CN112381351B - Power utilization behavior change detection method and system based on singular spectrum analysis - Google Patents
Power utilization behavior change detection method and system based on singular spectrum analysis Download PDFInfo
- Publication number
- CN112381351B CN112381351B CN202011105301.2A CN202011105301A CN112381351B CN 112381351 B CN112381351 B CN 112381351B CN 202011105301 A CN202011105301 A CN 202011105301A CN 112381351 B CN112381351 B CN 112381351B
- Authority
- CN
- China
- Prior art keywords
- data
- electric energy
- matrix
- change
- singular
- 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.)
- Active
Links
Images
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
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
-
- 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
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/06—Electricity, gas or water supply
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
- Y02D10/00—Energy efficient computing, e.g. low power processors, power management or thermal management
-
- 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
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/80—Management or planning
- Y02P90/82—Energy audits or management systems therefor
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Economics (AREA)
- Strategic Management (AREA)
- Mathematical Physics (AREA)
- Mathematical Optimization (AREA)
- Health & Medical Sciences (AREA)
- Marketing (AREA)
- Tourism & Hospitality (AREA)
- Pure & Applied Mathematics (AREA)
- General Business, Economics & Management (AREA)
- Educational Administration (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Development Economics (AREA)
- Mathematical Analysis (AREA)
- Entrepreneurship & Innovation (AREA)
- Software Systems (AREA)
- Game Theory and Decision Science (AREA)
- Computing Systems (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Quality & Reliability (AREA)
- General Engineering & Computer Science (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Operations Research (AREA)
- General Health & Medical Sciences (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a power utilization behavior change detection method and system based on singular spectrum analysis, wherein the method comprises the following steps: acquiring electric energy data of the electric energy metering device, and preprocessing the electric energy data; performing matrix construction and singular value decomposition on historical electric energy data and current data through a singular spectrum analysis method to obtain corresponding singular values and eigenvectors, and obtaining a change mode of the historical data and a change mode of the current data; calculating difference statistics between the historical data and the current data, and evaluating the change degree between the current data and the historical data; and (4) evaluating the change degree of the comprehensive electric energy data to obtain the judgment of the change of the electricity utilization behavior of the user. The embodiment provided by the invention can accurately identify the components with different change characteristics and high significance degree in the power consumption behavior of the user, realize the accurate identification of the different power consumption characteristics before and after the change, and effectively judge the power consumption behavior change trend of the user.
Description
Technical Field
The invention relates to the field of electric energy data processing, in particular to a power utilization behavior change detection method and system based on singular spectrum analysis.
Background
The user electricity utilization behavior is mainly to comprehensively analyze the historical electric energy data of the user so as to predict the electricity utilization of the user, judge the abnormality and the like.
At present, the user electricity consumption behavior is mainly obtained by analyzing a time sequence of user historical electric energy data, and corresponding analysis is carried out on the basis of characteristic information of the time sequence. However, in the use of a general time series analysis tool, the user electric energy data time series is often required to have smoothness, that is, smoothness of the user electric characteristics. However, in the electricity utilization process of an actual user, changes of electricity utilization behaviors such as capacity increase and decrease, technical transformation and the like often exist, the characteristics of the time series of the electric energy data can be changed due to the external factors, the change cannot be predicted through historical data, and only when the time series of the historical electric energy data of the user is analyzed, the change of the electricity utilization behaviors needs to be accurately recognized so as to analyze the time series of the electric energy data under different behavior characteristics.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a power consumption behavior change detection method and system based on singular spectrum analysis.
In order to solve the technical problem, an embodiment of the present invention provides a method for detecting a change in a power consumption behavior of a user based on singular spectrum analysis, where the method for detecting a change in a power consumption behavior of a user includes:
acquiring electric energy data of the electric energy metering device, and preprocessing the electric energy data;
performing matrix construction and singular value decomposition on historical electric energy data and current data through a singular spectrum analysis method to obtain corresponding singular values and characteristic vectors, and obtaining a change mode of the historical data and a change mode of the current data;
calculating difference statistics between the historical data and the current data, and evaluating the change degree between the current data and the historical data;
and (4) evaluating the change degree of the comprehensive electric energy data to obtain the judgment of the change of the electricity utilization behavior of the user.
The electrical energy data comprising:
freezing table codes at 0 point, an electric energy indicating curve at 96 points per day, a current curve at 96 points per day and a power curve at 96 points per day;
and the preprocessing comprises eliminating and fitting null values, maximum values and minimum values of the electric energy data.
