WO2024119919A1 - 一种多源配网监测数据的电网运行状态监测方法及系统 - Google Patents

一种多源配网监测数据的电网运行状态监测方法及系统 Download PDF

Info

Publication number
WO2024119919A1
WO2024119919A1 PCT/CN2023/116455 CN2023116455W WO2024119919A1 WO 2024119919 A1 WO2024119919 A1 WO 2024119919A1 CN 2023116455 W CN2023116455 W CN 2023116455W WO 2024119919 A1 WO2024119919 A1 WO 2024119919A1
Authority
WO
WIPO (PCT)
Prior art keywords
power supply
data
subgrid
grid
sub
Prior art date
Application number
PCT/CN2023/116455
Other languages
English (en)
French (fr)
Inventor
耿贞伟
李申章
杨天国
张贵鹏
沈宗云
范黎涛
周琦
Original Assignee
云南电网有限责任公司信息中心
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 云南电网有限责任公司信息中心 filed Critical 云南电网有限责任公司信息中心
Publication of WO2024119919A1 publication Critical patent/WO2024119919A1/zh

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
    • H02J13/00—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network
    • H02J13/00002—Circuit arrangements for providing remote indication of network conditions, e.g. an instantaneous record of the open or closed condition of each circuitbreaker in the network; Circuit arrangements for providing remote control of switching means in a power distribution network, e.g. switching in and out of current consumers by using a pulse code signal carried by the network characterised by monitoring
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
    • G06F17/10—Complex mathematical operations
    • G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
    • 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
    • G06Q10/20—Administration of product repair or maintenance
    • 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—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
    • G06Q50/06—Energy or water supply
    • 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
    • H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
    • H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
    • Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
    • Y04S10/00—Systems supporting electrical power generation, transmission or distribution
    • Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications

Definitions

  • the present invention relates to the field of smart grid technology, and in particular to a method and system for monitoring the operation status of a power grid using multi-source distribution network monitoring data.
  • the operation status of the power grid is complex, and the degree of data integration between the sub-grids of various parts is insufficient, making it difficult to conduct real-time status monitoring of the sub-grids in various regions and timely identify and handle abnormalities.
  • the different land attributes in various regions lead to different electricity consumption patterns in various regions. For example, the electricity consumption patterns in industrial areas and residential areas are different, making it difficult to detect abnormal conditions in time when multiple sub-grids are operating abnormally, resulting in further losses, making maintenance difficult and even requiring power outages for maintenance, causing inconvenience in electricity use and greater economic losses.
  • the present invention provides a method for monitoring the operation status of a power grid using multi-source distribution network monitoring data to solve the problem that abnormal operation of a sub-grid cannot be discovered in time, resulting in further loss and increased difficulty in maintenance.
  • the present invention provides the following technical solutions:
  • the present invention provides a method for monitoring power grid operation status using multi-source distribution network monitoring data, comprising:
  • the power supply law information and the power supply data matrix of the sub-grid to be tested are input into a power grid operation status monitoring model to determine the operation status of the sub-grid to be tested, thereby improving the convenience of maintenance.
  • the method comprises: determining the power supply regularity information according to the first historical power supply data of the subgrid to be tested and the second historical power supply data of the subgrid belonging to the same cluster as the subgrid to be tested, including:
  • the power supply regularity information is obtained according to the first distribution feature and the second distribution feature.
  • determining the first distribution characteristics of the load data of the sub-grid to be tested at multiple times includes: determining the load data at the same sampling time in multiple sampling periods;
  • the first candidate load data at multiple sampling moments are fitted to obtain the first distribution characteristics.
  • the load data at the same sampling time in multiple sampling periods are screened to obtain the first candidate load data corresponding to the sampling time, including:
  • the first candidate load data is screened according to the formula D 1 :m- ⁇ 1 ⁇ l j ⁇ m+ ⁇ 1 ⁇ , wherein D 1 is a first screening condition function, m is a mean of the load data, ⁇ is a variance of the load data, and ⁇ 1 is a first preset multiple.
  • determining the second distribution characteristics of the load data of the subgrid belonging to the same cluster as the subgrid to be tested at multiple times includes:
  • a weighted sum is performed on the third distribution characteristics of each subgrid to obtain the second distribution characteristics.
  • performing weighted summation on the third distribution characteristics of each sub-grid to obtain the second distribution characteristics comprises:
  • an element gap matrix between each subgrid and the subgrid to be tested is obtained, wherein the element in the i-th row and j-th column of the element gap matrix is the difference between the i-th element in the first data sequence and the j-th element in the reference data sequence;
  • ⁇ k is the weight of the third distribution feature of the k-th subgrid
  • E k (i, i) is the element in the i-th row and i-th column in the element gap matrix between the k-th subgrid and the subgrid to be measured
  • L (i, j) represents the set of elements of the target path in multiple paths from the (1, 1) element to the (i, j) element in the element gap matrix, wherein the target path is the path with the smallest sum of elements experienced;
  • weighted summation is performed on the third distribution characteristics of the sub-grids to obtain the second distribution characteristics.
  • the power supply law information and the power supply data matrix of the sub-grid to be tested are input into the power grid operation status monitoring model to determine the operation status of the sub-grid to be tested, including:
  • the operating status of the subgrid to be tested is determined according to the statistical information of the outliers.
  • the present invention provides a power grid operation status monitoring system for multi-source distribution network monitoring data, comprising:
  • the sampling module is used to obtain multiple types of power supply data of multiple sub-grids at multiple sampling moments, and use the geographical information of the power supply of each sub-grid, the geographical information including regional location information and geographical Block attribute information;
  • a matrix module used to obtain a power supply data matrix of each sub-grid according to the geographic information and the power supply data
  • a clustering module used for clustering the power supply data matrix to determine the sub-grid to be tested with abnormal power supply data
  • a rule module used to determine power supply rule information according to first historical power supply data of the subgrid to be tested and second historical power supply data of a subgrid belonging to the same cluster as the subgrid to be tested;
  • the operation status module is used to input the power supply law information and the power supply data matrix of the sub-grid to be tested into the power grid operation status monitoring model to determine the operation status of the sub-grid to be tested.
  • the present invention provides a computing device, comprising:
  • the memory is used to store computer executable instructions
  • the processor is used to execute the computer executable instructions.
  • the steps of the power grid operation status monitoring method of multi-source distribution network monitoring data are implemented.
  • the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of the method for monitoring the operation status of a power grid based on multi-source distribution network monitoring data.
  • the present invention clusters based on geographic information and power supply data, and determines the subgrid to be tested by feature distance. It can be applied to screening the subgrid to be monitored among multiple subgrids with complex geographic information, improve the screening efficiency, and thus detect abnormal conditions even when the geographic information is complex.
  • the weight of each subgrid can be determined by the element gap matrix, and then the power supply law information can be determined to improve the accuracy of the power supply law information, so that the power supply law information and the power supply data matrix of the subgrid to be tested can be processed by using the power grid operation status monitoring model to determine the operation status of the subgrid to be tested, which can improve the accuracy of determining the operation status, help reduce economic losses, and improve the convenience of maintenance.
  • FIG1 is a diagram of a power grid operation status monitoring method for multi-source distribution network monitoring data provided by an embodiment of the present invention. Schematic diagram of the basic flow of the test method;
  • one embodiment or “embodiment” as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention.
  • the term “in one embodiment” that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.
  • install, connect, connect should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components.
  • install, connect, connect should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components.
  • an embodiment of the present invention provides a method for monitoring the operation status of a power grid based on multi-source distribution network monitoring data.
  • FIG1 shows a flow chart of the method for monitoring the operation status of a power grid based on multi-source distribution network monitoring data according to an embodiment of the present invention. As shown in FIG1 , the method includes:
  • S1 Obtain the power supply data of the subgrid and the geographic information of the subgrid power supply, and construct the power supply data matrix of the subgrid;
  • the collected power supply data may include power supply current, power supply voltage, rated load, actual load, etc.
  • the present invention does not limit the specific type of power supply data.
  • a variety of power supply data of each sub-grid at multiple sampling times can be collected as a basis for subsequent processing.
  • the geographic information may include regional location information of the area powered by the subgrid, such as the location of the boundary of the area, the location coordinates of important power supply equipment in the area, the outline shape of the area, the area of the area, etc., and may also include land attribute information, such as industrial land, residential land, agricultural land, etc.
  • the sampling time of the power supply data of a subgrid, the power supply data and the geographic information can be combined into a row vector, and multiple row vectors can constitute the power supply data matrix of the subgrid, that is, the power supply data matrix may include multiple sampling times for sampling the power supply data of a subgrid, the power supply data collected at multiple sampling times, and the geographic information.
  • sampling moments may be multiple sampling moments within a preset time period
  • the preset time period may be any preset time period, for example, within a day, a week, a month, a quarter or a year.
  • the present invention does not limit the specific length of the preset time period.
  • S2 clustering the power supply data matrix to determine the sub-grid to be tested with abnormal power supply data
  • power supply data and geographic information can be used as the basis for clustering.
  • multiple subgrids with high similarity in geographic information and similar power supply data can be clustered into one cluster.
  • the geographic information of the two subgrids is industrial land, and the area is similar.
  • the power supply data both have higher electricity consumption during the day on weekdays and lower electricity consumption at night and on non-working days, and the total electricity consumption is similar, so the two can be clustered into the same cluster.
  • the higher the similarity between the power supply data and geographic information of the two subgrids the closer the distance between the two in the feature space formed during clustering, and vice versa, the farther the distance.
  • the characteristic distance between the subgrid and other subgrids with the same geographic information is far.
  • the geographic information of multiple subgrids is industrial land, and the power consumption pattern is high during the daytime on weekdays and low on non-working days.
  • the electricity consumption is low during the day and night, while the electricity consumption pattern of a subgrid whose geographical information is industrial land is different from that of other subgrids.
  • the electricity consumption pattern of this subgrid is high regardless of day or night, so the characteristic distance between this subgrid and other subgrids is far.
  • the subgrids to be tested with abnormal power supply data can be screened based on the clustering results of the power supply data matrices of multiple subgrids.
  • the subgrids whose characteristic distance to the cluster core in the cluster cluster is greater than or equal to the preset distance can be selected as the subgrids to be tested.
  • S3 determining power supply regularity information according to first historical power supply data of the subgrid to be tested and second historical power supply data of a subgrid belonging to the same cluster as the subgrid to be tested;
  • the first distribution characteristics of the load data of the subgrid to be tested at multiple times are determined; based on the second historical power supply data, the second distribution characteristics of the load data of the subgrid belonging to the same cluster as the subgrid to be tested at multiple times are determined; based on the first distribution characteristics and the second distribution characteristics, the power supply law information is obtained.
  • the load data includes the total energy consumption of the subgrid to be measured, for example, the product of the total power supply current and the power supply voltage of the subgrid to be measured is determined as the load data.
  • the load data of the subgrid to be measured at multiple historical moments can be counted to determine its first distribution feature.
  • the first distribution characteristics of the load data of the subgrid to be tested at multiple times are determined, including: determining the load data at the same sampling time in multiple sampling cycles; screening the load data at the same sampling time in multiple sampling cycles to obtain the first alternative load data corresponding to the sampling time; fitting the first alternative load data at multiple sampling times to obtain the first distribution characteristics.
  • the sampling cycle can be a longer periodic time period, for example, one week, one month, etc.
  • the same sampling moment in multiple sampling cycles is the sampling moment with the same serial number in each sampling cycle.
  • the sampling cycle is one week, and the same sampling moment in multiple sampling cycles is 10:00 am every Monday in multiple weeks.
  • the preset time period is one month, the load data at 10:00 am in each cycle within one month can be obtained. For example, 4 or 5 load data can be obtained.
  • data at each sampling moment with the same serial number in multiple sampling cycles can be obtained. For example, if the preset time period is one month, the load data at 10:00 a.m. every Monday in a month, the load data at 11:00 a.m. every Monday, the load data at 12:00 noon every Monday... the load data at 10:00 noon every Monday... the load data at 10:00 a.m. every Sunday... That is, data of multiple sampling cycles can be obtained at each same sampling moment.
  • the load data of the multiple sampling periods corresponding to each same sampling moment may contain erroneous data. Therefore, some erroneous data can be excluded to improve the calculation accuracy.
  • the load data at the same sampling moment in multiple sampling periods are screened to obtain first candidate load data corresponding to the sampling moment, including:
  • the first candidate load data is screened: D 1 :m- ⁇ 1 ⁇ l j ⁇ m+ ⁇ 1 ⁇ (1)
  • D1 is the first screening condition function
  • m is the mean of the load data
  • ⁇ is the variance of the load data
  • ⁇ 1 is the first preset multiple
  • ⁇ 1 can be set to empirical data such as 3, so as to screen out outliers, that is, load data that exceeds the range set by the first filtering condition function D1 .
  • These outliers cannot reflect the normal level of the load data. Therefore, these outliers can be excluded, and the load data that does not exceed the range set by the first filtering condition function D1 can be used as the first alternative load data, thereby improving the calculation accuracy.
