag method and a data analysis method. The signal injection method is to add signal equipment in the net rack, and judge the topological relation by injecting and reading signals, such as Chinese patent CN112468320A, CN110838758B; the data tag method is to add coded communicators to each level of equipment in the power distribution network and establish data tags to realize the self-identification of the equipment in the network, such as China patent CN112086965A, CN111600748A. The two methods have higher recognition accuracy, but auxiliary equipment is required to be added for recognition, and the problems of high cost, difficult operation and maintenance and the like exist. The data analysis method is based on the electrical quantity data such as voltage, current and the like acquired by the advanced measurement system, potential association relations among users are mined, and therefore topology relation identification is achieved. For example, chinese patent CN110768256B discloses a method, apparatus and system for identifying a topology of a transformer area based on a voltage harmonic spectrum, wherein the voltage harmonic spectrum distance between a transformer of the transformer area and a household table is compared to identify the topology of the transformer area, but the time-frequency converted voltage harmonic spectrum can only reflect the frequency domain characteristics of a voltage curve, resulting in loss of time sequence characteristic information. As further described in chinese patent CN111835006a, a low-voltage area topology identification method based on voltage curve and least square is disclosed, and area topology identification is completed by calculating voltage similarity and least square model related to active power. However, the least square model is calculated based on euclidean distance, which is a point-to-point distance calculation method, and the difference between the voltage curve morphology and the voltage level is coupled, so that the two differences cannot be analyzed independently.
Disclosure of Invention
This section is intended to outline some aspects of embodiments of the invention and to briefly introduce some preferred embodiments. Some simplifications or omissions may be made in this section as well as in the description summary and in the title of the application, to avoid obscuring the purpose of this section, the description summary and the title of the invention, which should not be used to limit the scope of the invention.
The present invention has been made in view of the above-mentioned and/or existing problems occurring in the low-voltage power distribution network user-variable topology identification method based on the SC-DTW algorithm.
Therefore, the problem to be solved by the invention is how to provide a low-voltage distribution network user transformer topology identification method based on an SC-DTW algorithm.
In order to solve the technical problems, the invention provides the following technical scheme: a low-voltage distribution network household transformer topology identification method based on an SC-DTW algorithm comprises the steps of obtaining a secondary side three-phase of a low-voltage distribution transformer of a to-be-identified area and an area daily voltage time sequence cluster jointly formed by all user sides measured by an area AMI system.
And (3) carrying out space polar logarithmic coordinate histogram calculation on each point in the district daily voltage time sequence cluster to obtain a shape context representation (SC) at each point.
And calculating a shape context cost matrix between each daily voltage time sequence.
The shape context cost matrix between each daily voltage time series is used to replace dynamic time warping Dynamic TimeWarping, namely Euclidean distance matrix used in the DTW algorithm, and a warping path which minimizes the accumulated shape context and meets the boundary condition, the monotone condition and the step length condition is calculated as the optimal alignment between each series.
And accumulating Euclidean distances between each pair of points according to the generated regular paths to obtain an accumulated distance matrix between each daily voltage time sequence.
And clustering by using a k-means algorithm to obtain labels of the phases of all users by taking an accumulated distance matrix among the daily voltage time sequences as a measure of the similarity among the sequences, so as to identify the user transformer topology of the low-voltage distribution transformer area.
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the daily voltage time sequence cluster for acquiring the to-be-identified area comprises a low-voltage distribution transformer secondary side three-phase daily voltage time sequence with a 3-dimensional vector of length L and each user side daily voltage time sequence with an N-dimensional vector of length L, which is metered by an area AMI system, wherein N is the number of users metered by the area AMI system, and L is the number of data points contained in each daily voltage time sequence.
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the daily voltage time sequence cluster of the area to be identified is a matrix of (N+3) x L.