The singular spectrum analysis method comprises the following steps:
and obtaining a basic matrix of the data by adopting a time delay embedding method, and then performing singular value decomposition on the basic matrix to obtain corresponding singular values and eigenvectors.
The singular spectrum analysis method comprises the following steps:
carrying out track matrix construction on a time sequence of an electric energy indicating value curve P (t) (t < i) before t = i, and carrying out singular value decomposition to obtain a corresponding singular value and a characteristic vector to obtain a change mode of historical data;
firstly, a basic track matrix of a P (t) time sequence before a time point i is obtained by a time delay embedding method, namely:
wherein t = i, ω is the window length, δ is the number of sampling windows;
then, singular value decomposition is carried out on the matrix B to obtain singular values and singular vectors, namely:
B＝USV T
wherein, U is left singular matrix, V is right singular matrix, S is singular value matrix, diagonal element is singular value called matrix B and is marked as sigma.
Further, because: BB T ＝USV T VS T U T ＝US 2 U T ；
The following can be obtained: (BB) T )U＝US 2 ；
Therefore, the matrix BB can be solved T The eigenvalue λ and eigenvector U obtain a corresponding matrix U, that is:
(BB T )u i ＝λ i u i
wherein U (t) = [ U = 1 ,…,u]，
Will be lambda i Arranged from large to small, the corresponding u i The main features of the P (t) time series before the time point i are shown.
And (3) constructing a track matrix of the time sequence of the electric energy indicating curve P (t) (t > i) after t = i, decomposing singular values to obtain corresponding singular values and eigenvectors, and obtaining the change mode of the current data.
Obtaining a test track matrix of the P (t) time sequence after the time point i by a time delay embedding method, namely:
where t = i, ω is the window length and γ is the number of sampling windows.
Then, singular value decomposition a = M Σ N is performed on the matrix a, where M is a left singular matrix, N is a right singular matrix, Σ is a singular value matrix, and diagonal elements thereof are called singular values of the matrix a.
Solving matrix AA in the same way T Characteristic value of
And a feature vector m, we can get:
wherein M (t) = [ M = 1 ,…,m ω ]，
Is AA T The eigenvalues of which are the squares of the singular values of the matrix a.
The calculating difference statistics between the historical data and the current data comprises: and selecting a characteristic vector with a larger singular value in the basic matrix, constructing to obtain a basic characteristic matrix corresponding to the main characteristics of the historical data, and then comparing the variation degree between the characteristic vector of the test matrix and the basic characteristic matrix to obtain the variation degree between the current data and the historical data.
The calculating difference statistics between the historical data and the current data comprises:
the first l (l) are selected according to the descending order of lambda<Omega) pairCombining the corresponding feature vectors U to obtain a basic feature matrix U l ；
According to
Arranging from large to small, selecting the first eta pieces (eta)<γ) corresponding feature vector m η (ii) a By calculating m η At U l Projection and m η The cosine value of the included angle therebetween, the corresponding change coefficient cp is obtained, namely:
m η at U l The projection vector is:
then the coefficient of variation is obtained:
at the same time, different weights are calculated according to the singular values of the test matrix to obtain the final change coefficient CP, that is
The evaluation of the change degree of the comprehensive electric energy data comprises the following steps: and selecting the user electric energy data time sequence for analysis, calculating the change degree of the electric energy data time sequence, and comprehensively obtaining the change degree of the user electric energy data time sequence so as to obtain a judgment conclusion of the change of the user power consumption behavior.
A user electricity consumption behavior change detection system based on singular spectrum analysis, comprising:
a preprocessing module: the system is used for acquiring the electric energy data of the electric energy metering device and preprocessing the electric energy data;
a data processing module: the method is used for carrying out matrix construction and singular value decomposition on historical electric energy data and current data through a singular spectrum analysis method to obtain corresponding singular values and characteristic vectors and obtain a change mode of the historical data and a change mode of the current data;
a comparison analysis module: the system is used for calculating difference statistics between the historical data and the current data and evaluating the change degree between the current data and the historical data;
a comprehensive evaluation module: and (4) evaluating the change degree of the comprehensive electric energy data to obtain the judgment of the change of the electricity utilization behavior of the user.