  • outliers can be excluded from the load data of multiple sampling periods corresponding to each same sampling moment, and the remaining first candidate load data can be fitted using the first candidate load data.
  • multiple first candidate data of each same sampling moment can be averaged, and the average value can be fitted according to time.
  • all first candidate data can also be directly used to fit according to time. After fitting, the fitting curve obtained is the first distribution feature.
  • the second distribution characteristics of the load data of other subgrids of the same cluster at multiple times can also be determined.
  • the process of determining the distribution characteristics of each subgrid is similar to the process of determining the first distribution characteristics of the subgrid to be tested. After determining the distribution characteristics of each subgrid, the distribution characteristics of multiple subgrids can be fused to obtain the second distribution characteristics.
  • the second distribution characteristics of the load data of the subgrid belonging to the same cluster as the subgrid to be tested at multiple times are determined, including: determining the load data of each subgrid at the same sampling time in multiple sampling cycles; screening the load data of each subgrid at the same sampling time in multiple sampling cycles to obtain second alternative load data of each subgrid corresponding to the sampling time; fitting the second alternative load data of each subgrid to obtain a third distribution characteristic of each subgrid; and performing weighted summation on the third distribution characteristics of each subgrid to obtain the second distribution characteristic.
  • the method of determining the third distribution characteristics of each subgrid is similar to the method of determining the first distribution characteristics of the subgrid to be measured, and will not be repeated here.
  • multiple third distribution characteristics can be fused, that is, weighted summation is performed to obtain the second distribution characteristics.
  • the weights can be set to be the same for each subgrid, or the weights of each subgrid can be determined based on the characteristic distance between each subgrid and the subgrid to be measured. For example, the farther the characteristic distance from the subgrid to be measured, the smaller the weight, and conversely, the closer the characteristic distance to the subgrid to be measured, the larger the weight.
  • performing weighted summation on the third distribution characteristics of each subgrid to obtain the second distribution characteristics includes: obtaining a first data sequence of load data of each subgrid according to load data of each subgrid at multiple times in multiple sampling periods;
  • an element gap matrix between each subgrid and the subgrid to be tested is obtained, wherein the element in the i-th row and j-th column of the element gap matrix is the difference between the i-th element in the first data sequence and the j-th element in the reference data sequence;
  • the weight of the third distribution characteristic of each sub-grid is determined:
  • ⁇ k is the weight of the third distribution feature of the kth subgrid
  • E k (i, i) is the element in the i-th row and i-th column in the element gap matrix between the kth subgrid and the subgrid to be measured
  • L (i, j) represents the set of elements of the target path in multiple paths from the (1, 1) element to the (i, j) element in the element gap matrix, wherein the target path is the path with the smallest sum of elements experienced;
  • weighted summation is performed on the third distribution characteristics of the sub-grids to obtain the second distribution characteristics.
  • the characteristic distance is usually determined by Euclidean distance or cosine distance.
  • the load data of each subgrid may not change at the same time, and there may be a time difference. For example, a factory may start work at 9:00 am, and another factory may start work at 8:30 am, then the load data of the two factories do not change at the same time.
  • the weight By determining the weight by Euclidean distance or predetermined distance, it may happen that the waveforms of the load data of the two subgrids are similar, but due to the time difference, the characteristic distance is farther and lower, and the weight is lower.
  • the load data of the two subgrids both include a peak and a trough, and the time difference between the peak and the trough of the load data of the two subgrids is Close, that is, the waveforms of the two sequences are similar, but due to the time difference in the changes of the load data of the two sub-grids, the trough of the load data of the first sub-grid and the peak of the load data of the second sub-grid appear close to each other, so the feature distance between the two is far and the weight is low.
  • an element gap matrix can be used to determine the minimum distance between the two, so as to solve the problem that the similarity between the two originally similar sub-grids is low and the feature distance is far due to the time difference.
  • the load data at multiple moments can be used to form a first data sequence.
  • the load data at multiple moments of the subgrid to be tested can be used to form a reference data sequence.
  • the element gap matrix is obtained using two data sequences, and the element in the i-th row and j-th column of the element gap matrix is the difference between the i-th element in the first data sequence and the j-th element in the reference data sequence.
  • each row and each column of the element gap matrix can be traversed to determine the minimum gap between the two data sequences.
  • the path between the element at the position (1, 1) in the upper left corner of the element gap matrix and the element at the position (n, n) in the lower right corner can traverse each row and each column of the element gap matrix, that is, any path from (1, 1) to (n, n) must pass through all rows and all columns.
  • at least one element in each row of the element gap matrix is included in the path, and at least one element in each column is included in the path.
  • the path will pass through the 1st row, the 2nd row... the nth row, and the path will also pass through the 1st column, the 2nd column... the nth column.
  • the path traverses each element in the first data sequence and each element in the reference data sequence.
  • the path having the smallest sum of elements traversed can be represented as the minimum distance between the first data sequence and the second data sequence.
  • the summation term in the numerator of formula (2) is the process of finding the path with the smallest sum of elements.
  • L(i, j) represents the set of elements of the target path in multiple paths from the (1, 1) element to the (i, j) element in the element gap matrix. Each time i and j increase, the minimum value of the three elements adjacent to (i, j) is added to the set to form the target path from the (1, 1) element to the (i, j) element, that is, the path with the smallest sum of elements experienced.
  • the sum of each element in the set, as well as the (1, 1) element and the (n, n) element can be obtained to obtain the sum of the elements of the target path in the path from the (1, 1) element to the (i, j) element, that is, the sum of the elements of the path with the smallest sum of elements experienced.
  • the denominator in formula (2) is the sum of the elements on the diagonal of the element gap matrix.
  • the diagonal is also the path from the (1, 1) element to the (i, j) element.
  • the elements traversed by this path are The sum is greater than or equal to the sum of the elements experienced by the target path.
  • the elements on the diagonal can represent the difference between the corresponding elements in the first data sequence and the reference data sequence. Therefore, the denominator of formula (2) is the sum of the differences of the corresponding elements.
  • the weight of the third distribution characteristic of the kth subgrid can be determined by formula (2).
  • the weight can be solved by formula (2), and the third distribution characteristics of each subgrid are weighted and summed using the weight of each subgrid to obtain the second distribution characteristic.
  • the element gap matrix can be used to traverse all load data in the first data sequence and the reference data sequence to determine the target path that minimizes the sum of the distances between the load data.
  • the weight of each subgrid can be determined by the element gap matrix distance.
  • the distance between all load data in the first data sequence and all load data in the reference data sequence can be used as a reference to reduce the problem of low weight accuracy caused by waveform offset due to time difference.
  • the power supply regularity information of the cluster can be determined.
  • the first distribution feature and the second distribution feature can be summed or weighted to obtain the power supply regularity information.
  • S4 Input the power supply law information and the power supply data matrix of the sub-grid to be tested into a power grid operation status monitoring model to determine the operation status of the sub-grid to be tested and improve the convenience of maintenance.
  • the power supply law information and the power supply data matrix of the sub-grid to be tested can be processed based on the power grid operation status monitoring model to determine whether the operation status of the sub-grid to be tested is normal.
  • outliers in the power supply data matrix of the sub-grid to be tested are determined; statistics are performed on the outliers to obtain a variety of statistical information of the outliers; and based on the statistical information of the outliers, the operating status of the sub-grid to be tested is determined.
  • outliers in the power supply data matrix are load data that deviate far from the power supply law information, for example, deviate by more than 50%, etc.
  • the present disclosure does not limit the degree of deviation.
  • statistics may be performed on the outliers to determine various statistical information of the outliers, for example, an average value of the deviation distances of the outliers may be determined, or the outliers may be fitted.
  • the outlier may just be the peak or trough value of power consumption. If the average value of the deviation distance is large, or the curve after the outlier is fitted is far from the power supply law information, then there may be an abnormality in the operation of the sub-grid, for example, there is abnormal power consumption, or the power supply efficiency is reduced, or there is component damage, leakage, etc. In this case, When the situation occurs, the sub-grid under test can be inspected and repaired to avoid further losses.
  • the above is a schematic scheme of a method for monitoring the operation status of a power grid using multi-source distribution network monitoring data in this embodiment.
  • the technical scheme of the power grid operation status monitoring system for accumulating multi-source distribution network monitoring data and the technical scheme of the above-mentioned method for monitoring the operation status of a power grid using multi-source distribution network monitoring data belong to the same concept.
  • This embodiment also provides a power grid operation status monitoring system for multi-source distribution network monitoring data, characterized by comprising:
  • a sampling module used to obtain multiple types of power supply data of multiple sub-grids at multiple sampling moments, and use geographic information of power supply of each sub-grid, wherein the geographic information includes regional location information and plot attribute information;
  • a matrix module used to obtain a power supply data matrix of each sub-grid according to the geographic information and the power supply data
  • a clustering module used for clustering the power supply data matrix to determine the sub-grid to be tested with abnormal power supply data
  • a rule module used to determine power supply rule information according to first historical power supply data of the subgrid to be tested and second historical power supply data of a subgrid belonging to the same cluster as the subgrid to be tested;
  • the operation status module is used to input the power supply law information and the power supply data matrix of the sub-grid to be tested into the power grid operation status monitoring model to determine the operation status of the sub-grid to be tested.
  • Memory used to store programs
  • a processor is used to load the program to execute the power grid operation status monitoring method of multi-source distribution network monitoring data.
  • This embodiment also provides a computer-readable storage medium storing a program, and when the program is executed by a processor, the power grid operation status monitoring method of multi-source distribution network monitoring data is implemented.
  • the storage medium proposed in this embodiment and the power grid operation status monitoring method of multi-source distribution network monitoring data proposed in the above embodiment belong to the same inventive concept.
  • the technical details not fully described in this embodiment can be referred to the above embodiment, and this embodiment has the same beneficial effects as the above embodiment.
  • the present invention can be implemented by means of software and necessary general hardware, and of course can also be implemented by hardware, but in many cases the former is a better implementation method.
  • the technical solution of the present invention, or the part that contributes to the prior art can be embodied in the form of a software product, which can be stored in a computer-readable storage medium, such as a computer's floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, server, or network device, etc.) to execute the methods of various embodiments of the present invention.
  • a computer-readable storage medium such as a computer's floppy disk, read-only memory (ROM), random access memory (RAM), flash memory (FLASH), hard disk or optical disk, etc.
  • an embodiment of the present invention provides a method for monitoring the operation status of a power grid using multi-source distribution network monitoring data. In order to verify its beneficial effects, comparison results of two solutions are provided.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Mathematical Optimization (AREA)
  • Pure & Applied Mathematics (AREA)
  • Mathematical Analysis (AREA)
  • Computational Mathematics (AREA)
  • Human Resources & Organizations (AREA)
  • Economics (AREA)
  • Software Systems (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Algebra (AREA)
  • Operations Research (AREA)
  • Strategic Management (AREA)
  • Marketing (AREA)
  • Databases & Information Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Evolutionary Biology (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Primary Health Care (AREA)
  • General Health & Medical Sciences (AREA)
  • Water Supply & Treatment (AREA)
  • Public Health (AREA)
  • Computing Systems (AREA)
  • Power Engineering (AREA)
  • Remote Monitoring And Control Of Power-Distribution Networks (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