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the step of obtaining the spatial polar log coordinates includes,
equally dividing the whole space into 12 directions in angle by taking the data point as the center;
according toDivided into 5 layers along the radial direction, and the radius R of the innermost layer inner Taking the minimum interval delta t of the solar voltage time sequence and the radius R of the outermost layer outer R is taken inner 2 of (2) 5 =32 times;
the stretch/shrink factor sf=73, the entire space is divided into 60 cells.
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the shape context is a spatial polar logarithmic coordinate histogram of each point on each daily voltage time sequence obtained by calculating the number of points falling into each small lattice.
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the shape context cost matrix C of the time series of solar voltages p and q is expressed by the following formula,
wherein C (p) i ,q j ) Representing a match p i And q j Cost of p i And q j Respectively represent the ith and jth points, h on the p-th and q-th daily voltage time sequences i (k) And h j (k) Respectively represent p i And q j In the shape context, k=60, K being the number of divisions of the lattice.
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the euclidean distance matrix used in the dynamic time warping DTW algorithm is replaced with a shape context cost matrix between each daily voltage time series, calculated to minimize the cumulative shape context, expressed by the following formula,
wherein p and q represent the p-th and q-th time series of daily voltages, w k Representing the value of the kth element on the regular path.
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the cumulative distance matrix between the time series of daily voltages is expressed by the following formula,
D(i,j)＝DTW(p(1:i),q(1:j))
where D represents the cumulative distance matrix.
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the step of clustering by using k-means algorithm to obtain labels of the phases of the users by taking the accumulated distance matrix D between the daily voltage time sequences as a measure of the similarity between the sequences comprises,
randomly selecting k sequences from a daily voltage time sequence cluster V of the platform area as initial centroids;
calculating each sequence V in the sequence cluster V i And centroid V μ Distance d of (2) ij ＝D(i end ,j end )；
Will V i The minimum mark is d iμ The corresponding category lambda i Update V λi ＝V λi ∪{V i Re-computing centroid
As a preferable scheme of the SC-DTW algorithm-based low-voltage distribution network user variable topology identification method, the invention comprises the following steps: the step of clustering with the k-means algorithm to obtain labels of the phases to which each user belongs further comprises,
iterating until all centroids are unchanged or the maximum iteration times are reached, stopping iterating and obtaining labels of all the users;
and identifying the household transformer topology of the low-voltage distribution transformer area.
Compared with the technology based on signal injection and data labeling, the method has the advantages that the method is calculated based on the measurement data of the existing AMI system, does not need to be additionally provided with extra hardware equipment, saves the installation and operation cost, and has more feasibility in engineering; compared with the classical DTW algorithm, the SC-DTW algorithm introduces morphological characteristics into the measurement of the similarity of the voltage time sequence, and the Euclidean distance matrix used by the classical DTW algorithm is replaced by the shape context cost matrix of the data points, so that the optimal alignment avoids the pathological alignment caused by using the Euclidean distance, the alignment effect is obviously better, and the accuracy of the user clustering label is further improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein:
FIG. 1 is a diagram of a shape context outline region;
FIG. 2 is a graph of the optimal normal path and corresponding optimal alignment using a classical DTW algorithm solution;
FIG. 3 is a diagram of an optimal regular path and corresponding optimal alignment solved using the SC-DTW algorithm proposed by the present invention;
fig. 4 is a result of a topology relationship of a certain low-voltage power distribution network user identified by using the SC-DTW algorithm provided by the invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will become more readily apparent, a more particular description of the invention will be rendered by reference to specific embodiments thereof which are illustrated in the appended drawings.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, but the present invention may be practiced in other ways other than those described herein, and persons skilled in the art will readily appreciate that the present invention is not limited to the specific embodiments disclosed below.
Further, reference herein to "one embodiment" or "an embodiment" means that a particular feature, structure, or characteristic can be included in at least one implementation of the invention. The appearances of the phrase "in one embodiment" in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments.