The preprocessing module comprises a 0-point freezing meter code, a 96-point-per-day electric energy indicating value curve, a 96-point-per-day current curve and a 96-point-per-day power curve; the preprocessing comprises eliminating and fitting null values and maximum and minimum values of the electric energy data;
the data processing module obtains a basic matrix of the data by adopting a time delay embedding method through a singular spectrum analysis method, and then performs singular value decomposition on the basic matrix to obtain corresponding singular values and eigenvectors;
the comparative analysis module comprises: selecting a characteristic vector with a large singular value in a basic matrix, constructing to obtain a basic characteristic matrix corresponding to main characteristics of historical data, and then obtaining the change degree between current data and historical data by comparing the change degree between the characteristic vector of a test matrix and the basic characteristic matrix;
the comprehensive evaluation module comprises: and selecting the user electric energy data time sequence for analysis, calculating the change degree of the electric energy data time sequence, and comprehensively obtaining the change degree of the user electric energy data time sequence so as to obtain a judgment conclusion of the change of the user power consumption behavior.
The invention provides a power utilization behavior change detection method and system based on singular spectrum analysis. And comparing the change degree between the characteristic vectors of the electric energy data before and after a certain time point, thereby obtaining the change evaluation of the electric energy data. And (4) comprehensively evaluating the change of different types of electric energy data of the user to obtain a judgment conclusion of the change of the electricity utilization behavior of the user. The method can accurately identify different change characteristics and components with high significance degree in the power utilization behavior of the user by using a singular spectrum analysis method, thereby realizing accurate identification of the different power utilization characteristics before and after change and effectively judging the power utilization behavior change trend of the user.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a flow chart schematic diagram of a power consumption behavior change detection method based on singular spectrum analysis.
Fig. 2 is a schematic structural diagram of a power utilization behavior change detection system based on singular spectrum analysis.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Examples
Referring to fig. 1, fig. 1 is a schematic flow chart of a power consumption behavior change detection method based on singular spectrum analysis.
As shown in the figure, a method for detecting a change in a user power consumption behavior based on singular spectrum analysis includes:
s101, electric energy data of the electric energy metering device are obtained and preprocessed.
And acquiring a user electric energy data time sequence, including but not limited to a user 96-point daily electric energy indicating value curve P (t), a 96-point daily voltage curve U (t), a 96-point daily current curve I (t) and a 96-point daily power curve W (t) time sequence.
And preprocessing the acquired electric energy data time sequence. And calculating standard deviations corresponding to various time sequences, and deleting abnormal data points with the numerical values larger than 6 times of the standard deviations according to the standard deviations of 6 times as a threshold.
And fitting by a front-back interpolation method to obtain a fitting data value of the deleted abnormal point. And similarly, fitting the data points with null values by adopting a method of forward and backward interpolation.
S102, performing matrix construction and singular value decomposition on historical electric energy data and current data through a singular spectrum analysis method to obtain corresponding singular values and eigenvectors, and obtaining a change mode of the historical data and a change mode of the current data.
The singular spectrum analysis method comprises the steps of obtaining a basic track matrix of data by adopting a time delay embedding method, namely the basic matrix for short, and then carrying out singular value decomposition on the basic matrix to obtain corresponding singular values and eigenvectors.
Carrying out track matrix construction on a time sequence of an electric energy indicating value curve P (t) (t < i) before t = i, and carrying out singular value decomposition to obtain a corresponding singular value and a characteristic vector to obtain a change mode of historical data;
firstly, a basic track matrix of a P (t) time sequence before a time point i is obtained by a time delay embedding method, namely:
wherein t = i, ω is the window length, δ is the number of sampling windows;
then, singular value decomposition is carried out on the matrix B to obtain singular values and singular vectors, namely:
B＝USV T
where S is a diagonal matrix, and the diagonal elements, i.e., singular values called the matrix, are denoted as σ.
Further, because: BB T ＝USV T VS T U T ＝US 2 U T ；
The following can be obtained: (BB) T )U＝US 2 ；
Therefore, the matrix BB can be solved T The eigenvalue and eigenvector of (a) get the corresponding U, i.e.:
(BBT)u i ＝λ i u i
wherein U (t) = [ U = 1 ,…,u]，
Will be lambda i Arranged from large to small, the corresponding u i The main features of the P (t) time series before the time point i are shown.
And (3) performing track matrix construction on the time sequence of the electric energy indicating value curve P (t) (t > i) after t = i by a singular spectrum analysis method, and performing singular value decomposition to obtain corresponding singular values and characteristic vectors to obtain a change mode of the current data.
As above, the test track matrix after the time point i of the P (t) time sequence is obtained by the method of delay embedding, that is:
where t = i, ω is the window length and γ is the number of sampling windows.
Then, performing singular value decomposition on the matrix a, a = M Σ N, and obtaining:
wherein M (t) = [ M = 1 ,…,m ω ]，
Is AA T The eigenvalues of which are the squares of the singular values of the matrix a.
S103, calculating difference statistics between the historical data and the current data, and evaluating the change degree between the current data and the historical data.
And selecting a characteristic vector with a larger singular value in the basic matrix, constructing to obtain a basic characteristic matrix corresponding to the main characteristics of the historical data, and then comparing the variation degree between the characteristic vector of the test matrix and the basic characteristic matrix to obtain the variation degree between the current data and the historical data.
The calculating difference statistics between the historical data and the current data comprises:
the first l (l) are selected according to the arrangement of the lambda from large to small<Omega) to obtain a basic feature matrix U l ；
According to the following
Arranging from large to small, selecting the first eta pieces (eta)<γ) corresponding feature vector m η (ii) a By calculating m η At U l Projection and m η The cosine value of the included angle therebetween, the corresponding change coefficient cp is obtained, namely:
m η at U l The projection vector is:
then the coefficient of variation can be found:
meanwhile, different weights are calculated according to singular values of the test matrix to obtain a final change coefficient CP, namely
And S104, evaluating the change degree of the comprehensive electric energy data to obtain the judgment of the change of the electricity utilization behavior of the user.
And selecting a plurality of user electric energy data time sequences for analysis, respectively calculating the change degrees of the different electric energy data time sequences, and comprehensively obtaining the change degrees of the various user electric energy data time sequences so as to obtain a judgment conclusion of the change of the user electricity consumption behavior.
And performing singular spectrum analysis on the rest electric energy data to obtain corresponding change degree evaluation. Acquiring other types of electric energy data of a user, such as a current indicating value curve I (t) and a power curve W (t), repeating S101, S102 and S103 to obtain a change coefficient CP corresponding to the current indicating value curve and the power curve I 、CP W And when the change coefficient is larger than 0.5, the electricity utilization behavior of the user is considered to be changed.
A power consumption behavior change detection method based on singular spectrum analysis is characterized in that a track matrix of a user electric energy data time sequence is constructed, singular spectrum decomposition is carried out, main characteristic information in the time sequence is extracted, and accurate judgment of power consumption behavior change is finally achieved by comparing change degrees between main characteristics.
Fig. 2 and fig. 2 are schematic structural diagrams of a power utilization behavior change detection system based on singular spectrum analysis.
As shown in fig. 2, a system for detecting a change in power consumption behavior of a user based on singular spectrum analysis includes:
the preprocessing module 201: the system is used for acquiring the electric energy data of the electric energy metering device and preprocessing the electric energy data;
the data processing module 202: the method is used for carrying out matrix construction and singular value decomposition on historical electric energy data and current data through a singular spectrum analysis method to obtain corresponding singular values and characteristic vectors and obtain a change mode of the historical data and a change mode of the current data;
the comparative analysis module 203: the system is used for calculating difference statistics between historical data and current data and evaluating the change degree between the current data and the historical data;
the comprehensive evaluation module 204: and the system is used for carrying out singular spectrum analysis on the rest electric energy data to obtain corresponding change degree evaluation, and integrating the change degree evaluation of the electric energy data of each category to obtain the judgment of the power utilization behavior change of the user.
The preprocessing module 201 comprises a 0-point freezing table code, a 96-point-per-day electric energy indicating value curve, a 96-point-per-day current curve and a 96-point-per-day power curve; the preprocessing comprises eliminating and fitting null values and maximum and minimum values of the electric energy data;
the singular spectrum analysis method 202 is to obtain a basic trajectory matrix of the data, called basic matrix for short, by using a time delay embedding method for the data, and then perform singular value decomposition on the basic matrix to obtain corresponding singular values and eigenvectors;
the comparative analysis module 203 comprises: selecting a characteristic vector with a large singular value in a basic matrix, constructing to obtain a basic characteristic matrix corresponding to main characteristics of historical data, and then obtaining the change degree between current data and historical data by comparing the change degree between the characteristic vector of a test matrix and the basic characteristic matrix;
the comprehensive evaluation module 204 includes: and selecting a plurality of user electric energy data time sequences for analysis, respectively calculating the change degrees of the different electric energy data time sequences, and comprehensively obtaining the change degrees of the various user electric energy data time sequences so as to obtain a judgment conclusion of the change of the user electricity consumption behavior.
The invention provides a power utilization behavior change detection method and system based on singular spectrum analysis. And comparing the change degree between the characteristic vectors of the electric energy data before and after a certain time point to obtain the change evaluation of the electric energy data. And (4) comprehensively evaluating the change of different types of electric energy data of the user to obtain a judgment conclusion of the change of the electricity utilization behavior of the user. The method can accurately identify the components with different change characteristics and high significance degree in the power consumption behavior of the user by using a singular spectrum analysis method, thereby realizing the accurate identification of the different power consumption characteristics before and after the change and effectively judging the change trend of the power consumption behavior of the user.
Those skilled in the art will appreciate that all or part of the steps in the methods of the above embodiments may be implemented by hardware related to instructions of a program, and the program may be stored in a computer-readable storage medium, and the storage medium may include: a Read Only Memory (ROM), a Random Access Memory (RAM), a magnetic or optical disk, or the like.
In addition, the power utilization behavior change detection method and system based on singular spectrum analysis provided by the embodiment of the invention are described in detail, a specific embodiment is adopted to explain the principle and the implementation mode of the invention, and the description of the embodiment is only used for helping to understand the method and the core idea of the invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed, and in summary, the content of the present specification should not be construed as a limitation to the present invention.
Claims (6)
1. A user electricity consumption behavior change detection method based on singular spectrum analysis is characterized by comprising the following steps:
acquiring electric energy data of the electric energy metering device, and preprocessing the electric energy data;
the power data includes: freezing table codes at 0 point, an electric energy indicating curve at 96 points per day, a current curve at 96 points per day and a power curve at 96 points per day;
the preprocessing comprises the step of eliminating and fitting null values, maximum values and minimum values of the electric energy data;
performing matrix construction and singular value decomposition on historical electric energy data and current data through a singular spectrum analysis method to obtain corresponding singular values and eigenvectors, and obtaining a change mode of the historical data and a change mode of the current data;
calculating difference statistics between the historical data and the current data, and evaluating the change degree between the current data and the historical data;
comparing the change degree between the characteristic vectors of the electric energy data before and after a certain time point to obtain the change evaluation of the electric energy data; the change evaluation of different types of electric energy data of the user is integrated to obtain a judgment conclusion of the change of the electricity utilization behavior of the user;
the calculating difference statistics between the historical data and the current data comprises:
according to the arrangement of lambda from large to small, selectingSelecting the first l corresponding eigenvectors U, wherein the number l of the selected eigenvectors is not more than the number delta of the sampling windows of the basic trajectory matrix B, and obtaining the basic characteristic matrix U l (ii) a Wherein λ is the square of the singular value of the base trajectory matrix B;
according to
Arranging from large to small, selecting the first eta corresponding eigenvectors m η (ii) a The number eta of the selected characteristic vectors is not more than the number gamma of sampling windows of the test track matrix A, wherein,
the square of the singular value of the test trajectory matrix A;
by calculating m η At U l Projection and m η The cosine value of the included angle therebetween, the corresponding change coefficient CP is obtained, namely:
m η at U l The projection is as follows:
then the coefficient of variation can be found:
at the same time, different weights are calculated according to the singular values of the test matrix to obtain the final change coefficient CP, that is
2. The method for detecting the change of the power consumption behaviors of the user according to claim 1, wherein the singular spectrum analysis method comprises the following steps:
and obtaining a basic matrix of the data by adopting a time delay embedding method, and then performing singular value decomposition on the basic matrix to obtain corresponding singular values and eigenvectors.
3. The method for detecting the change of the electricity consumption behavior of the user according to claim 1, wherein the singular spectrum analysis method comprises:
constructing a track matrix of the electric energy data P (t) time sequence before t = i, and performing singular value decomposition to obtain corresponding singular values and characteristic vectors and obtain a change mode of historical data;
obtaining a basic track matrix of the time sequence of the electric energy data P (t) before a time point i by a time delay embedding method, namely:
wherein t = i, ω is the window length, δ is the number of sampling windows;
performing singular value decomposition on the basic track matrix B to obtain singular values and singular vectors, namely:
B＝USV T ；
wherein U is left singular matrix, V is right singular matrix, S is singular value matrix, its diagonal angle element is singular value called matrix, and is marked as sigma, sigma i Is the ith singular value;
solving matrix BB T The eigenvalues and eigenvectors of (a) result in corresponding U, i.e.:
(BB T )u i ＝λ i u i ；
wherein U (t) = [ U = 1 ,…,u i ]，
Will be lambda i Arranged from large to small, lambda i Is the ith characteristic value, then corresponding u i The main features of the P (t) time series before the time point i are shown.
4. The method for detecting changes in user electricity consumption behaviors of claim 3, wherein the singular spectrum analysis method comprises:
carrying out track matrix construction on the electric energy data P (t) time sequence after t = i and carrying out singular value decomposition to obtain corresponding singular values and characteristic vectors and obtain a change mode of current data;
obtaining a test track matrix of the P (t) time sequence after a time point i by a time delay embedding method, namely:
where t = i, ω is the window length and γ is the number of sampling windows;
then, performing singular value decomposition on the test track matrix a, wherein a = M Σ N, and obtaining the following result in the same way:
wherein M (t) = [ M = 1 ,…,m ω ]，
Is AA T The eigenvalue of (a) is the square of the singular value of the test trajectory matrix a.
5. A user power consumption behavior change detection system based on singular spectrum analysis is characterized in that the user power consumption behavior change detection system comprises:
a preprocessing module: the electric energy data acquisition unit is used for acquiring electric energy data of the electric energy metering device and carrying out preprocessing;
the power data includes: freezing table codes at 0 point, an electric energy indicating curve at 96 points per day, a current curve at 96 points per day and a power curve at 96 points per day;
the preprocessing comprises the step of eliminating and fitting null values, maximum values and minimum values of the electric energy data;
a data processing module: the method is used for carrying out matrix construction and singular value decomposition on historical electric energy data and current data through a singular spectrum analysis method to obtain corresponding singular values and characteristic vectors and obtain a change mode of the historical data and a change mode of the current data;
a comparison analysis module: the system is used for calculating difference statistics between the historical data and the current data and evaluating the change degree between the current data and the historical data;
a comprehensive evaluation module: the method is used for comparing the change degree between the characteristic vectors of the electric energy data before and after a certain time point so as to obtain the change evaluation of the electric energy data; the change evaluation of different types of electric energy data of the user is integrated to obtain a judgment conclusion of the change of the electricity utilization behavior of the user;
the calculating of difference statistics between historical data and current data includes:
according to the arrangement of lambda from large to small, the first l corresponding eigenvectors U are selected, the number l of the selected eigenvectors is not more than the number delta of sampling windows of the basic track matrix B, and the basic eigenvector U is obtained l (ii) a Wherein λ is the square of the singular value of the base trajectory matrix B;
according to
Arranging from big to small, selecting the first eta corresponding eigenvectors m η (ii) a The number η of the selected feature vectors is not more than the number γ of sampling windows of the test trajectory matrix a, wherein,
the square of the singular value of the test trajectory matrix A;
by calculating m η At U l Projection and m η The cosine value of the included angle therebetween, the corresponding change coefficient CP is obtained, namely:
m η at U l The projection is as follows:
then the coefficient of variation is obtained:
meanwhile, different weights are calculated according to singular values of the test matrix to obtain a final change coefficient CP, namely
6. The system according to claim 5, wherein the data processing module obtains a basis matrix of the data by using a singular spectrum analysis method and using a time delay embedding method for the data, and then performs singular value decomposition on the basis matrix to obtain corresponding singular values and eigenvectors.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011105301.2A CN112381351B (en) | 2020-10-15 | 2020-10-15 | Power utilization behavior change detection method and system based on singular spectrum analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011105301.2A CN112381351B (en) | 2020-10-15 | 2020-10-15 | Power utilization behavior change detection method and system based on singular spectrum analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112381351A CN112381351A (en) | 2021-02-19 |
| CN112381351B true CN112381351B (en) | 2022-10-11 |
Family
ID=74581601
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011105301.2A Active CN112381351B (en) | 2020-10-15 | 2020-10-15 | Power utilization behavior change detection method and system based on singular spectrum analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112381351B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113255960A (en) * | 2021-04-14 | 2021-08-13 | 北京市燃气集团有限责任公司 | Gas consumption prediction method and device based on ISSA-SVR fusion |
| CN113378063B (en) * | 2021-07-09 | 2023-07-28 | 小红书科技有限公司 | Method for determining content diversity based on sliding spectrum decomposition and content sorting method |
| CN113792766A (en) * | 2021-08-26 | 2021-12-14 | 广西电网有限责任公司 | Singular spectrum analysis-based line-variable relation identification method |
| CN115600061B (en) * | 2022-12-14 | 2023-03-28 | 嘉兴索罗威新能源有限公司 | Inverter zero voltage drop data processing method based on machine learning |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103077402B (en) * | 2012-12-28 | 2016-05-11 | 国家电网公司 | Partial discharge of transformer mode identification method based on singular value decomposition algorithm |
| CN107024694A (en) * | 2017-05-19 | 2017-08-08 | 武汉大学 | Anomalous of the ionosphere detection method and system based on singular spectrum analysis |
| CN109345013A (en) * | 2018-09-25 | 2019-02-15 | 广西电网有限责任公司电力科学研究院 | A kind of distribution user power utilization behavior prediction method |
| CN111311292B (en) * | 2018-12-12 | 2023-08-04 | 北京嘀嘀无限科技发展有限公司 | User classification method and system |
| US11493911B2 (en) * | 2018-12-27 | 2022-11-08 | Utopus Insights, Inc. | System and method for proactive handling of multiple faults and failure modes in an electrical network of energy assets |
| CN109993368A (en) * | 2019-04-08 | 2019-07-09 | 福州大学 | Power forecasting method based on unusual spectral factorization and shot and long term memory network |
| CN111103497A (en) * | 2019-11-05 | 2020-05-05 | 深圳供电局有限公司 | Device and method for detecting abnormal electricity utilization of user |
| CN111062502B (en) * | 2019-12-17 | 2023-07-14 | 国网湖南省电力有限公司 | User electricity consumption behavior subdivision method and fault analysis method thereof |
-
2020
- 2020-10-15 CN CN202011105301.2A patent/CN112381351B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN112381351A (en) | 2021-02-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112381351B (en) | Power utilization behavior change detection method and system based on singular spectrum analysis | |
| CN112699913B (en) | Method and device for diagnosing abnormal relationship of household transformer in transformer area | |
| CN115858630B (en) | Abnormality detection method for energy storage data of energy storage power station | |
| CN107885928B (en) | Stepping stress acceleration performance degradation reliability analysis method considering measurement error | |
| CN106067034B (en) | Power distribution network load curve clustering method based on high-dimensional matrix characteristic root | |
| CN108921424B (en) | Power data anomaly detection method, device, equipment and readable storage medium | |
| CN112308124B (en) | Intelligent electricity larceny prevention method for electricity consumption information acquisition system | |
| CN112152201A (en) | Electricity load prediction method and system based on convolution length time memory neural network | |
| CN110672905A (en) | CNN-based self-supervision voltage sag source identification method | |
| CN116660759B (en) | Battery life prediction method and device based on BMS battery management system | |
| CN116739829B (en) | Big data-based power data analysis method, system and medium | |
| CN115686124B (en) | Energy storage battery output power self-adjusting system and method based on safety protection | |
| Chen et al. | Cluster-based profile analysis in phase I | |
| CN112816881A (en) | Battery differential pressure abnormality detection method, battery differential pressure abnormality detection device and computer storage medium | |
| CN113533995B (en) | Consistency detection method for power battery | |
| CN114881157A (en) | Method, device and equipment for detecting working state of converter valve and storage medium | |
| CN112595918A (en) | Low-voltage meter reading fault detection method and device | |
| CN117154263A (en) | Lithium battery cascade utilization charging and discharging system and control method | |
| CN112101673A (en) | Power grid development trend prediction method and system based on hidden Markov model | |
| CN116520068A (en) | Diagnostic method, device, equipment and storage medium for electric power data | |
| CN116611003A (en) | Transformer fault diagnosis method, device and medium | |
| CN110991741A (en) | Section constraint probability early warning method and system based on deep learning | |
| CN115980588A (en) | Lithium ion battery health state estimation method based on self-encoder extraction features | |
| CN111505445B (en) | Credibility detection method and device for mutual-user relationship of transformer area and computer equipment | |
| CN110187167B (en) | Method and device for detecting load switch event based on manifold classification |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |