CN117273284A - Abnormal data monitoring system for enterprise electricity balance - Google Patents
Abnormal data monitoring system for enterprise electricity balance Download PDFInfo
- Publication number
- CN117273284A CN117273284A CN202311548889.2A CN202311548889A CN117273284A CN 117273284 A CN117273284 A CN 117273284A CN 202311548889 A CN202311548889 A CN 202311548889A CN 117273284 A CN117273284 A CN 117273284A
- Authority
- CN
- China
- Prior art keywords
- data
- power consumption
- abnormal
- historical power
- cluster
- 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.)
- Withdrawn
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 130
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 113
- 230000005611 electricity Effects 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 claims abstract description 55
- 238000007405 data analysis Methods 0.000 claims abstract description 7
- 238000010606 normalization Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000012545 processing Methods 0.000 claims description 12
- 238000013528 artificial neural network Methods 0.000 claims description 11
- 230000005856 abnormality Effects 0.000 claims description 3
- 238000011156 evaluation Methods 0.000 claims description 3
- 230000008859 change Effects 0.000 description 9
- 230000033228 biological regulation Effects 0.000 description 7
- 230000000694 effects Effects 0.000 description 4
- 230000001105 regulatory effect Effects 0.000 description 3
- 239000000758 substrate Substances 0.000 description 3
- 238000012549 training Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 241001155961 Baris Species 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000011217 control strategy Methods 0.000 description 1
- 238000007418 data mining Methods 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Classifications
-
- 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
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2337—Non-hierarchical techniques using fuzzy logic, i.e. fuzzy clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Human Resources & Organizations (AREA)
- Economics (AREA)
- General Physics & Mathematics (AREA)
- Strategic Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Tourism & Hospitality (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Evolutionary Computation (AREA)
- Entrepreneurship & Innovation (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Operations Research (AREA)
- Educational Administration (AREA)
- General Health & Medical Sciences (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Development Economics (AREA)
- Primary Health Care (AREA)
- Game Theory and Decision Science (AREA)
- Fuzzy Systems (AREA)
- Automation & Control Theory (AREA)
- Quality & Reliability (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The invention relates to the technical field of data analysis, in particular to an abnormal data monitoring system for enterprise electricity balance, which comprises the following components: the data acquisition module is used for acquiring historical power consumption data of all feeder lines under all buses and constructing an abnormal monitoring space; the data classification module is used for acquiring a plurality of initial data clusters and data points in each initial data cluster according to the abnormal monitoring space; the data analysis module is used for acquiring a first membership factor and a second membership factor of each data point belonging to each initial data cluster in the abnormal monitoring space; the data clustering module is used for acquiring the membership degree of each feeder line, which is affiliated to each initial data cluster; and obtaining a plurality of data clusters, wherein the data monitoring module is used for judging whether abnormal data exist in the enterprise electricity balance data according to the data clusters. According to the method, the purpose of accurately identifying the abnormal data in the enterprise electricity balance data is achieved by analyzing the historical electricity consumption data.
Description
Technical Field
The invention relates to the technical field of data analysis, in particular to an abnormal data monitoring system for enterprise electricity balance.
Background
The power utilization balance of an enterprise refers to reasonably and efficiently managing and utilizing power resources in the operation process of the enterprise so as to realize the balance state between power utilization requirements and supply. It relates to aspects of enterprise electricity usage including energy consumption, load distribution, energy cost control, and environmental sustainability.
And (3) carrying out current multi-use load management analysis and historical data mining on the power balance regulation and control of the enterprise so as to obtain historical power characteristics of the enterprise, predicting and planning future power trend according to the power characteristics, realizing a peak-staggering power utilization strategy, and guiding the enterprise to transfer part of power load to a low peak or a non-peak period. In the process of acquiring the historical electricity utilization characteristics, the method has higher accuracy requirements on the distribution strategy of the historical electricity utilization of the enterprise and the monitored load data, if the monitored load data is abnormal, the abnormal data is an outlier compared with the normal data in the normal case, and the electricity utilization regulation and control is mainly used for meeting all electricity utilization conditions, the outlier data can cause larger deviation when the enterprise electricity utilization balance is predicted and planned, so that energy waste and energy efficiency are caused.
Disclosure of Invention
The invention provides an abnormal data monitoring system for enterprise electricity balance, which aims to solve the existing problems: the clustering result of the traditional fuzzy C clustering has errors, so that whether abnormal data exist in the enterprise electricity balance data cannot be accurately judged.