Example 1
Referring to fig. 1 to fig. 4, in a first embodiment of the present invention, the embodiment provides a low voltage power distribution network user transformer topology identification method based on SC-DTW algorithm, including:
s1: acquiring a transformer area daily voltage time sequence cluster consisting of a low-voltage distribution transformer secondary side three phase of a transformer area to be identified and each user side metered by an AMI system of the transformer area;
further, a daily voltage time sequence cluster V of the to-be-identified area is obtained, and the cluster V consists of two parts, namely a low-voltage distribution transformer secondary side three-phase daily voltage time sequence, which is expressed by the following formula:
secondly, the daily voltage time sequence of each user side measured by the platform area AMI system is expressed by the following formula:
is an nxl matrix, so that the daily voltage time series cluster of the to-be-identified area is a (3+n) ×l matrix, where N is the number of users that the area AMI system measures and includes, and L is the number of data points that each daily voltage time series includes.
S2: carrying out space polar logarithmic coordinate histogram calculation on each point in the daily voltage time sequence cluster of the platform area to obtain a Shape Context (SC) representation at each point;
further, each point in the daily voltage time series cluster of the area obtained in the step S1 is subjected to spatial polar logarithmic coordinate histogram calculation to obtain a shape context representation at each point. The specific method is that the whole space centered by the data point is divided into 12 directions in an angle and then is divided into two directions according to the following wayDivided into 5 layers along the radial direction, the radius Rinner of the outermost layer takes the minimum interval delta t of the solar voltage time sequence, and the radius R of the outermost layer outer Taking out Rinner 2 of (2) 5 =32 times. For time series, the x-axis represents time scale, the y-axis represents voltage, the x-and y-coordinates have different units and different orders of magnitude, whereas the shape context uses a circular (isotropic) neighborhood, so the y-coordinate must be linearly stretched or contracted to ensure that the y-coordinate is comparable to the x-coordinate, as shown in the following formula:
wherein V is i Is the original voltage value of a certain sequence in a daily voltage time sequence cluster of a station area, and the denominator max { V ] i }-min{V i The function of } is to set V i The value of (1) maps to [0,1 ]]In between, SF is a stretch/shrink factor, and sf=73 is taken in the present invention, so that the dots can be uniformly dispersed in the shape context, neither too sparse nor too compact. The entire space is thus divided into 60 small lattices as shown in fig. 2. The spatial polar logarithmic coordinate histogram of each point on each daily voltage time series, i.e. the shape context, is obtained by calculating the number of points falling into each small grid.
S3: calculating a shape context cost matrix among the daily voltage time sequences;
and (3) calculating a shape context cost matrix C of the daily voltage time sequences p and q according to the shape context between the daily voltage time sequences obtained in the step S2. Since the shape context is a distribution represented by a histogram, i.e., both rows and columns are two-class disordered variables, χ will naturally be 2 The concept of statistics is migrated into a measure of the correlation of two sequences as shown in the following:
wherein C (p) i ,q j ) Representing a match p i And q j Cost of p i And q j Respectively represent the ith and jth points, h on the p-th and q-th daily voltage time sequences i (k) And h j (k) Respectively represent p i And q j In the shape context, K is the number of division of the lattices in step S2, and k=60.
S4: replacing dynamic time warping with a shape context cost matrix among the daily voltage time sequences, namely, a Euclidean distance matrix used in a DTW algorithm, and calculating a regular path which minimizes the accumulated shape context and meets boundary conditions, monotonic conditions and step length conditions to be used as the optimal alignment among the sequences;
s5: accumulating Euclidean distances between each pair of points according to the generated regular paths to obtain an accumulated distance matrix between each daily voltage time sequence;
s6: and clustering by using a k-means algorithm to obtain labels of the phases of all users by taking an accumulated distance matrix among the daily voltage time sequences as a measure of the similarity among the sequences, so as to identify the user transformer topology of the low-voltage distribution transformer area.
Example 2
Referring to fig. 1 to 4, a second embodiment of the present invention is different from the first embodiment in that: and also comprises