本发明公开了一种多源配网监测数据的电网运行状态监测方法，包括：获取子电网供电数据和使用子电网供电的地理信息，构建子电网的供电数据矩阵；对所述供电数据矩阵进行聚类，确定供电数据异常的待测子电网；根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息；将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态；能够在地理信息复杂的情况下及时发现异常状况，减少经济损失，提升检修的便利性。

Description

一种多源配网监测数据的电网运行状态监测方法及系统 技术领域
本发明涉及智能电网技术领域，尤其涉及一种多源配网监测数据的电网运行状态监测方法及系统。
背景技术
在相关技术中，电网运行状态复杂，各个部分的子电网之间的数据融合程度不足，难以对各个地区的子电网进行实时的状态监测以及及时的异常识别和处理，并且，各个地区的地块属性不同，导致各个地区的用电规律互不相同，例如，工业区和居住区的用电规律不相同，从而难以在多个子电网运行异常时及时发现异常状况，导致损失进一步扩大，造成检修困难甚至需要停电维修，造成用电不便和较大的经济损失。
发明内容
本部分的目的在于概述本发明的实施例的一些方面以及简要介绍一些较佳实施例。在本部分以及本申请的说明书摘要和发明名称中可能会做些简化或省略以避免使本部分、说明书摘要和发明名称的目的模糊，而这种简化或省略不能用于限制本发明的范围。
鉴于上述现有存在的问题，提出了本发明。
因此，本发明提供了一种多源配网监测数据的电网运行状态监测方法解决无法及时发现子电网运行异常，导致损失进一步扩大，增加检修难度的问题。
为解决上述技术问题，本发明提供如下技术方案：
第一方面，本发明提供了一种多源配网监测数据的电网运行状态监测方法，包括：
获取子电网供电数据和使用子电网供电的地理信息，构建子电网的供电数据矩阵；
对所述供电数据矩阵进行聚类，确定供电数据异常的待测子电网；
根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息；
将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态，提升检修的便利性。
作为本发明所述的多源配网监测数据的电网运行状态监测方法的一种优 选方案，其中：根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息，包括：
根据所述第一历史供电数据，确定所述待测子电网在多个时刻的负载数据的第一分布特征；
根据所述第二历史供电数据，确定所述与所述待测子电网属于同一聚类簇的子电网在多个时刻的负载数据的第二分布特征；
根据所述第一分布特征和所述第二分布特征，获得所述供电规律信息。
作为本发明所述的多源配网监测数据的电网运行状态监测方法的一种优选方案，其中：根据所述第一历史供电数据，确定所述待测子电网在多个时刻的负载数据的第一分布特征，包括：确定多个采样周期中相同采样时刻的负载数据；
对多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的第一备选负载数据；
对多个采样时刻的第一备选负载数据进行拟合，获得所述第一分布特征。
作为本发明所述的多源配网监测数据的电网运行状态监测方法的一种优选方案，其中：对多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的第一备选负载数据，包括：
确定多个采样周期中相同采样时刻的负载数据的均值和方差；
根据公式D1：m-θ1σ≤lj≤m+θ1σ，筛选第一备选负载数据，其中，D1为第一筛选条件函数，m为负载数据的均值，σ为负载数据的方差，θ1为第一预设倍数。
作为本发明所述的多源配网监测数据的电网运行状态监测方法的一种优选方案，其中：根据所述第二历史供电数据，确定所述与所述待测子电网属于同一聚类簇的子电网在多个时刻的负载数据的第二分布特征，包括：
确定各子电网在多个采样周期中相同采样时刻的负载数据；
对各子电网在多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的各子电网的第二备选负载数据；
对各子电网的所述第二备选负载数据进行拟合，获得各子电网的第三分布特征；
对各子电网的第三分布特征进行加权求和，获得所述第二分布特征。
作为本发明所述的多源配网监测数据的电网运行状态监测方法的一种优选方案，其中：对各子电网的第三分布特征进行加权求和，获得所述第二分布特征，包括：
根据所述各子电网在多个采样周期中多个时刻的负载数据，获得各子电网的负载数据的第一数据序列；
根据所述待测子电网在在多个采样周期中多个时刻的负载数据，获得参考数据序列；
根据各子电网的第一数据序列和所述参考数据序列，获得各子电网与待测子电网之间的元素差距矩阵，其中，元素差距矩阵中第i行第j列的元素为第一数据序列中第i个元素与参考数据序列中第j个元素之差；
根据公式
确定所述各子电网的第三分布特征的权值，其中，αk为第k个子电网的第三分布特征的权值，Ek(i，i)为第k个子电网与待测子电网之间的元素差距矩阵中第i行第i列的元素，L(i，j)表示在所述元素差距矩阵中，从(1，1)元素至(i，j)元素的多个路径中的目标路径的元素的集合，其中，所述目标路径为经历的元素之和最小的路径；
根据所述各子电网的第三分布特征的权值，对各子电网的第三分布特征进行加权求和，获得所述第二分布特征。
作为本发明所述的多源配网监测数据的电网运行状态监测方法的一种优选方案，其中：将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态，包括：
根据所述供电规律信息，确定所述待测子电网的供电数据矩阵中的离群点；
对所述离群点进行统计，获得离群点的多种统计信息；
根据离群点的统计信息，确定所述待测子电网的运行状态。
第二方面，本发明提供了一种多源配网监测数据的电网运行状态监测系统，包括，
采样模块，用于在多个采样时刻，获取多个子电网的多个类型的供电数据，以及使用各子电网供电的地理信息，所述地理信息包括区域位置信息，以及地 块属性信息；
矩阵模块，用于根据所述地理信息和所述供电数据，获得各个子电网的供电数据矩阵；
聚类模块，用于对所述供电数据矩阵进行聚类，确定供电数据异常的待测子电网；
规律模块，用于根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息；
运行状态模块，用于将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态。
第三方面，本发明提供了一种计算设备，包括：
存储器和处理器；
所述存储器用于存储计算机可执行指令，所述处理器用于执行所述计算机可执行指令，该计算机可执行指令被处理器执行时实现所述多源配网监测数据的电网运行状态监测方法的步骤。
第四方面，本发明提供了一种计算机可读存储介质，其存储有计算机可执行指令，该计算机可执行指令被处理器执行时实现所述多源配网监测数据的电网运行状态监测方法的步骤。
本发明的有益效果：本发明基于地理信息和供电数据进行聚类，并通过特征距离确定待测子电网，可适用于在地理信息复杂的多个子电网中筛选需要监测的子电网，提高筛选效率，从而可在地理信息复杂的情况下即使发现异常状况。并可通过元素差距矩阵确定各子电网的权值，进而确定供电规律信息，提升供电规律信息的准确性，从而利用电网运行状态监测模型来对供电规律信息和待测子电网的供电数据矩阵进行处理，确定待测子电网的运行状态，可提升确定运行状态的准确性，有利于减少经济损失，提升检修的便利性。
附图说明
为了更清楚地说明本发明实施例的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动性的前提下，还可以根据这些附图获得其它的附图。其中：
图1为本发明一个实施例提供的一种多源配网监测数据的电网运行状态监 测方法的基本流程示意图；
具体实施方式
为使本发明的上述目的、特征和优点能够更加明显易懂，下面结合说明书附图对本发明的具体实施方式做详细的说明，显然所描述的实施例是本发明的一部分实施例，而不是全部实施例。基于本发明中的实施例，本领域普通人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本发明的保护的范围。
在下面的描述中阐述了很多具体细节以便于充分理解本发明，但是本发明还可以采用其他不同于在此描述的其它方式来实施，本领域技术人员可以在不违背本发明内涵的情况下做类似推广，因此本发明不受下面公开的具体实施例的限制。
其次，此处所称的“一个实施例”或“实施例”是指可包含于本发明至少一个实现方式中的特定特征、结构或特性。在本说明书中不同地方出现的“在一个实施例中”并非均指同一个实施例，也不是单独的或选择性的与其他实施例互相排斥的实施例。
本发明结合示意图进行详细描述，在详述本发明实施例时，为便于说明，表示器件结构的剖面图会不依一般比例作局部放大，而且所述示意图只是示例，其在此不应限制本发明保护的范围。此外，在实际制作中应包含长度、宽度及深度的三维空间尺寸。
同时在本发明的描述中，需要说明的是，术语中的“上、下、内和外”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本发明和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本发明的限制。此外，术语“第一、第二或第三”仅用于描述目的，而不能理解为指示或暗示相对重要性。
本发明中除非另有明确的规定和限定，术语“安装、相连、连接”应做广义理解，例如：可以是固定连接、可拆卸连接或一体式连接；同样可以是机械连接、电连接或直接连接，也可以通过中间媒介间接相连，也可以是两个元件内部的连通。对于本领域的普通技术人员而言，可以具体情况理解上述术语在本发明中的具体含义。
实施例1
参照图1，为本发明的一个实施例，提供了一种多源配网监测数据的电网运行状态监测方法，图1示出根据本发明实施例的多源配网监测数据的电网运行状态监测方法的流程图，如图1所示，所述方法包括：
S1：获取子电网供电数据和使用子电网供电的地理信息，构建子电网的供电数据矩阵；
更进一步的，采集到的所述供电数据可包括供电电流、供电电压、额定负载、实际负载等，本发明对供电数据的具体类型不做限制。可采集每个子电网在多个采样时刻的多种供电数据，作为后续处理的依据。
更进一步的，所述地理信息可包括使用子电网供电的区域的区域位置信息，例如，区域的边界的位置、区域内重要供电设备的位置坐标、区域的轮廓形状、区域的面积等，还可包括地块属性信息，例如，工业用地、居住用地、农业用地等。
更进一步的，可将某个子电网的供电数据的采样时刻、供电数据和地理信息组成行向量，多个行向量即可组成该子电网的供电数据矩阵，即，供电数据矩阵中可包括对某个子电网的供电数据进行采样的多个采样时刻，以及多个采样时刻采集到的供电数据，以及地理信息。
更进一步的，采样时刻可以是预设时间段内的多个采样时刻，预设时间段可以是预设的任意时间段，例如，一天内，一周内、一个月内、一个季度内或一年内，本发明对预设时间段的具体长度不做限制。
S2：对所述供电数据矩阵进行聚类，确定供电数据异常的待测子电网；
更进一步的，在聚类过程中，可将供电数据和地理信息作为聚类的依据，例如，地理信息相似度高，供电数据接近的多个子电网可聚类成为一个聚类簇。在示例中，两个子电网的地理信息均为工业用地，且面积接近，对供电数据进行分析，二者均为工作日白天用电量较高，夜晚和非工作日用电量较低，且用电总量接近，则二者可聚类至同一聚类簇中。并且，两个子电网的供电数据和地理信息的相似度越高，则二者在聚类时构成的特征空间中的距离越近，反之，则距离越远。
更进一步的，在聚类过程中，如果出现供电数据异常的子电网，则在该子电网与地理信息相同的其他子电网之间的特征距离较远。在示例中，多个子电网的地理信息均为工业用地，且用电规律均为工作日白天用电量高，非工作日 和夜晚用电量低，而某个地理信息为工业用地的子电网的用电规律与其他子电网有差异，例如，该子电网的用电规律为不论白天和黑夜，用电量均较高，则该子电网与其他子电网的特征距离较远。
更进一步的，可基于等多个子电网的供电数据矩阵的聚类结果来筛选供电数据异常的待测子电网。在示例中，可选择与聚类簇中的聚类核心的特征距离大于或等于预设距离的子电网作为待测子电网。
S3：根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息；
更进一步的，根据所述第一历史供电数据，确定所述待测子电网在多个时刻的负载数据的第一分布特征；根据所述第二历史供电数据，确定所述与所述待测子电网属于同一聚类簇的子电网在多个时刻的负载数据的第二分布特征；根据所述第一分布特征和所述第二分布特征，获得所述供电规律信息。
更进一步的，所述负载数据包括待测子电网的总能耗，例如，将待测子电网的总供电电流与供电电压的乘积确定为所述负载数据。可对待测子电网的多个历史时刻的负载数据进行统计，确定其第一分布特征。
更进一步的，根据所述第一历史供电数据，确定所述待测子电网在多个时刻的负载数据的第一分布特征，包括：确定多个采样周期中相同采样时刻的负载数据；对多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的第一备选负载数据；对多个采样时刻的第一备选负载数据进行拟合，获得所述第一分布特征。
更进一步的，所述采样周期可以是较长的周期性的时间段，例如，一周、一个月等，多个采样周期中的相同采样时刻为各采样周期中序号相同的采样时刻，例如，采样周期为一周，多个采样周期中相同的采样时刻为多个星期中的每个周一上午10:00，如果所述预设时间段为一个月，则可获得一个月内每个周期上午10:00的负载数据，例如，可获得4个或5个负载数据。
更进一步的，可获得多个采样周期中的每个序号相同的采样时刻的数据，例如，如果所述预设时间段为一个月，则可获得一个月内每个周一上午10:00的负载数据，每个周一上午11:00的负载数据，每个周一中午12:00的负载数据…每个周一中午10:00的负载数据…每个周日上午10:00的负载数据…即，每个相同的采样时刻均可获取多个采样周期的数据。
应说明的是，在上述与每个相同采样时刻对应的多个采样周期的负载数据中，可能含有错误数据，因此，可排除一些错误数据，提升运算准确性。
更进一步的，对多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的第一备选负载数据，包括：
确定多个采样周期中相同采样时刻的负载数据的均值和方差；
根据公式(1)，筛选第一备选负载数据：
D1：m-θ1σ≤lj≤m+θ1σ      (1)
其中，D1为第一筛选条件函数，m为负载数据的均值，σ为负载数据的方差，θ1为第一预设倍数。
应说明的是，在示例中，θ1可设置为3等经验数据，从而筛选出离群点，即，超出第一筛选条件函数D1设定的范围的负载数据，这些离群点不能反映负载数据的正常水平，因此，可将这些离群点排除，并将未超出第一筛选条件函数D1设定的范围的负载数据作为第一备选负载数据，从而提升运算准确性。
更进一步的，经过上述筛选处理，与每个相同采样时刻对应的多个采样周期的负载数据中，可排除离群点，剩余第一备选负载数据，可利用第一备选负载数据进行拟合。在示例中，可将每个相同采样时刻的多个第一备选数据进行求平均，并将平均值按照时间进行拟合。当然，也可直接利用所有第一备选数据按照时间进行拟合。在拟合后，获得的拟合曲线即为所述第一分布特征。
更进一步的，在确定聚类簇的供电规律信息时，除了待测子电网的第一分布特征外，还可确定相同聚类簇的其他子电网在多个时刻的负载数据的第二分布特征。对于每个子电网确定其分布特征的处理与上述确定待测子电网的第一分布特征的处理类似，在确定出每个子电网的分布特征后，可将多个子电网的分布特征进行融合，从而获得第二分布特征。
更进一步的，根据所述第二历史供电数据，确定所述与所述待测子电网属于同一聚类簇的子电网在多个时刻的负载数据的第二分布特征，包括：确定各子电网在多个采样周期中相同采样时刻的负载数据；对各子电网在多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的各子电网的第二备选负载数据；对各子电网的所述第二备选负载数据进行拟合，获得各子电网的第三分布特征；对各子电网的第三分布特征进行加权求和，获得所述第二分布特征。
更进一步的，如上所述，确定各子电网的第三分布特征的方式与以上确定待测子电网的第一分布特诊的方式类似，在此不再赘述。在获得每个子电网的第三分布特征后，可对多个第三分布特征进行融合，即，进行加权求和，获得第二分布特征。在示例中，可将权值设为各子电网相同，也可根据各子电网与待测子电网之间的特征距离确定个子电网的权值，例如，与待测子电网的特征距离越远，则权值越小，反之，与待测子电网的特征距离越近，则权值越大。
更进一步的，对各子电网的第三分布特征进行加权求和，获得所述第二分布特征，包括：根据所述各子电网在多个采样周期中多个时刻的负载数据，获得各子电网的负载数据的第一数据序列；
根据所述待测子电网在在多个采样周期中多个时刻的负载数据，获得参考数据序列；
根据各子电网的第一数据序列和所述参考数据序列，获得各子电网与待测子电网之间的元素差距矩阵，其中，元素差距矩阵中第i行第j列的元素为第一数据序列中第i个元素与参考数据序列中第j个元素之差；
根据公式(2)，确定所述各子电网的第三分布特征的权值：
其中，αk为第k个子电网的第三分布特征的权值，Ek(i，i)为第k个子电网与待测子电网之间的元素差距矩阵中第i行第i列的元素，L(i，j)表示在所述元素差距矩阵中，从(1，1)元素至(i，j)元素的多个路径中的目标路径的元素的集合，其中，所述目标路径为经历的元素之和最小的路径；
根据所述各子电网的第三分布特征的权值，对各子电网的第三分布特征进行加权求和，获得所述第二分布特征。
更进一步的，在如上所述的求解特征距离作为确定权值的依据时，通常通过欧氏距离或余弦距离来确定所述特征距离，然而，各子电网的负载数据可能不是同时变化，可能存在时间差，例如，某个工厂可能上午9:00开工，另一个工厂为上午8:30开工，则两个工厂的负载数据并非同时变化。通过欧式距离或预先距离确定权值，可能出现两个子电网的负载数据的波形类似，但由于时间差而导致特征距离较远较低，权值较低的情况。例如，两个子电网的负载数据均包括一个波峰和一个波谷，且两个子电网的负载数据的波峰和波谷的时间差 相近，即，两个序列的波形相似，但由于两个子电网的负载数据的变化存在时间差，导致第一个子电网的负载数据的波谷和第二个子电网的负载数据中的波峰出现的时刻接近，进而二者特征距离较远，权值较低。
更进一步的，针对上述问题，可采用元素差距矩阵来确定二者之间的最小距离，解决由于时间差导致原本相似的两个子电网之间相似度较低，特征距离较远的问题。
更进一步的，针对第k个子电网，可利用其多个时刻的负载数据组成第一数据序列。并且，利用待测子电网的多个时刻的负载数据组成参考数据序列。从而利用两个数据序列获取所述元素差距矩阵，元素差距矩阵中第i行第j列的元素为第一数据序列中第i个元素与参考数据序列中第j个元素之差。
更进一步的，可遍历元素差距矩阵的每一行和每一列，即可确定两个数据序列之间的最小差距。在示例中，元素差距矩阵左上角的(1，1)位置处的元素至右下角的(n，n)位置处的元素之间的路径，即可遍历元素差距矩阵的每一行和每一列，即，从(1，1)至(n，n)的任意路径，要经过所有行，也要经过所有列，换言之，元素差距矩阵中的每一行中均会至少有一个元素包括在路径中，每一列中至少有一个元素包括在路径中，该路径会途径第1行、第2行…第n行，该路径也会途径第1列、第2列…第n列。由于第i行第j列的元素为第一数据序列中第i个元素与参考数据序列中第j个元素之差，因此，该路径遍历了第一数据序列中的每个元素，也遍历了参考数据序列中的每个元素。在众多路径中，所经历的元素之和最小的路径，该路径经历的元素之和则可表示为第一数据序列和第二数据序列之间的最小距离。
更进一步的，公式(2)中分子部分的求和项即为寻找元素之和最小的路径的过程，L(i，j)表示在所述元素差距矩阵中，从(1，1)元素至(i，j)元素的多个路径中的目标路径的元素的集合，i和j每次增大，均在集合中新增与(i，j)相邻的三个元素中的最小值，构成从(1，1)元素至(i，j)元素的目标路径，即，经历的元素之和最小的路径。最终将集合中的各个元素，以及(1，1)元素和(n，n)元素进行求和，即可得到从(1，1)元素至(i，j)的路径中的目标路径的元素之和，即，经历的元素之和最小的路径的元素之和。
更进一步的，公式(2)中分母部分为元素差距矩阵的对角线上的元素之和，对角线同样是从(1，1)元素至(i，j)元素的路径，该路径经历的元素 之和大于或等于目标路径经历的元素之和。对角线上的元素可表示第一数据序列和参考数据序列中对应元素之差，因此，公式(2)的分母即为对应元素之差进行求和。
更进一步的，通过公式(2)可确定第k个子电网的第三分布特征的权值，针对各子电网，均可使用公式(2)求解权值，利用各子电网的权值对各子电网的第三分布特征进行加权求和，获得第二分布特征。
通过这种方式，可通过元素差距矩阵遍历第一数据序列与参考数据序列中所有负载数据，确定使各负载数据之间的距离之和最小的目标路径，通过元素差距矩阵距离确定各子电网的权值，可参考第一数据序列中所有负载数据和参考数据序列中所有负载数据之间的距离，减少由于时间差异造成波形偏移而导致权值准确度较低的问题。
更进一步的，在获得上述第一分布特征和第二分布特征后，则可确定该聚类簇的供电规律信息。例如，可对第一分布特征和第二分布特征进行求和或加权求和，获得所述供电规律信息。
S4：将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态，提升检修的便利性。
更进一步的，可基于电网运行状态监测模型对供电规律信息和待测子电网的供电数据矩阵进行处理，确定待测子电网的运行状态是否正常。
更进一步的，根据所述供电规律信息，确定所述待测子电网的供电数据矩阵中的离群点；对所述离群点进行统计，获得离群点的多种统计信息；根据离群点的统计信息，确定所述待测子电网的运行状态。
应说明的是，供电数据矩阵中的离群点为偏离供电规律信息较远的负载数据，例如，偏离50％以上等，本公开对偏离的程度不做限制。
更进一步的，可对离群点进行统计，确定离群点的多种统计信息，例如，确定离群点的偏离距离的平均值，或者对离群点进行拟合等。
更进一步的，如果偏离距离的平均值较小，或者离群点拟合后的曲线与供电规律信息接近，则子电网的运行状态不存在异常，离群点则可能只是用电量的波峰或波谷值等。如果偏离距离的平均值较大，或者离群点拟合后的曲线与供电规律信息差距较大，则子电网的运行状态可能存在异常，例如，存在异常的用电，或者供电效率下降，又或者存在元件损坏、漏电等现象，在出现这种 情况时，可对待测子电网进行检修，避免损失扩大。
上述为本实施例的一种多源配网监测数据的电网运行状态监测方法的示意性方案。需要说明的是，该累计多源配网监测数据的电网运行状态监测系统的技术方案与上述的累计多源配网监测数据的电网运行状态监测方法的技术方案属于同一构思，本实施例中多源配网监测数据的电网运行状态监测系统的技术方案未详细描述的细节内容，均可以参见上述多源配网监测数据的电网运行状态监测方法的技术方案的描述。
本实施例还提供一种多源配网监测数据的电网运行状态监测系统，其特征在于，包括：
采样模块，用于在多个采样时刻，获取多个子电网的多个类型的供电数据，以及使用各子电网供电的地理信息，所述地理信息包括区域位置信息，以及地块属性信息；
矩阵模块，用于根据所述地理信息和所述供电数据，获得各个子电网的供电数据矩阵；
聚类模块，用于对所述供电数据矩阵进行聚类，确定供电数据异常的待测子电网；
规律模块，用于根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息；
运行状态模块，用于将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态。
更进一步的，还包括：
存储器，用于存储程序；
处理器，用于加载所述程序以执行所述的多源配网监测数据的电网运行状态监测方法。
本实施例还提供一种计算机可读存储介质，其存储有程序，所述程序被处理器执行时，实现所述的多源配网监测数据的电网运行状态监测方法。
本实施例提出的存储介质与上述实施例提出的多源配网监测数据的电网运行状态监测方法属于同一发明构思，未在本实施例中详尽描述的技术细节可参见上述实施例，并且本实施例与上述实施例具有相同的有益效果。
通过以上关于实施方式的描述，所属领域的技术人员可以清楚地了解到， 本发明可借助软件及必需的通用硬件来实现，当然也可以通过硬件实现，但很多情况下前者是更佳的实施方式。基于这样的理解，本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来，该计算机软件产品可以存储在计算机可读存储介质中，如计算机的软盘、只读存储器(ReadOnly，Memory，ROM)、随机存取存储器(Random Access Memory，RAM)、闪存(FLASH)、硬盘或光盘等，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本发明各个实施例的方法。
实施例2
参照表1，为本发明的一个实施例，提供了一种多源配网监测数据的电网运行状态监测方法，为了验证其有益效果，提供了两种方案的对比结果。
表1对比表
从表1可以看出，我方对于电网运行状态监测处理较为细致，能够基于子电网的供电数据和地理信息确定供电规律信息，从而确定子电网的运行状态，能够在地理信息复杂的情况下及时发现异常状况，减少经济损失，提升检修的便利性。
应说明的是，以上实施例仅用以说明本发明的技术方案而非限制，尽管参照较佳实施例对本发明进行了详细说明，本领域的普通技术人员应当理解，可以对本发明的技术方案进行修改或者等同替换，而不脱离本发明技术方案的精神和范围，其均应涵盖在本发明的权利要求范围当中。

Claims (10)

  1. 一种多源配网监测数据的电网运行状态监测方法，其特征在于，包括：
    获取子电网供电数据和使用子电网供电的地理信息，构建子电网的供电数据矩阵；
    对所述供电数据矩阵进行聚类，确定供电数据异常的待测子电网；
    根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息；
    将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态，提升检修的便利性。
  2. 如权利要求1所述的多源配网监测数据的电网运行状态监测方法，其特征在于：根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息，包括：
    根据所述第一历史供电数据，确定所述待测子电网在多个时刻的负载数据的第一分布特征；
    根据所述第二历史供电数据，确定所述与所述待测子电网属于同一聚类簇的子电网在多个时刻的负载数据的第二分布特征；
    根据所述第一分布特征和所述第二分布特征，获得所述供电规律信息。
  3. 如权利要求1或2所述的多源配网监测数据的电网运行状态监测方法，其特征在于：根据所述第一历史供电数据，确定所述待测子电网在多个时刻的负载数据的第一分布特征，包括：确定多个采样周期中相同采样时刻的负载数据；
    对多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的第一备选负载数据；
    对多个采样时刻的第一备选负载数据进行拟合，获得所述第一分布特征。
  4. 如权利要求3所述的多源配网监测数据的电网运行状态监测方法，其特征在于：对多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的第一备选负载数据，包括：
    确定多个采样周期中相同采样时刻的负载数据的均值和方差；
    根据公式D1：m-θ1σ≤lj≤m+θ1σ，筛选第一备选负载数据，其中，D1为第一筛选条件函数，m为负载数据的均值，σ为负载数据的方差，θ1为第一预设倍数。
  5. 如权利要求4所述的多源配网监测数据的电网运行状态监测方法，其特征在于：根据所述第二历史供电数据，确定所述与所述待测子电网属于同一聚类簇的子电网在多个时刻的负载数据的第二分布特征，包括：
    确定各子电网在多个采样周期中相同采样时刻的负载数据；
    对各子电网在多个采样周期中相同采样时刻的负载数据进行筛选，获得与所述采样时刻对应的各子电网的第二备选负载数据；
    对各子电网的所述第二备选负载数据进行拟合，获得各子电网的第三分布特征；
    对各子电网的第三分布特征进行加权求和，获得所述第二分布特征。
  6. 如权利要求5所述的多源配网监测数据的电网运行状态监测方法，其特征在于：对各子电网的第三分布特征进行加权求和，获得所述第二分布特征，包括：
    根据所述各子电网在多个采样周期中多个时刻的负载数据，获得各子电网的负载数据的第一数据序列；
    根据所述待测子电网在在多个采样周期中多个时刻的负载数据，获得参考数据序列；
    根据各子电网的第一数据序列和所述参考数据序列，获得各子电网与待测子电网之间的元素差距矩阵，其中，元素差距矩阵中第i行第j列的元素为第一数据序列中第i个元素与参考数据序列中第j个元素之差；
    根据公式：
    确定所述各子电网的第三分布特征的权值，其中，αk为第k个子电网的第三分布特征的权值，Ek(i，i)为第k个子电网与待测子电网之间的元素差距矩阵中第i行第i列的元素，L(i，j)表示在所述元素差距矩阵中，从(1，1)元素至(i，j)元素的多个路径中的目标路径的元素的集合，其中，所述目标路径为经历的元素之和最小的路径；
    根据所述各子电网的第三分布特征的权值，对各子电网的第三分布特征进行加权求和，获得所述第二分布特征。
  7. 如权利要求4-6任一项所述的多源配网监测数据的电网运行状态监测方 法，其特征在于：将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态，包括：
    根据所述供电规律信息，确定所述待测子电网的供电数据矩阵中的离群点；
    对所述离群点进行统计，获得离群点的多种统计信息；
    根据离群点的统计信息，确定所述待测子电网的运行状态。
  8. 一种多源配网监测数据的电网运行状态监测系统，其特征在于，包括：
    采样模块，用于在多个采样时刻，获取多个子电网的多个类型的供电数据，以及使用各子电网供电的地理信息，所述地理信息包括区域位置信息，以及地块属性信息；
    矩阵模块，用于根据所述地理信息和所述供电数据，获得各个子电网的供电数据矩阵；
    聚类模块，用于对所述供电数据矩阵进行聚类，确定供电数据异常的待测子电网；
    规律模块，用于根据所述待测子电网的第一历史供电数据，以及与所述待测子电网属于同一聚类簇的子电网的第二历史供电数据，确定供电规律信息；
    运行状态模块，用于将所述供电规律信息和所述待测子电网的供电数据矩阵输入电网运行状态监测模型，确定所述待测子电网的运行状态。
  9. 一种电子设备，其特征在于，包括：
    存储器和处理器；
    所述存储器用于存储计算机可执行指令，所述处理器用于执行所述计算机可执行指令，该计算机可执行指令被处理器执行时实现权利要求1至7任意一项所述多源配网监测数据的电网运行状态监测方法的步骤。
  10. 一种计算机可读存储介质，其存储有计算机可执行指令，该计算机可执行指令被处理器执行时实现权利要求1至7任意一项所述多源配网监测数据的电网运行状态监测方法的步骤。
PCT/CN2023/116455 2022-12-08 2023-09-01 一种多源配网监测数据的电网运行状态监测方法及系统 WO2024119919A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211571855.0 2022-12-08
CN202211571855.0A CN116317101A (zh) 2022-12-08 2022-12-08 一种多源配网监测数据的电网运行状态监测方法及系统