The abnormal data monitoring system for enterprise electricity balance adopts the following technical scheme:
the method comprises the following modules:
the data acquisition module is used for acquiring historical power consumption data of all feeder lines under all buses; constructing an abnormal monitoring space according to the historical power consumption data of all feeder lines under all buses, and short-circuiting the historical power consumption data points in the abnormal monitoring space as data points;
the data classification module is used for acquiring a plurality of initial data clusters and data points in each initial data cluster according to the abnormal monitoring space;
the data analysis module is used for acquiring a first membership factor of each data point belonging to each initial data cluster in the abnormal monitoring space according to the historical power consumption of different feeder lines under the same bus in the initial data cluster; acquiring a second membership factor of each data point belonging to each initial data cluster in the abnormal monitoring space according to the historical power consumption under different buses in the initial data clusters;
the data clustering module is used for acquiring the membership degree of each feeder line, which is affiliated to each initial data cluster, according to the first membership factor of each data point affiliated to each initial data cluster in the abnormal monitoring space and the second membership factor of each data point affiliated to each initial data cluster in the abnormal monitoring space; constructing objective functions of all initial data clusters in the abnormal monitoring space according to the membership degree of each feeder belonging to each initial data cluster, and carrying out fuzzy C clustering on data points in the abnormal monitoring space by utilizing the objective functions to obtain a plurality of data clusters;
and the data monitoring module is used for judging whether abnormal data exist in the enterprise electricity balance data according to the plurality of data clusters.
Preferably, the collecting historical power consumption data of all feeder lines under all buses includes the following specific steps:
presetting an acquisition time intervalThe method comprises the steps of carrying out a first treatment on the surface of the Use of a power sensor at preset acquisition time intervals +.>Collecting enterprisesHistorical power data of all buses and all feeders in a substation-bus system, and will +.>And recording seconds as a moment to obtain historical power data of all buses and all feeder lines at all moments, and performing integral operation on the historical power data and time of all buses and all feeder lines at all moments to obtain historical power consumption data of all buses and all feeder lines.
Preferably, the method for constructing the anomaly monitoring space according to the historical power consumption data of all feeder lines under all buses includes the following specific steps:
performing maximum normalization processing on the historical power consumption data of all buses and all feeder lines to obtain historical power consumption data results of all buses and all feeder lines subjected to the maximum normalization processing, wherein the historical power consumption data results of all buses and all feeder lines subjected to the maximum normalization processing are the firstTime lower->Historical power consumption of the individual bus, marked +.>Time lower->Historical power consumption of individual bus bars->The +.>Time lower->No. under the individual bus bar>The historical power consumption of the individual feeders, denoted +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeder lines->The method comprises the steps of carrying out a first treatment on the surface of the And finally, constructing a three-dimensional coordinate system by taking time as a horizontal axis, taking the historical power consumption as a vertical axis and taking the buses and the feeder lines as vertical axes, and filling the historical power consumption data results of all buses and all feeder lines subjected to maximum normalization processing into the three-dimensional coordinate system to obtain an anomaly monitoring space.
Preferably, the acquiring a plurality of initial data clusters according to the anomaly monitoring space and the data points in each initial data cluster includes the following specific methods:
iterative evaluation is carried out on all data points in the abnormal monitoring space through an elbow method, so that clustering centers of all data points in the abnormal monitoring space are obtained, and the number of the clustering centers of all the data points in the abnormal monitoring space is recorded asAccording to the number of clustering centers of all data points in the abnormal monitoring space, performing primary iteration of fuzzy C clustering on all data points in the abnormal monitoring space to obtain +.>A number of initial data clusters, and data points in each initial data cluster.
Preferably, the obtaining, according to the historical power consumption of different feeders under the same busbar in the initial data cluster, a first membership factor that each data point in the anomaly monitoring space belongs to each initial data cluster includes the following specific calculation formula:
in the method, in the process of the invention,indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A first membership factor of the initial data cluster; />Indicate->Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders;indicate->Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->Time lower->Historical power consumption of the individual bus bars; />Indicate->Time lower->Historical power consumption of the individual bus bars; />Indicate->The +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->The +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->The +.>Time lower->Number of data points under the individual bus bars; />An exponential function based on a natural constant; />Representing an absolute value operation; />Is a preset super parameter.
Preferably, the obtaining the second membership factor of each data point belonging to each initial data cluster in the anomaly monitoring space according to the historical power consumption under different buses in the initial data cluster includes the following specific calculation formulas:
in the method, in the process of the invention,indicate->No. under the individual bus bar>A number of data points in the feeder lines; />Indicate->The +.>Time lower->Number of data points under the individual bus bars; />Is indicated at +.>The +.>Time lower->No. under the individual bus bar>A number of data points in the feeder lines; />Is indicated at +.>The first data cluster in the initial data clusterTime lower->No. under the individual bus bar>A number of data points in the feeder lines; />Indicate->At the moment ofFirst->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A second membership factor for the initial data cluster; />Representing the total number of bus bars; />Representing a linear normalization function.
Preferably, the obtaining the membership degree of each feeder line belonging to each initial data cluster includes the following specific calculation formula:
in the method, in the process of the invention,indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A first membership factor of the initial data cluster; />Indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A second membership factor for the initial data cluster; />Representing the number of moments in the anomaly monitoring space; />Representing the number of cluster centers for all data points in the anomaly monitoring space; />Indicate->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>Membership of the individual initial data clusters.
Preferably, the objective function of all the initial data clusters in the anomaly monitoring space is constructed according to the membership degree of each feeder line belonging to each initial data cluster, and the objective function is used for carrying out fuzzy C clustering on the data points in the anomaly monitoring space to obtain a plurality of data clusters, comprising the following specific methods:
the objective function of all initial data clusters is:
in the method, in the process of the invention,an objective function representing all initial data clusters in the anomaly monitoring space; />Representing the number of all data points in the anomaly monitoring space; />Representing the number of cluster centers for all data points in the anomaly monitoring space; />Is indicated in the abnormality monitoring space +.>Time lower->No. under the individual bus bar>Historical power consumption data distance of individual feeder lines +.>Euclidean distance of cluster centers of the initial data clusters; />Representing preset fuzzy weights; />Indicate->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>Initial number ofMembership of the clusters;
fuzzy C clustering is carried out on all data points in the abnormal monitoring space through objective functions of all initial data clusters in the abnormal monitoring space, and thus obtainingAnd data clusters.
Preferably, the method for judging whether abnormal data exists in the power consumption balance data of the enterprise according to the plurality of data clusters includes the following specific steps:
taking the data cluster with the largest data quantity in a plurality of data clusters as a reference data cluster, inputting the data in the reference data cluster into a neural network, judging whether abnormal data exist in the enterprise electricity balance data, and sending an alarm signal if the abnormal data exist; if no abnormal data exists, no alarm signal is sent out.
Preferably, the neural network is an LSTM neural network.
The technical scheme of the invention has the beneficial effects that: the membership degree of the traditional fuzzy C-cluster is obtained through the distance between a data point to be clustered in an abnormal monitoring space and each data point of an initial data cluster, but in the electricity utilization balance data, the influence of buses, feeder lines and feeder lines is caused by a regulation strategy, so that the membership degree of the data point to be clustered and the initial data cluster is caused to generate errors, the clustering result is caused to generate errors, and the clustering result is inaccurate, namely whether the abnormal data exists in the enterprise electricity utilization balance data cannot be accurately judged.
According to the method, fuzzy C clustering is carried out on data points in the abnormal monitoring space according to the objective function of all initial data clusters reconstructed in the abnormal monitoring space, errors of clustering results are avoided, and whether abnormal data exist in enterprise electricity balance data is judged according to the clustering results.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a block diagram of an anomaly data monitoring system for enterprise electricity balance according to the present invention.
Detailed Description
In order to further describe the technical means and effects adopted by the invention to achieve the preset aim, the following detailed description is given below of a specific implementation, structure, characteristics and effects of the abnormal data monitoring system for enterprise electricity balance according to the invention in combination with the accompanying drawings and the preferred embodiment. In the following description, different "one embodiment" or "another embodiment" means that the embodiments are not necessarily the same. Furthermore, the particular features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.
The following specifically describes a specific scheme of the abnormal data monitoring system for enterprise power consumption balance provided by the invention with reference to the accompanying drawings.
Referring to fig. 1, a block diagram of an abnormal data monitoring system for enterprise electricity balance according to an embodiment of the present invention is shown, where the system includes the following modules:
the data acquisition module 101 is used for acquiring historical power consumption data of all feeder lines under all buses; and constructing an abnormal monitoring space according to the historical power consumption data of all feeder lines under all buses, and short-circuiting the historical power consumption data points in the abnormal monitoring space as data points.
It should be noted that, the present embodiment is used as an abnormal data monitoring system for enterprise electricity balance, which aims to realize timely early warning of enterprise electricity abnormal conditions, so that historical electricity consumption data of all buses and all feeders are needed first.
In particular, the method comprises the steps of,presetting an acquisition time interval，/>The specific size of (2) can be set according to the actual situation, the hard requirement is not made in the present embodiment, in the present embodiment +.>Second to describe; use of a power sensor at preset acquisition time intervals +.>Collecting historical power data of all buses and all feeder lines in a substation-bus system of an enterprise, and adding +.>The second is recorded as a moment to obtain the historical power data of all buses and all feeder lines at all moments, and the historical power data and time of all buses and all feeder lines at all moments are subjected to integral operation to obtain the historical power consumption data of all buses and all feeder lines.
It should be further noted that, in order to better analyze the historical power consumption data of all the buses and all the feeders, it is necessary to monitor the abnormal data in the historical power consumption data of all the buses and all the feeders, and also construct an abnormal monitoring space.
Specifically, performing maximum normalization processing on the historical power consumption data of all buses and all feeder lines to obtain the historical power consumption data result of all buses and all feeder lines after the maximum normalization processing, and performing maximum normalization processing on the historical power consumption data result of all buses and all feeder linesTime lower->Personal bus barIs marked by>Time lower->Historical power consumption of individual bus bars->The +.>Time lower->No. under the individual bus bar>The historical power consumption of the individual feeders, denoted +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeder lines->The method comprises the steps of carrying out a first treatment on the surface of the And finally, taking time as a horizontal axis, taking the historical power consumption as a vertical axis, taking the buses and the feeder lines as vertical axes, constructing a three-dimensional coordinate system, filling the historical power consumption data results of all buses and all feeder lines subjected to maximum normalization processing into the three-dimensional coordinate system, obtaining an abnormal monitoring space, and short-term data points of the historical power consumption in the abnormal monitoring space.
Thus, an anomaly monitoring space and data points in the anomaly monitoring space are obtained.
A data classification module 102 for acquiring according to the anomaly monitoring spaceA number of initial data clusters, and data points in each initial data cluster.
In the transformer substation-bus system, when the output power of the transformer substation is unchanged, the plurality of buses are mutually regulated and controlled according to the power consumption, namely when the power consumption of a part of buses is increased, the power consumption of other buses is correspondingly reduced; similarly, for a bus-sub-line system, the power consumption of one sub-line increases and the power consumption of the other sub-line decreases. Therefore, the embodiment provides a method for obtaining the membership degree according to the influence of coordination association between the sub-line and the bus and the influence of change of each data point in each data point to be clustered and each data point in the cluster, and obtaining a standard initial data cluster by combining with fuzzy C clustering, so as to obtain a decision model of power utilization balance.
It should be further noted that, the fuzzy C-cluster is a clustering algorithm based on the objective function, and since different data distribution types have different characteristics, it is generally necessary to give the initial cluster center number of the objective function when constructing the objective function for the fuzzy C-cluster. Therefore, the elbow method is selected to obtain the number of initial clustering centers, the elbow method is used for forming an error square sum curve through iterating the number of the initial clustering centers, the curve is similar to the elbow, namely, an obvious inflection point exists, the clustering effect before the inflection point is increased along with the increase of the clustering centers, the clustering effect after the inflection point is not greatly changed or even reduced, and then the inflection point is the optimal clustering center number.
Specifically, iteration evaluation is performed on all data points in the abnormal monitoring space through an elbow method to obtain a clustering center of all data points in the abnormal monitoring space, and the elbow method is a well-known prior art, so that redundant description is omitted in the embodiment; the number of clustering centers of all data points in the anomaly monitoring space is recorded asAccording to the number of clustering centers of all data points in the abnormal monitoring space, performing primary iteration of fuzzy C clustering on all data points in the abnormal monitoring space to obtain +.>The initial data clusters and the data points in each initial data cluster are not described in this embodiment, since the initial iteration of the fuzzy C-cluster is a well-known prior art.
To this end, we obtainAn initial data cluster, and data points in each initial data cluster.
The data analysis module 103 is configured to obtain a first membership factor of each data point in the anomaly monitoring space, where the first membership factor belongs to each initial data cluster, according to the historical power consumption of different feeder lines under the same busbar in the initial data cluster; and acquiring a second membership factor of each data point belonging to each initial data cluster in the abnormal monitoring space according to the historical power consumption under different buses in the initial data clusters.
It should be noted that, the membership degree of the fuzzy C-cluster is obtained by the distance between the data point to be clustered and each data point of the initial data cluster in the abnormal monitoring space, but in the electricity balance data, the influence of the bus and the bus, the bus and the feeder, and the feeder is caused by the regulation and control strategy, so that the membership degree of the data point to be clustered and the initial data cluster generates errors, the clustering result generates errors, and the clustering result is inaccurate.
1. A first membership factor is obtained.
It should be further noted that, for a certain historical power consumption data point in the anomaly monitoring space, if the number of other data points in the feeder line where the historical power consumption data point is located is greater and the number of other data points are affiliated to a certain initial data cluster, the historical power consumption data point is more likely to be affiliated to the initial data cluster; however, in the actual calculation process, each feeder line is not only affected by the influence of the feeder lines, but also affected by the regulation and control of the bus lines, namely, the sum of the historical power consumption data of all feeder lines under one bus line is different at different moments, so that the embodiment calculates the historical power consumption data of each feeder line under each bus line under all moments by the change difference of the historical power consumption data at adjacent moments, and the historical power consumption data belongs to the first membership factor of each initial data cluster; that is, for a certain data point to be clustered in a bus, when the sum of the change value of the data point to be clustered and the change value of the clustered data point in the initial data cluster is close to the change amount of the bus, the more likely that the data point to be clustered is affiliated to the initial data cluster.
Specifically, for the calculation ofTime lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>The specific calculation formula of the first membership factor of each initial data cluster is as follows:
in the method, in the process of the invention,indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A first membership factor of the initial data cluster; />Indicate->Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders;indicate->Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->Time lower->Historical power consumption of the individual bus bars; />Indicate->Time lower->Historical power consumption of the individual bus bars; />Indicate->The +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->The +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->The +.>Time lower->Number of data points under the individual bus bars; />An exponential function based on a natural constant; />Representing an absolute value operation; />Is a preset super parameter->The specific size of (2) can be set according to the actual situation, and in the embodiment, the method is as followsThe purpose of this description is to prevent the denominator from being 0 when the partial operation is performed.
It should be noted that the number of the substrates,indicate->Time and->Time->Historical power consumption difference of each bus, wherein the difference is a regulated change value, +>Indicate->Time and->Time->No. under the individual bus bar>Historical power consumption difference of individual feeder,/>For data points to be clustered +.>Indicate->The sum of the change values of the clustered data points in the initial data cluster, thus +.>The closer the value is to 1, the more data points to be clustered ∈>The more likely it is to be subordinate to +.>Initial data clusters; thus->The larger the value of (2) is the data point to be clustered +.>The more likely it is to be subordinate to +.>Initial data clusters.
Thus far, get the firstTime lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A first membership factor of the initial data cluster; and similarly, acquiring historical power consumption data of each feeder line under each busbar at all moments, wherein the historical power consumption data belongs to a first membership factor of each initial data cluster, namely the first membership factor of each data point in the abnormal monitoring space belongs to each initial data cluster.
2. And obtaining a second membership factor.
It should be further noted that, not only the feeder lines under the same bus are mutually affected by regulation and control, but also the regulation and control between different bus lines can also affect the feeder lines, when the power consumption of the bus line is unchanged, the accumulated value of the power consumption of all the feeder lines below the same is the power consumption of the bus line, but along with the change between the regulated bus lines, the method also has the change, so the data points of the feeder lines under different bus lines in adjacent time are in the first timeAnd obtaining a second membership factor by changing the duty ratio in each initial data cluster.
Specifically, for the calculation ofTime lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>The specific calculation formula of the first membership factor of each initial data cluster is as follows:
in the method, in the process of the invention,indicate->No. under the individual bus bar>A number of data points in the feeder lines; />Indicate->The +.>Time lower->Number of data points under the individual bus bars; />Is indicated at +.>The +.>Time lower->No. under the individual bus bar>A number of data points in the feeder lines; />Is indicated at +.>The first data cluster in the initial data clusterTime lower->No. under the individual bus bar>A number of data points in the feeder lines; />Indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A second membership factor for the initial data cluster; />Representing the total number of bus bars; />Indicated at->The +.>No. under the individual bus bar>A number of data points in the feeder lines;representing a linear normalization function.
It should be noted that the number of the substrates,indicated at->The +.>No. under the individual bus bar>The number of data points in the individual feeder, +.>The greater the value of +.>Time lower->No. under the individual bus bar>The more likely the historical power consumption data of the individual feeders are to be affiliated to +.>In the initial data clusters; />Indicated at->Difference in number of data points at adjacent time points in each initial data cluster, < >>The greater the value of +.>Time lower->No. under the individual bus bar>The more likely the historical power consumption data of the individual feeders are to be affiliated to +.>In the initial data clusters; thus->The greater the value of (2)Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders, the more likely it is to be subordinate to +.>Initial data clusters.
Thus far, get the firstTime lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A second membership factor for the initial data cluster; and similarly, acquiring historical power consumption data of each feeder line under each busbar at all moments, wherein the historical power consumption data belongs to a second membership factor of each initial data cluster, namely the second membership factor of each data point in the abnormal monitoring space belongs to each initial data cluster.
The data clustering module 104 is configured to obtain a membership degree of each feeder line, where the membership degree is subordinate to each initial data cluster, according to a first membership factor of each data point in the anomaly monitoring space, where the first membership factor of each initial data cluster is subordinate to each data point in the anomaly monitoring space, and a second membership factor of each data point in the anomaly monitoring space; according to the membership degree of each feeder line belonging to each initial data cluster, constructing an objective function of all initial data clusters in the abnormal monitoring space, and carrying out fuzzy C-clustering on data points in the abnormal monitoring space to obtainAnd data clusters.
It should be noted that, the data analysis module 103 obtains the first membership factor and the second membership factor of each data point in the abnormal monitoring space, where each data point belongs to each initial data cluster, and when the first membership factor and the second membership factor of each data point in the abnormal monitoring space belong to a certain initial data cluster, the larger the data point in the abnormal monitoring space is, the more likely the data point belongs to the initial data cluster; therefore, the membership degree of each feeder line to each initial data cluster can be obtained according to the first membership factor and the second membership factor of each data point in the abnormal monitoring space.
Specifically, according to the first membership factor and the second membership factor of each data point in the abnormal monitoring space, the membership degree of each feeder line belonging to each initial data cluster is obtained, and a specific calculation formula is as follows:
in the method, in the process of the invention,indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A first membership factor of the initial data cluster; />Indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A second membership factor for the initial data cluster; />Representing the number of moments in the anomaly monitoring space; />Representing the number of cluster centers for all data points in the anomaly monitoring space; />Indicate->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>Membership of the individual initial data clusters.
It should be noted that the number of the substrates,the larger the value of (2), the description of +.>No. under the individual bus bar>The more likely the historical power consumption data of the individual feeders are to be affiliated to +.>Initial data clusters.
It should be further noted that after the membership degree of each feeder line belonging to each initial data cluster is obtained, an objective function of all initial data clusters in the anomaly monitoring space may be constructed according to the membership degree of each feeder line belonging to each initial data cluster, and fuzzy C-clustering may be performed on data points in the anomaly monitoring space to obtain a plurality of initial data clusters.
Specifically, the objective function of all initial data clusters is:
in the method, in the process of the invention,an objective function representing all initial data clusters in the anomaly monitoring space; />Representing the number of all data points in the anomaly monitoring space; />Representing the number of cluster centers for all data points in the anomaly monitoring space; />Is indicated in the abnormality monitoring space +.>Time lower->No. under the individual bus bar>Historical power consumption data distance of individual feeder lines +.>Euclidean distance of cluster centers of the initial data clusters; />Representing preset fuzzy rights, +.>The specific size of (2) can be set according to the actual situation, the hard requirement is not required in the embodiment, and +_ is adopted in the embodiment>Calculating; />Indicate->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>Membership of the individual initial data clusters.
Fuzzy C clustering is carried out on all data points in the abnormal monitoring space through objective functions of all initial data clusters in the abnormal monitoring space, and thus obtainingThe fuzzy C-cluster is a known technology, and therefore will not be described in detail in this embodiment.
So far as the above-mentioned is obtained,and data clusters.
A data monitoring module 105 forAnd judging whether abnormal data exist in the enterprise electricity balance data or not by the data clusters.
It should be noted that, the present embodiment is used as an abnormal data monitoring system for enterprise electricity balance, and its final purpose is to use all buses at all timesAnalyzing historical power consumption data in all feeder lines, and early warning the abnormal power consumption condition of the enterprise in time; and is obtained by the data clustering module 104And the data clusters are used for monitoring abnormal data of enterprise electricity balance.
Specifically, it willThe data cluster with the largest data quantity in the data clusters is used as a reference data cluster, the data in the reference data cluster is input into the neural network, whether abnormal data exist in enterprise electricity balance data is judged, and if abnormal data exist, an alarm signal is sent out; if no abnormal data exists, no alarm signal is sent out.
The neural network used in the embodiment is an LSTM neural network, the training method of the network takes data in a reference data cluster as a data set, each data sequence to be encoded in the data set is artificially allocated with a label, and if the data sequence to be encoded contains abnormal operation parameters, the allocated label is 1; if the data sequence to be encoded does not contain an abnormal operation parameter, the assigned tag is 0. The neural network is trained using the set of data. The data set and the label train the neural network, the loss function used for training is a cross entropy loss function, the training method of the network is well known, and detailed description is omitted in this embodiment.
This embodiment is completed.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the invention, but any modifications, equivalent substitutions, improvements, etc. within the principles of the present invention should be included in the scope of the present invention.
Claims (10)
1. An abnormal data monitoring system for enterprise electricity balance is characterized by comprising the following modules:
the data acquisition module is used for acquiring historical power consumption data of all feeder lines under all buses; constructing an abnormal monitoring space according to the historical power consumption data of all feeder lines under all buses, and short-circuiting the historical power consumption data points in the abnormal monitoring space as data points;
the data classification module is used for acquiring a plurality of initial data clusters and data points in each initial data cluster according to the abnormal monitoring space;
the data analysis module is used for acquiring a first membership factor of each data point belonging to each initial data cluster in the abnormal monitoring space according to the historical power consumption of different feeder lines under the same bus in the initial data cluster; acquiring a second membership factor of each data point belonging to each initial data cluster in the abnormal monitoring space according to the historical power consumption under different buses in the initial data clusters;
the data clustering module is used for acquiring the membership degree of each feeder line, which is affiliated to each initial data cluster, according to the first membership factor of each data point affiliated to each initial data cluster in the abnormal monitoring space and the second membership factor of each data point affiliated to each initial data cluster in the abnormal monitoring space; constructing objective functions of all initial data clusters in the abnormal monitoring space according to the membership degree of each feeder belonging to each initial data cluster, and carrying out fuzzy C clustering on data points in the abnormal monitoring space by utilizing the objective functions to obtain a plurality of data clusters;
and the data monitoring module is used for judging whether abnormal data exist in the enterprise electricity balance data according to the plurality of data clusters.
2. The abnormal data monitoring system for enterprise electricity balance according to claim 1, wherein the collecting historical power consumption data of all feeder lines under all buses comprises the following specific steps:
presetting an acquisition time intervalThe method comprises the steps of carrying out a first treatment on the surface of the Use of a power sensor at preset acquisition time intervals +.>Collecting historical power data of all buses and all feeder lines in a substation-bus system of an enterprise, and adding +.>And recording seconds as a moment to obtain historical power data of all buses and all feeder lines at all moments, and performing integral operation on the historical power data and time of all buses and all feeder lines at all moments to obtain historical power consumption data of all buses and all feeder lines.
3. The abnormal data monitoring system for enterprise electricity balance according to claim 1, wherein the constructing the abnormal monitoring space according to the historical power consumption data of all feeder lines under all buses comprises the following specific steps:
performing maximum normalization processing on the historical power consumption data of all buses and all feeder lines to obtain historical power consumption data results of all buses and all feeder lines subjected to the maximum normalization processing, wherein the historical power consumption data results of all buses and all feeder lines subjected to the maximum normalization processing are the firstTime lower->Historical power consumption of the individual bus, marked +.>Time lower->Historical power consumption of individual bus bars->The +.>At the moment ofFirst->No. under the individual bus bar>The historical power consumption of the individual feeders, denoted +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeder lines->The method comprises the steps of carrying out a first treatment on the surface of the And finally, constructing a three-dimensional coordinate system by taking time as a horizontal axis, taking the historical power consumption as a vertical axis and taking the buses and the feeder lines as vertical axes, and filling the historical power consumption data results of all buses and all feeder lines subjected to maximum normalization processing into the three-dimensional coordinate system to obtain an anomaly monitoring space.
4. The abnormal data monitoring system for enterprise electricity balance according to claim 1, wherein the acquiring a plurality of initial data clusters according to the abnormal monitoring space and the data points in each initial data cluster comprises the following specific methods:
iterative evaluation is carried out on all data points in the abnormal monitoring space through an elbow method, so that clustering centers of all data points in the abnormal monitoring space are obtained, and the number of the clustering centers of all the data points in the abnormal monitoring space is recorded asAccording to the number of clustering centers of all data points in the abnormal monitoring space, performing primary iteration of fuzzy C clustering on all data points in the abnormal monitoring space to obtain +.>A number of initial data clusters, and data points in each initial data cluster.
5. The abnormal data monitoring system for enterprise electricity balance according to claim 3, wherein the obtaining the first membership factor of each data point belonging to each initial data cluster in the abnormal monitoring space according to the historical electricity consumption of different feeder lines under the same busbar in the initial data cluster comprises the following specific calculation formula:
in the method, in the process of the invention,indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A first membership factor of the initial data cluster; />Indicate->Time lower->Personal motherUnder line->Historical power consumption of the individual feeders;indicate->Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->Time lower->Historical power consumption of the individual bus bars; />Indicate->Time lower->Historical power consumption of the individual bus bars; />Indicate->The +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->The +.>Time lower->No. under the individual bus bar>Historical power consumption of the individual feeders; />Indicate->The +.>Time lower->Number of data points under the individual bus bars; />An exponential function based on a natural constant; />Representing an absolute value operation; />Is a preset super parameter.
6. The abnormal data monitoring system for enterprise electricity balance according to claim 1, wherein the specific calculation formula for obtaining the second membership factor of each data point belonging to each initial data cluster in the abnormal monitoring space according to the historical electricity consumption under different buses in the initial data cluster is as follows:
in the method, in the process of the invention,indicate->No. under the individual bus bar>A number of data points in the feeder lines; />Indicate->The +.>Time lower->Number of data points under the individual bus bars; />Is indicated at +.>The +.>At the lower part of the timeNo. under the individual bus bar>A number of data points in the feeder lines; />Is indicated at +.>The +.>Time lower->No. under the individual bus bar>A number of data points in the feeder lines; />Indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A second membership factor for the initial data cluster; />Representing the total number of bus bars;representing a linear normalization function.
7. The abnormal data monitoring system for enterprise electricity balance according to claim 1, wherein the obtaining the membership degree of each feeder line belonging to each initial data cluster comprises the following specific calculation formula:
in the method, in the process of the invention,indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A first membership factor of the initial data cluster; />Indicate->Time lower->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>A second membership factor for the initial data cluster; />Representing the number of moments in the anomaly monitoring space; />Representing the number of cluster centers for all data points in the anomaly monitoring space; />Indicate->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>Membership of the individual initial data clusters.
8. The abnormal data monitoring system for enterprise electricity balance according to claim 1, wherein the constructing an objective function of all initial data clusters in an abnormal monitoring space according to membership of each feeder line to each initial data cluster, and performing fuzzy C-clustering on data points in the abnormal monitoring space by using the objective function to obtain a plurality of data clusters comprises the following specific methods:
the objective function of all initial data clusters is:
in the method, in the process of the invention,an objective function representing all initial data clusters in the anomaly monitoring space; />Representing the number of all data points in the anomaly monitoring space; />Representing the number of cluster centers for all data points in the anomaly monitoring space; />Is indicated in the abnormality monitoring space +.>Time lower->No. under the individual bus bar>Historical power consumption data distance of individual feeder lines +.>Euclidean distance of cluster centers of the initial data clusters; />Representing preset fuzzy weights; />Indicate->No. under the individual bus bar>Historical power consumption data of individual feeders belonging to +.>Membership of the initial data clusters;
fuzzy C clustering is carried out on all data points in the abnormal monitoring space through objective functions of all initial data clusters in the abnormal monitoring space, and thus obtainingAnd data clusters.
9. The abnormal data monitoring system for enterprise electricity balance according to claim 1, wherein the specific method for judging whether abnormal data exists in the enterprise electricity balance data according to a plurality of data clusters comprises the following steps:
taking the data cluster with the largest data quantity in a plurality of data clusters as a reference data cluster, inputting the data in the reference data cluster into a neural network, judging whether abnormal data exist in the enterprise electricity balance data, and sending an alarm signal if the abnormal data exist; if no abnormal data exists, no alarm signal is sent out.
10. The enterprise electricity balance oriented anomaly data monitoring system of claim 9, wherein the neural network is an LSTM neural network.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311548889.2A CN117273284A (en) | 2023-11-21 | 2023-11-21 | Abnormal data monitoring system for enterprise electricity balance |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311548889.2A CN117273284A (en) | 2023-11-21 | 2023-11-21 | Abnormal data monitoring system for enterprise electricity balance |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117273284A true CN117273284A (en) | 2023-12-22 |
Family
ID=89206664
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311548889.2A Withdrawn CN117273284A (en) | 2023-11-21 | 2023-11-21 | Abnormal data monitoring system for enterprise electricity balance |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117273284A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117540327B (en) * | 2024-01-09 | 2024-04-02 | 中山市环境保护技术中心 | Enterprise environment autonomous management data acquisition and processing system |
-
2023
- 2023-11-21 CN CN202311548889.2A patent/CN117273284A/en not_active Withdrawn
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117540327B (en) * | 2024-01-09 | 2024-04-02 | 中山市环境保护技术中心 | Enterprise environment autonomous management data acquisition and processing system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110518880B (en) | Photovoltaic power station state diagnosis method and device | |
| CN103914741A (en) | Line loss intelligent evaluation and assistant decision-making system for power distribution network | |
| CN117273284A (en) | Abnormal data monitoring system for enterprise electricity balance | |
| CN105305488B (en) | A kind of evaluation method for considering new-energy grid-connected and power transmission network utilization rate being influenced | |
| CN113269468A (en) | Power dispatching system based on block chain and data processing method thereof | |
| CN111476397B (en) | Bus load prediction method and system applied to medium and long term electric quantity safety check | |
| CN116599151A (en) | Source network storage safety management method based on multi-source data | |
| CN115693652A (en) | Power distribution network frame optimization method and device based on power balance and performance cost | |
| Zou et al. | Hierarchical evaluation framework for coupling effect enhancement of renewable energy and thermal power coupling generation system | |
| CN115186888A (en) | Energy saving and consumption reduction method based on double carbon targets | |
| Bai et al. | Robust mean-variance optimization model for grid-connected microgrids | |
| CN112330089B (en) | Comprehensive energy efficiency monitoring method and monitoring system for equipment manufacturing enterprises | |
| CN105958474A (en) | Power transmission line dynamic capacity increasing method and system used for power grid regulation and control system | |
| CN117477609A (en) | Energy management system and method of hydrogen electric coupling system for park | |
| Eltohamy et al. | Technical investigation for power system flexibility | |
| Wang et al. | Short-term wind power prediction based on DBSCAN clustering and support vector machine regression | |
| Chen et al. | Interval prediction of photovoltaic power using improved NARX network and density peak clustering based on kernel mahalanobis distance | |
| CN115441440A (en) | Distribution network elasticity improving method and system based on distributed new energy aggregation | |
| CN112491049A (en) | Multi-energy access power grid optimized scheduling method considering on-line coal consumption curve | |
| Di | Power system short term load forecasting based on weather factors | |
| CN116845985B (en) | Source network load storage collaborative scheduling system based on data analysis | |
| Huang et al. | Stackelberg Competition Between Merchant and Regulated Storage Investment under Locational Marginal Pricing | |
| CN117595517B (en) | Intelligent cluster control method and system based on distributed photovoltaic | |
| CN113408892B (en) | Layered cooperative regulation and control system for guaranteeing multi-energy complementary microgrid and regulation and control method thereof | |
| CN107785933A (en) | A kind of new-energy grid-connected electric energy quality monitoring method based on Intelligent Recognition |
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 | ||
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20231222 |