S4: replacing dynamic time warping with a shape context cost matrix among the daily voltage time sequences, namely, a Euclidean distance matrix used in a DTW algorithm, and calculating a regular path which minimizes the accumulated shape context and meets boundary conditions, monotonic conditions and step length conditions to be used as the optimal alignment among the sequences;
further, the euclidean distance matrix used in the DTW algorithm is replaced by the shape context cost matrix between each daily voltage time series obtained in the step S3, and a regular path which minimizes the cumulative shape context and satisfies the boundary condition, the monotonic condition and the step length condition is calculated as the optimal alignment between each series. The essence of the DTW algorithm is the optimization problem-finding the path that satisfies certain constraints and has the smallest total cost, as shown in the following equation:
constraint conditions
1) Boundary conditions: w (w) 1 = (1, 1) and w k ＝(m,n)
2) Continuity: if w k = (a, b) and w k-1 = (a ', b'), then a-a '. Ltoreq.1 and b-b'. Ltoreq.1 must be satisfied
3) Monotonicity: if w k-1 = (a ', b'), and w k = (a, b), then a-a '. Gtoreq.0 and b-b'. Gtoreq.0 must be satisfied
Wherein p and q represent the p-th and q-th time series of daily voltages, w, respectively k Representing the first on the regular pathThe value of k elements, w of classical DTW algorithm k The invention substitutes the Euclidean distance matrix C between the time series of each day voltage for the Euclidean distance matrix used in the DTW algorithm, namely w k ＝(i,j) k ＝C ij Then solving by a dynamic programming algorithm, wherein the following formula is shown:
where γ (i, j) is the cumulative distance, d (p) i ,q j ) Is p i And q j Distance of (C) ij . The optimal regular path and the corresponding optimal alignment obtained by using the classical DTW algorithm and the SC-DTW algorithm provided by the invention are respectively shown in the figures 2 and 3, so that compared with the optimal alignment obtained by using the classical DTW, the SC-DTW algorithm provided by the invention avoids the pathological alignment caused by using the Euclidean distance, and the alignment effect is obviously better.
S5: accumulating Euclidean distances between each pair of points according to the generated regular paths to obtain an accumulated distance matrix between each daily voltage time sequence;
further, the optimal alignment between every two solar voltage sequences obtained in the step S4 is performed, and the shape context cost between each pair of points is accumulated to obtain an accumulated distance matrix D between each solar voltage time sequence, as shown in the following formula:
D(i,j)＝DTW(p(1:i),q(1:j))
s6: taking an accumulated distance matrix among the daily voltage time sequences as a measure of similarity among the sequences, and clustering by using a k-means algorithm to obtain labels of the phases of all users, so that the user transformer topology of the low-voltage distribution transformer area is identified;
furthermore, the accumulated distance matrix D between the time series of each daily voltage obtained in the step S5 is used as a measure of the similarity between the series, and the k-means algorithm is used for clustering to obtain the labels of the phases of each user. Firstly, k sequences are randomly selected from a daily voltage time sequence cluster V of a platform area to be used as initial barycenters, and the phase sequence possibly belonging to a user isA. B, C, thus taking k=3, the initial centroid is V μ ＝{V 1 ,V 2 ,V 3 -a }; then each sequence V in the sequence cluster V is calculated by using the accumulated distance matrix D in the step S5 i And centroid V μ Distance of (2): d, d ij ＝D(i end ,j end ) V is set up i The minimum mark is d iμ The corresponding category lambda i Update V λi ＝V λi ∪{V i Re-computing centroidAnd iterating until all centroids are unchanged or the maximum iteration times are reached, stopping iterating and obtaining labels of all the phases of the users, thereby identifying the user variable topology of the low-voltage distribution transformer area.
In summary, through the steps, the user variable topology relationship of the low-voltage distribution network can be identified according to the daily voltage curve collected by the user ammeter, as shown in fig. 4, so that the power grid company can conveniently and timely and correctly master the topology connection relationship of the low-voltage distribution network when the low-voltage user generates the transformer area and the phase sequence is transferred, and the three-phase unbalance degree and the transformer area line loss calculation are more accurate.
It should be noted that the above embodiments are only for illustrating the technical solution of the present invention and not for limiting the same, and although the present invention has been described in detail with reference to the preferred embodiments, it should be understood by those skilled in the art that the technical solution of the present invention may be modified or substituted without departing from the spirit and scope of the technical solution of the present invention, which is intended to be covered in the scope of the claims of the present invention.