Publications (1)

Publication Number Publication Date
WO2024119919A1 true WO2024119919A1 (zh) 2024-06-13

Family

ID=86789420

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/116455 WO2024119919A1 (zh) 2022-12-08 2023-09-01 一种多源配网监测数据的电网运行状态监测方法及系统

Country Status (2)

Country Link
CN (1) CN116317101A (zh)
WO (1) WO2024119919A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN116317101A (zh) * 2022-12-08 2023-06-23 云南电网有限责任公司信息中心 一种多源配网监测数据的电网运行状态监测方法及系统

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3107174A1 (de) * 2015-06-19 2016-12-21 Siemens Aktiengesellschaft Verfahren, steuereinrichtung und system zum betreiben eines teilnetzes eines energieversorgungsnetzes
CN113742650A (zh) * 2021-08-16 2021-12-03 国网河南省电力公司电力科学研究院 一种分布式传感数据处理方法及装置
CN114662994A (zh) * 2022-05-18 2022-06-24 国网山西省电力公司晋城供电公司 一种整县式光伏分区方法、存储设备及终端
CN114676883A (zh) * 2022-03-02 2022-06-28 深圳江行联加智能科技有限公司 基于大数据的电网运行管理方法、装置、设备及存储介质
CN116317101A (zh) * 2022-12-08 2023-06-23 云南电网有限责任公司信息中心 一种多源配网监测数据的电网运行状态监测方法及系统

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP3107174A1 (de) * 2015-06-19 2016-12-21 Siemens Aktiengesellschaft Verfahren, steuereinrichtung und system zum betreiben eines teilnetzes eines energieversorgungsnetzes
CN113742650A (zh) * 2021-08-16 2021-12-03 国网河南省电力公司电力科学研究院 一种分布式传感数据处理方法及装置
CN114676883A (zh) * 2022-03-02 2022-06-28 深圳江行联加智能科技有限公司 基于大数据的电网运行管理方法、装置、设备及存储介质
CN114662994A (zh) * 2022-05-18 2022-06-24 国网山西省电力公司晋城供电公司 一种整县式光伏分区方法、存储设备及终端
CN116317101A (zh) * 2022-12-08 2023-06-23 云南电网有限责任公司信息中心 一种多源配网监测数据的电网运行状态监测方法及系统

Also Published As

Publication number Publication date
CN116317101A (zh) 2023-06-23

Similar Documents

Publication Publication Date Title
WO2024119919A1 (zh) 一种多源配网监测数据的电网运行状态监测方法及系统
Bourdeau et al. Classification of daily electric load profiles of non-residential buildings
US8989910B1 (en) Systems and methods for data mining of energy consumption data
CN111160404B (zh) 配电网线损标杆合理值的分析方法及装置
CN116976707B (zh) 基于用电信息采集的用户用电数据异常分析方法及系统
Evans et al. Multivariate data analytics in PV manufacturing—Four case studies using manufacturing datasets
Milton et al. Characterization and classification of daily electricity consumption profiles: shape factors and k-means clustering technique
Saad et al. Data analytics-based anomaly detection in smart distribution network
CN114417971A (zh) 一种基于k近邻密度峰值聚类的电力数据异常值检测算法
CN117391499A (zh) 光伏电站可靠性评价方法及其装置
CN111680851A (zh) 基于k线图的企业用电趋势评估方法
Rashid et al. Collect, compare, and score: a generic data-driven anomaly detection method for buildings
CN117236800B (zh) 一种多场景自适应的用电负荷预测方法及系统
CN106298582A (zh) 制程异常诊断系统及方法
CN107274025B (zh) 一种实现用电模式智能识别与管理的系统和方法
CN117828485A (zh) 一种逆变器低效诊断方法及装置
ES2967447T3 (es) Método, sistema y producto de programa informático para la evaluación del consumo de energía en entornos industriales
CN117217726A (zh) 一种综合能源运维的异常数据监测方法及系统
CN104714205B (zh) 电表错置侦测系统及其方法
CN105678388A (zh) 一种基于基线的运维管理健康分析方法
CN116108362A (zh) 一种基于多时空尺度数据的电网异常识别方法及系统
CN105573269A (zh) 半导体制造机台的参数监控系统及方法
CN115905360A (zh) 一种基于随机构建矩阵的异常数据量测辨识方法和装置
CN115579862A (zh) 基于多维多层关联规则的电压暂降预测分析方法
CN114116837A (zh) 数据处理方法、装置、计算机设备和存储介质