Claims (10)

1. A low-voltage distribution network user transformer topology identification method based on an SC-DTW algorithm is characterized by comprising the following steps of: comprising the steps of (a) a step of,
acquiring a transformer area daily voltage time sequence cluster consisting of a low-voltage distribution transformer secondary side three phase of a transformer area to be identified and each user side metered by an AMI system of the transformer area;
carrying out space polar logarithmic coordinate histogram calculation on each point in the daily voltage time sequence cluster of the platform area to obtain a Shape Context (SC) representation at each point;
calculating a shape context cost matrix among the daily voltage time sequences;
the method comprises the steps that a shape context cost matrix among each daily voltage time sequence is used for replacing dynamic time warping Dynamic TimeWarping, namely, euclidean distance matrix used in a DTW algorithm, and a warping path which minimizes the accumulated shape context and meets boundary conditions, monotonic conditions and step length conditions is calculated to be used as the optimal alignment among each sequence;
accumulating Euclidean distances between each pair of points according to the generated regular paths to obtain an accumulated distance matrix between each daily voltage time sequence;
and clustering by using a k-means algorithm to obtain labels of the phases of all users by taking an accumulated distance matrix among the daily voltage time sequences as a measure of the similarity among the sequences, so as to identify the user transformer topology of the low-voltage distribution transformer area.
2. The SC-DTW algorithm-based low-voltage distribution network user-variable topology identification method according to claim 1, wherein the method is characterized by comprising the following steps of: the daily voltage time sequence cluster for acquiring the to-be-identified area comprises a low-voltage distribution transformer secondary side three-phase daily voltage time sequence with a 3-dimensional vector of length L and each user side daily voltage time sequence with an N-dimensional vector of length L, which is metered by an area AMI system, wherein N is the number of users metered by the area AMI system, and L is the number of data points contained in each daily voltage time sequence.
3. The SC-DTW algorithm-based low-voltage distribution network user-variable topology identification method according to claim 2, wherein the method is characterized by comprising the following steps of: the daily voltage time sequence cluster of the area to be identified is a matrix of (N+3) x L.
4. The SC-DTW algorithm-based low-voltage distribution network user-variable topology identification method of claim 3, wherein the method comprises the following steps of: the step of obtaining the spatial polar log coordinates includes,
equally dividing the whole space into 12 directions in angle by taking the data point as the center;
according to log r 2 Divided into 5 layers along the radial direction, and the radius R of the innermost layer inner Taking the minimum interval delta t of the solar voltage time sequence and the radius R of the outermost layer outer R is taken inner 2 of (2) 5 =32 times;
the stretch/shrink factor sf=73, the entire space is divided into 60 cells.
5. The method for identifying the household transformer topology of the low-voltage power distribution network based on the SC-DTW algorithm as claimed in claim 4, wherein the method comprises the following steps: the shape context is a spatial polar logarithmic coordinate histogram of each point on each daily voltage time sequence obtained by calculating the number of points falling into each small lattice.
6. The method for identifying the household transformer topology of the low-voltage power distribution network based on the SC-DTW algorithm as claimed in claim 5, wherein the method comprises the following steps: the shape context cost matrix C of the time series of solar voltages p and q is expressed by the following formula,
wherein C (p) i ,q j ) Representing a match p i And q j Cost of p i And q j Respectively represent the ith and jth points, h on the p-th and q-th daily voltage time sequences i (k) And h j (k) Respectively represent p i And q j In the shape context, k=60, K being the number of divisions of the lattice.
7. The method for identifying the household transformer topology of the low-voltage power distribution network based on the SC-DTW algorithm as claimed in claim 6, wherein the method is characterized in that: the euclidean distance matrix used in the dynamic time warping DTW algorithm is replaced with a shape context cost matrix between each daily voltage time series, calculated to minimize the cumulative shape context, expressed by the following formula,
wherein p and q represent the p-th and q-th time series of daily voltages, w k Representing the value of the kth element on the regular path.
8. The SC-DTW algorithm-based low-voltage distribution network user-variable topology identification method of claim 7, wherein the method is characterized by comprising the following steps of: the cumulative distance matrix between the time series of daily voltages is expressed by the following formula,
D(i,j)＝DTW(p(1:i),q(1:j))
where D represents the cumulative distance matrix.
9. The SC-DTW algorithm-based low-voltage distribution network user-variable topology identification method of claim 8, wherein the method is characterized by comprising the following steps of: the step of clustering by using k-means algorithm to obtain labels of the phases of the users by taking the accumulated distance matrix D between the daily voltage time sequences as a measure of the similarity between the sequences comprises,
randomly selecting k sequences from a daily voltage time sequence cluster V of the platform area as initial centroids;
calculating each sequence V in the sequence cluster V i And centroid V μ Distance d of (2) ij ＝D(i end ,j end )；
Will V i The minimum mark is d iμ The corresponding category lambda i Update V λi ＝V λi ∪{V i Re-computing centroid
10. The SC-DTW algorithm-based low-voltage distribution network user-variable topology identification method of claim 9, wherein the method is characterized by comprising the following steps of: the step of clustering with the k-means algorithm to obtain labels of the phases to which each user belongs further comprises,
iterating until all centroids are unchanged or the maximum iteration times are reached, stopping iterating and obtaining labels of all the users;
and identifying the household transformer topology of the low-voltage distribution transformer area.
CN202211226698.XA 2022-10-09 2022-10-09 SC-DTW algorithm-based low-voltage distribution network user variable topology identification method Pending CN116454856A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202211226698.XA CN116454856A (en) 2022-10-09 2022-10-09 SC-DTW algorithm-based low-voltage distribution network user variable topology identification method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202211226698.XA CN116454856A (en) 2022-10-09 2022-10-09 SC-DTW algorithm-based low-voltage distribution network user variable topology identification method

Publications (1)

Publication Number Publication Date
CN116454856A true CN116454856A (en) 2023-07-18

Family

ID=87129034

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211226698.XA Pending CN116454856A (en) 2022-10-09 2022-10-09 SC-DTW algorithm-based low-voltage distribution network user variable topology identification method

Country Status (1)

Country Link
CN (1) CN116454856A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117148023A (en) * 2023-10-31 2023-12-01 威海海泰电子有限公司 Intelligent power adapter production detection method and system

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117148023A (en) * 2023-10-31 2023-12-01 威海海泰电子有限公司 Intelligent power adapter production detection method and system
CN117148023B (en) * 2023-10-31 2024-01-26 威海海泰电子有限公司 Intelligent power adapter production detection method and system

Similar Documents

Publication Publication Date Title
Ardakanian et al. On identification of distribution grids
Ferreira et al. Time series clustering via community detection in networks
CN110443302B (en) Load identification method based on feature fusion and deep learning and application thereof
McGovern et al. Identifying predictive multi-dimensional time series motifs: an application to severe weather prediction
CN105468677A (en) Log clustering method based on graph structure
CN106792514A (en) User's duty residence analysis method based on signaling data
Huang et al. Power quality disturbances classification using rotation forest and multi‐resolution fast S‐transform with data compression in time domain
CN104765768A (en) Mass face database rapid and accurate retrieval method
CN103728507A (en) Grid fault diagnosis method based on data mining
Wu et al. Online detection of events with low-quality synchrophasor measurements based on $ i $ Forest
CN104966161A (en) Electric energy quality recording data calculating analysis method based on Gaussian mixture model
CN110210723B (en) Electricity stealing identification method based on analytic hierarchy process and isolated forest
CN110444011B (en) Traffic flow peak identification method and device, electronic equipment and storage medium
CN116454856A (en) SC-DTW algorithm-based low-voltage distribution network user variable topology identification method
CN110263945A (en) Based on markovian Line Loss of Distribution Network System processing system
CN116108984A (en) Urban flow prediction method based on flow-POI causal relationship reasoning
CN107590604B (en) Coherent unit grouping method and system combining S transformation and 2DPCA
CN115618249A (en) Low-voltage power distribution station area phase identification method based on LargeVis dimension reduction and DBSCAN clustering
CN116757498A (en) Method, equipment and medium for pushing benefit-enterprise policy
CN114124721B (en) Topology automatic identification method based on power carrier
CN115544070A (en) Similarity query optimization method based on trajectory representation learning
CN114758257A (en) CART algorithm-based power transmission line degraded insulator identification method
CN104376139A (en) Graph isomorphism judging method based on spectrum analysis
Chen et al. State Monitoring and Fault Diagnosis of HVDC System via KNN Algorithm with Knowledge Graph: A Practical China Power Grid Case
CN114444589A (en) Harmonic pollution partitioning method based on kDBA clustering

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination