CN110910028A - Photovoltaic electricity stealing discovery method and system based on time characteristic analysis - Google Patents
Photovoltaic electricity stealing discovery method and system based on time characteristic analysis Download PDFInfo
- Publication number
- CN110910028A CN110910028A CN201911192256.6A CN201911192256A CN110910028A CN 110910028 A CN110910028 A CN 110910028A CN 201911192256 A CN201911192256 A CN 201911192256A CN 110910028 A CN110910028 A CN 110910028A
- Authority
- CN
- China
- Prior art keywords
- user
- time
- sample user
- time measuring
- measuring point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000005611 electricity Effects 0.000 title claims abstract description 73
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000005259 measurement Methods 0.000 claims abstract description 48
- 238000010248 power generation Methods 0.000 claims abstract description 41
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 37
- 238000007621 cluster analysis Methods 0.000 claims abstract description 21
- 238000005286 illumination Methods 0.000 claims description 12
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000000630 rising effect Effects 0.000 claims description 3
- 230000002123 temporal effect Effects 0.000 claims 6
- 230000006399 behavior Effects 0.000 abstract description 11
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000000605 extraction Methods 0.000 abstract description 2
- 238000010586 diagram Methods 0.000 description 3
- 241000220317 Rosa Species 0.000 description 1
- 230000007423 decrease Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 239000004576 sand Substances 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
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
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- 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
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- 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
-
- 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
- Y02E—REDUCTION OF GREENHOUSE GAS [GHG] EMISSIONS, RELATED TO ENERGY GENERATION, TRANSMISSION OR DISTRIBUTION
- Y02E40/00—Technologies for an efficient electrical power generation, transmission or distribution
- Y02E40/70—Smart grids as climate change mitigation technology in the energy generation sector
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Health & Medical Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Artificial Intelligence (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Primary Health Care (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Water Supply & Treatment (AREA)
- Public Health (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- General Engineering & Computer Science (AREA)
- Game Theory and Decision Science (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a photovoltaic electricity stealing discovery method and a photovoltaic electricity stealing discovery system based on time characteristic analysis, wherein the method comprises the following steps: calculating key time measurement points: the key time measuring points comprise a time measuring point of output improvement, a time measuring point of output stability and a time measuring point of output reduction in an output curve of a sample user; or the key time measuring points comprise a time measuring point of output improvement, a time measuring point of highest output and a time measuring point of output reduction in the output curve of the sample user; calculating the daily power generation change proportion; performing cluster analysis: and performing cluster analysis on the sample users by using an HCM algorithm according to the daily power generation amount change proportion and the key time measuring point, and confirming the electricity stealing users. The invention provides an effective, practical and scientific photovoltaic curve characteristic extraction method, which is suitable for finding photovoltaic electricity stealing behaviors, is beneficial to reducing the economic loss of a power grid and improving the social fairness.
Description
Technical Field
The invention belongs to the technical field of power systems, and particularly relates to a photovoltaic electricity stealing discovery method and system based on time characteristic analysis.
Background
Because subsidies enjoyed by distributed photovoltaic power generation mainly depend on self power generation amount, certain users enable the distributed photovoltaic on-line electricity meters to measure more power generation amount through certain technical means under the drive of benefits, and further obtain the risk of high-volume subsidies, and the behavior of cheating the subsidies is called as photovoltaic electricity stealing behavior. The behavior of cheating and subsidizing photovoltaic electricity stealing seriously affects the implementation of Chinese new energy planting policies, the fairness of the power generation market, huge potential safety hazards are brought to power supply and distribution due to private line switching of users due to electricity stealing, and the normal development of the photovoltaic power generation industry is affected. At present, no perfect, practical and low-cost photovoltaic electricity stealing behavior discovery algorithm exists in the industry.
Disclosure of Invention
Aiming at the defects of the prior art, the invention discloses a photovoltaic electricity stealing discovery method and a photovoltaic electricity stealing discovery system based on time characteristic analysis.
The technical scheme adopted by the invention is as follows:
a photovoltaic electricity stealing discovery method based on time characteristic analysis is characterized by comprising the following steps:
calculating key time measurement points: the key time measuring points comprise a time measuring point of output improvement, a time measuring point of output stability and a time measuring point of output reduction in an output curve of a sample user; or the key time measuring points comprise a time measuring point of output improvement, a time measuring point of highest output and a time measuring point of output reduction in the output curve of the sample user;
calculating the daily power generation change proportion;
performing cluster analysis: and performing cluster analysis on the sample users by using an HCM algorithm according to the daily power generation amount change proportion and the key time measuring point, and confirming the electricity stealing users.
The further technical scheme is that the key time measuring points specifically comprise: a first time measuring point t of the sample user, wherein the difference value of the load of the sample user at two adjacent time measuring points is higher than a load change threshold value0(ii) a Second time measurement point t at which the load rise of a sample user is first below a load change threshold1(ii) a Third time measurement point t at which load drop of sample user is first higher than load change threshold2(ii) a The fourth time measuring point t when the load reduction amount of the sample user is lower than the load change threshold value for the first time3。
The further technical scheme is that the key time measuring points specifically comprise: a first time measuring point t of the sample user, wherein the difference value of the load of the sample user at two adjacent time measuring points is higher than a load change threshold value0(ii) a Second time measuring point t for changing load of sample user from rising to falling1(ii) a Third time measurement point at which load drop amount of sample user is first lower than load change thresholdt3。
The further technical scheme is that the method for carrying out cluster analysis on the sample users comprises the following steps:
judging the relation between the difference between the daily power generation change proportion of the sample user and the clustering center of the sample user and the load change threshold of the sample user, and carrying out first clustering analysis on the sample user by taking the daily power generation change proportion as a parameter;
performing cluster analysis with the key time measurement points as parameters:
set S of users to be examined after first clustering analysisuThe sample user in (1), defining a first distance dt01The following were used:
dt01=|t0+t1-ct0-ct1|
ct0to sunrise time, ct1Performing second clustering analysis on the sample by using an HCM algorithm on the basis of the first distance for the time when the illumination intensity reaches the output limit of the equipment for the first time;
set S of users to be examined after second clustering analysisuThe sample user in (1), defining a second distance dt23The following were used:
dt23=|t2+t3-ct2-ct3|
ct2the time when the illumination intensity is weaker than the output limit of the equipment for the first time in the middle and later days, ct3And performing third clustering analysis on the samples by using an HCM algorithm on the basis of the second distance for the sunset time to confirm the electricity stealing users.
The further technical scheme is that the specific method of the first clustering analysis is as follows:
if △ -c△≥δ△Then the corresponding sample user is charged into the electricity stealing user set Ss;
If c is△-△≥δ△If the corresponding sample user is counted as the failure user set Sb;
The residual samples are counted into a user set S to be examinedu;
δ△Is the load change threshold of the sample user; c. C△Is the clustering center of the sample user, and △ is the daily generated energy change proportion of the sample user.
The further technical scheme is that the daily power generation change proportion △ of the photovoltaic sample user is as follows:
in the formula, sum (E)p) Is the total generating capacity of photovoltaic users on the day, sum (E'p) Is the total power generation amount of photovoltaic users in the previous day, EpIs the generated energy of the photovoltaic user at the time of the day measurement point p, Ep' is the power generation amount of the photovoltaic user at the previous day time measurement point p.
The further technical scheme is that on the basis of the first distance, the HCM algorithm is utilized to perform the second clustering analysis on the samples, and the specific method comprises the following steps: will dt01Set to the set of users to be examined S as 0uCluster center of (d)t01=ct0+ct1Set to failed user set SbCluster center of (d)t01=2ct0+2ct1Set as electricity stealing subscriber set SsThe cluster center of (a); if the sample user is distant from the electricity stealing user set SsWhen the cluster center is nearest, the sample user is counted as the electricity stealing user set Ss(ii) a If the sample user is far from the faulty user set SbWhen the cluster center is nearest, the sample user is counted as a fault user set Sb(ii) a The residual samples are counted into a user set S to be examinedu。
The further technical scheme is that on the basis of the second distance, the HCM algorithm is utilized to carry out the third clustering analysis on the samples, and the specific method comprises the following steps: will dt23Set to 0 as the normal user set SnCluster center of (d)t23=ct2+ct3Set to failed user set SbCluster center of (d)t23=2ct2+2ct3Set as electricity stealing subscriber set SsThe cluster center of (2). If the sample user is distant from the electricity stealing user set SsCluster center of (2)In the near, the sample user is counted as the electricity stealing user set Ss(ii) a If the sample user is far from the faulty user set SbWhen the cluster center is nearest, the sample user is counted as a fault user set Sb(ii) a The remaining samples are counted in the normal user set Sn。
A photovoltaic electricity stealing discovery system based on time feature analysis comprises:
a critical time measurement point calculation module: the key time measuring points comprise a time measuring point of output improvement, a time measuring point of output stability and a time measuring point of output reduction in an output curve of a sample user; or the key time measuring points comprise a time measuring point of output improvement, a time measuring point of highest output and a time measuring point of output reduction in the output curve of the sample user;
the daily generated energy change proportion calculation module: calculating the daily power generation change proportion of two adjacent days of the sample user;
HCM algorithm cluster analysis module: and performing cluster analysis on the sample users by using an HCM algorithm according to the daily power generation amount change proportion and the key time measuring point, and confirming the electricity stealing users.
The further technical scheme is that the HCM algorithm module comprises:
a first clustering analysis module: judging the relation between the difference between the daily power generation change proportion of the sample user and the clustering center of the sample user and the load change threshold of the sample user, and carrying out first clustering analysis on the sample user;
a second clustering analysis module: set S of users to be examined after first clustering analysisuThe sample user in (1), defining a first distance dt01The following were used:
dt01=|t0+t1-ct0-ct1|
ct0to sunrise time, ct1Performing second clustering analysis on the sample by using an HCM algorithm on the basis of the first distance for the time when the illumination intensity reaches the output limit of the equipment for the first time;
a third clustering analysis module: to-be-examined after the second clustering analysisSet of scout users SuThe sample user in (1), defining a second distance dt23The following were used:
dt23=|t2+t3-ct2-ct3|
ct2the time when the illumination intensity is weaker than the output limit of the equipment for the first time in the middle and later days, ct3And performing third clustering analysis on the samples by using an HCM algorithm on the basis of the second distance for the sunset time to confirm the electricity stealing users.
The invention has the following beneficial effects:
the invention designs a photovoltaic electricity stealing behavior discovery algorithm, and discovers an electricity stealing behavior by calculating a daily generated energy change proportion parameter and a key time measuring point parameter of a photovoltaic user and carrying out cluster analysis by utilizing an HCM algorithm, and belongs to the first time in the industry.
The method fully excavates the data information of the photovoltaic curve, has less dependence on other information of equipment and users, and has higher universality.
The invention provides an effective, practical and scientific photovoltaic curve characteristic extraction method, which is suitable for finding photovoltaic electricity stealing behaviors, is beneficial to reducing the economic loss of a power grid and improving the social fairness.
Drawings
Fig. 1 is a schematic flow chart of embodiment 1 of the present invention.
Fig. 2 is a schematic flow chart of embodiment 3 of the present invention.
Fig. 3 is a schematic diagram of the key time measurement points in embodiment 3 of the present invention.
Fig. 4 is a schematic structural framework diagram of embodiment 5 of the present invention.
Detailed Description
The method comprises the steps of taking a day as a unit, and extracting time characteristics of a photovoltaic output curve of a sample user; and comparing the photovoltaic curves of two adjacent days to extract the change characteristics of the daily generated energy. And selecting standard photovoltaic users, analyzing each index by adopting an HCM algorithm, and finding users suspected of stealing electricity.
The following describes a specific embodiment of the present embodiment with reference to the drawings.
Example 1.
Fig. 1 is a schematic flow chart of embodiment 1 of the present invention. As shown in fig. 1, a photovoltaic electricity stealing discovery method based on time characteristic analysis includes:
s101, calculating a key time measuring point; the key time measuring points comprise a time measuring point of output improvement, a time measuring point of output stability and a time measuring point of output reduction in the output curve of the sample user. Based on this, the key time measurement points specifically include:
first time measuring point t0: the time measurement point when the difference between the load of the sample user measured at two adjacent time measurement points is higher than the load change threshold value, that is, the time measurement point when the difference between the load measured by the sample user at one time measurement point and the load measured at the previous time measurement point adjacent to the time measurement point is higher than the load change threshold value.
Second time measuring point t1: measuring a point t at a first time0Thereafter, the time measurement point at which the load rise of the sample user is first below the load change threshold.
Third time measuring point t2: measuring point t at a second time1And then, the load reduction amount of the sample user is firstly higher than the time measurement point of the load change threshold value.
Fourth time measuring point t3: measuring point t at a third time2Thereafter, the load drop amount of the sample user is first lower than the time measurement point of the load change threshold.
S102, calculating the daily power generation change proportion;
s103, performing clustering analysis by using an HCM algorithm: and performing cluster analysis on the sample users by using an HCM algorithm according to the daily power generation amount change proportion and the key time measuring point, and confirming the electricity stealing users.
Example 2.
In embodiment 2, a photovoltaic electricity stealing discovery method based on time feature analysis includes:
s201, calculating a key time measuring point: the key time measuring points comprise a time measuring point of output improvement, a time measuring point of highest output and a time measuring point of output reduction in an output curve of a sample user; based on this, the key time measurement points specifically include:
first time measuring point t0: the time measurement point when the difference between the load of the sample user measured at two adjacent time measurement points is higher than the load change threshold value, that is, the time measurement point when the difference between the load measured by the sample user at one time measurement point and the load measured at the previous time measurement point adjacent to the time measurement point is higher than the load change threshold value.
Second time measuring point t1: the point in time at which the load of the sample user changes from rising to falling is measured. Measuring a point t at a first time0Thereafter, if there is a second time measurement point t1Then, at a second time, measure point t1Previously, the load of the sample user rose; second time measuring point t1Thereafter, the load of the sample user decreases.
Third time measuring point t3: measuring point t at a second time1Thereafter, the load drop amount of the sample user is first lower than the time measurement point of the load change threshold.
S202, calculating the daily power generation change proportion;
s203, clustering analysis is carried out by using an HCM algorithm: and performing cluster analysis on the sample users by using an HCM algorithm according to the daily power generation amount change proportion and the key time measuring point, and confirming the electricity stealing users.
Example 3.
Fig. 2 is a schematic flow chart of embodiment 3 of the present invention. As shown in fig. 2, embodiment 3 includes:
s301, calculating a key time measuring point; the key time measuring points comprise a time measuring point of output improvement, a time measuring point of output stability and a time measuring point of output reduction in the output curve of the sample user. Fig. 3 is a graph showing the output curves of example 3 of the present invention. In fig. 3, the abscissa is the time measurement point number and the ordinate is the output value. As shown in fig. 3, the critical time measurement points include:
first time measuring point t0: of sample usersThe time measuring point when the difference value of the load between two adjacent time measuring points is higher than the load change threshold value, that is, the time measuring point when the difference value of the load measured by the sample user at one time measuring point and the load measured by the adjacent previous time measuring point before the time measuring point is higher than the load change threshold value.
Second time measuring point t1: measuring a point t at a first time0Thereafter, the time measurement point at which the load rise of the sample user is first below the load change threshold.
Third time measuring point t2: measuring point t at a second time1And then, the load reduction amount of the sample user is firstly higher than the time measurement point of the load change threshold value.
Fourth time measuring point t3: measuring point t at a third time2Thereafter, the load drop amount of the sample user is first lower than the time measurement point of the load change threshold.
S302, calculating the daily power generation change proportion △ of the photovoltaic users is as follows:
in the formula, sum (E)p) Is the total generating capacity of photovoltaic users on the day, sum (E'p) The total power generation amount of the photovoltaic users in the previous day. E, EpIs the generated energy of the photovoltaic user at the time of the day measurement point p, Ep' is the power generation amount of the photovoltaic user at the previous day time measurement point p.
S303, carrying out first clustering analysis on the sample users by taking the daily generated energy change proportion as a parameter:
if △ -c△≥δ△Then the corresponding sample user is charged into the electricity stealing user set Ss;
If c is△-△≥δ△If the corresponding sample user is counted as the failure user set Sb;
The residual samples are counted into a user set S to be examinedu;
△ is the daily power generation change proportion of the sample user;δ△is the load change threshold of the sample user; c. C△Is the cluster center of the sample user.
S304, collecting the users to be examined after the first clustering analysis SuThe sample user in (1), defining a first distance dt01The following were used:
dt01=|t0+t1-ct0-ct1|
ct0to sunrise time, ct1Performing second clustering analysis on the sample by using an HCM algorithm on the basis of the first distance for the time when the illumination intensity reaches the output limit of the equipment for the first time; respectively recording the users corresponding to the samples into a user set S to be examineduSet of faulty users SbAnd a set of electricity stealing subscribers Ss。
Based on the first distance, the HCM algorithm is utilized to perform the second clustering analysis on the samples, and the specific method comprises the following steps: will dt01Set to the set of users to be examined S as 0uCluster center of (d)t01=ct0+ct1Set to failed user set SbCluster center of (d)t01=2ct0+2ct1Set as electricity stealing subscriber set SsThe cluster center of (a); if the first distance parameter of the sample user is distant from the electricity stealing user set SsWhen the cluster center is nearest, the sample user is counted as the electricity stealing user set Ss(ii) a If the first distance parameter of the sample user is far from the fault user set SbWhen the cluster center is nearest, the sample user is counted as a fault user set Sb(ii) a The residual samples are counted into a user set S to be examinedu。
S305, collecting the users to be examined after the second clustering analysis SuThe sample user in (1), defining a second distance dt23The following were used:
dt23=|t2+t3-ct2-ct3|
ct2the time when the illumination intensity is weaker than the output limit of the equipment for the first time in the middle and later days, ct3Performing third clustering analysis on the sample by using HCM algorithm based on the second distance for sunset timeRespectively recording the users corresponding to the samples into a normal user set SnSet of faulty users SbAnd a set of electricity stealing subscribers SsAnd confirming the electricity stealing user.
Based on the second distance, the HCM algorithm is utilized to carry out the third clustering analysis on the samples, and the specific method comprises the following steps: will dt23Set to 0 as the normal user set SnCluster center of (d)t23=ct2+ct3Set to failed user set SbCluster center of (d)t23=2ct2+2ct3Set as electricity stealing subscriber set SsThe cluster center of (2). If the second distance parameter of the sample user is distant from the electricity stealing user set SsWhen the cluster center is nearest, the sample user is counted as the electricity stealing user set Ss(ii) a If the second distance parameter of the sample user is far from the fault user set SbWhen the cluster center is nearest, the sample user is counted as a fault user set Sb(ii) a The remaining samples are counted in the normal user set Sn。
Step S304 and step S305 are cluster analysis using the key time measurement point as a parameter. And selecting a credible photovoltaic user, and taking the output curve of the photovoltaic user as a clustering center.
Example 4.
Example 4 is based on the procedure of example 3, and a practical specific operation procedure was combined once. Embodiment 4 targets 236 photovoltaic users in Zhejiang province near the sea city as a search target. The daily sampling frequency of the 236 photovoltaic users is 288 points. Among the users, one user has a long-term cooperative relationship with the electric power supply company in the near-sea city and belongs to a credible photovoltaic user, so that the index of the user is set as a clustering center.
And (4) analyzing the data of 7 months in 2019, and extracting a daily power generation amount change proportion parameter and a key time measuring point parameter of each sample. And analyzing the daily generated energy change proportion parameter and the key time measuring point parameter in sequence.
And in the step of carrying out clustering analysis according to the daily generated electricity quantity change parameters, finding that the suspected electricity stealing behavior exists in the 2 households. And (3) performing a clustering analysis link by taking the key time measuring point as a parameter to find that the suspected electricity stealing behavior exists in the user 3. And 5, confirming that the electricity stealing behavior of the household is true through the verification of the staff at home.
Example 5.
Fig. 4 is a schematic structural frame diagram of embodiment 5. As shown in fig. 4, the photovoltaic electricity stealing discovery system based on the time characteristic analysis includes:
a critical time measurement point calculation module: the key time measuring points comprise a time measuring point of output improvement, a time measuring point of output stability and a time measuring point of output reduction in an output curve of a sample user; or the key time measuring points comprise a time measuring point of output improvement, a time measuring point of highest output and a time measuring point of output reduction in the output curve of the sample user;
a daily power generation change proportion calculation module; the method is used for calculating the daily power generation change proportion of two adjacent days of the photovoltaic user.
A cluster analysis module: and performing cluster analysis on the sample users by using an HCM algorithm according to the daily power generation amount change proportion and the key time measuring point, and confirming the electricity stealing users.
Example 6.
Embodiment 6 is based on embodiment 5, and it is further clear that the cluster analysis algorithm module includes:
a first clustering analysis module: and judging the relation between the difference between the daily power generation change proportion of the sample user and the clustering center of the sample user and the load change threshold of the sample user, and carrying out first clustering analysis on the sample user by taking the daily power generation change proportion as a parameter.
A second clustering analysis module: set S of users to be examined after first clustering analysisuThe sample user in (1), defining a first distance dt01The following were used:
dt01=|t0+t1-ct0-ct1|
ct0to sunrise time, ct1Performing second clustering analysis on the sample by using an HCM algorithm on the basis of the first distance for the time when the illumination intensity reaches the output limit of the equipment for the first time;
a third clustering analysis module: set S of users to be examined after second clustering analysisuThe sample user in (1), defining a second distance dt23The following were used:
dt23=|t2+t3-ct2-ct3|
ct2the time when the illumination intensity is weaker than the output limit of the equipment for the first time in the middle and later days, ct3And performing third clustering analysis on the samples by using an HCM algorithm on the basis of the second distance for the sunset time to confirm the electricity stealing users.
The second clustering analysis module and the third clustering analysis module are used for clustering analysis with key time measuring points as parameters.
The foregoing description is illustrative of the present invention and is not to be construed as limiting thereof, the scope of the invention being defined by the appended claims, which may be modified in any manner without departing from the basic structure thereof.
Claims (10)
1. A photovoltaic electricity stealing discovery method based on time characteristic analysis is characterized by comprising the following steps:
calculating key time measurement points: the key time measuring points comprise a time measuring point of output improvement, a time measuring point of output stability and a time measuring point of output reduction in an output curve of a sample user; or the key time measuring points comprise a time measuring point of output improvement, a time measuring point of highest output and a time measuring point of output reduction in the output curve of the sample user;
calculating the daily power generation change proportion;
performing cluster analysis: and performing cluster analysis on the sample users by using an HCM algorithm according to the daily power generation amount change proportion and the key time measuring point, and confirming the electricity stealing users.
2. The photovoltaic electricity stealing discovery method based on time characteristic analysis according to claim 1, wherein the key time measurement points specifically include: sample user load isA first time measuring point t with the difference value of two adjacent time measuring points higher than a load change threshold value0(ii) a Second time measurement point t at which the load rise of a sample user is first below a load change threshold1(ii) a Third time measurement point t at which load drop of sample user is first higher than load change threshold2(ii) a The fourth time measuring point t when the load reduction amount of the sample user is lower than the load change threshold value for the first time3。
3. The photovoltaic electricity stealing discovery method based on temporal feature analysis according to claim 1, wherein the critical time measurement points specifically include: a first time measuring point t of the sample user, wherein the difference value of the load of the sample user at two adjacent time measuring points is higher than a load change threshold value0(ii) a Second time measuring point t for changing load of sample user from rising to falling1(ii) a Third time measurement point t at which load drop of sample user is first lower than load change threshold3。
4. The photovoltaic electricity stealing discovery method based on temporal feature analysis according to claim 1, wherein the clustering analysis is performed on the sample users by:
judging the relation between the difference between the daily power generation change proportion of the sample user and the clustering center of the sample user and the load change threshold of the sample user, and carrying out first clustering analysis on the sample user by taking the daily power generation change proportion as a parameter;
performing cluster analysis with the key time measurement points as parameters:
set S of users to be examined after first clustering analysisuThe sample user in (1), defining a first distance dt01The following were used:
dt01=|t0+t1-ct0-ct1|
ct0to sunrise time, ct1Performing second clustering analysis on the sample by using an HCM algorithm on the basis of the first distance for the time when the illumination intensity reaches the output limit of the equipment for the first time;
to the secondSet of users to be examined after sub-cluster analysis SuThe sample user in (1), defining a second distance dt23The following were used:
dt23=|t2+t3-ct2-ct3|
ct2the time when the illumination intensity is weaker than the output limit of the equipment for the first time in the middle and later days, ct3And performing third clustering analysis on the samples by using an HCM algorithm on the basis of the second distance for the sunset time to confirm the electricity stealing users.
5. The photovoltaic electricity stealing discovery method based on temporal feature analysis according to claim 4, wherein the specific method of the first clustering analysis is:
if △ -c△≥δ△Then the corresponding sample user is charged into the electricity stealing user set Ss;
If c is△-△≥δ△If the corresponding sample user is counted as the failure user set Sb;
The residual samples are counted into a user set S to be examinedu;
δ△Is the load change threshold of the sample user; c. C△Is the clustering center of the sample user, and △ is the daily generated energy change proportion of the sample user.
6. The photovoltaic electricity stealing discovery method based on the temporal feature analysis according to claim 4 or 5, wherein the daily power generation amount variation ratio △ of the photovoltaic sample users is:
in the formula, sum (E)p) Is the total generating capacity of photovoltaic users on the day, sum (E'p) Is the total power generation amount of photovoltaic users in the previous day, EpIs the generated energy of the photovoltaic user at the time of the day measurement point p, Ep' is the power generation amount of the photovoltaic user at the previous day time measurement point p.
7. The photovoltaic electricity stealing discovery method based on time characteristic analysis according to claim 1, wherein the specific method for performing the second clustering analysis on the samples by using the HCM algorithm based on the first distance is as follows: will dt01Set to the set of users to be examined S as 0uCluster center of (d)t01=ct0+ct1Set to failed user set SbCluster center of (d)t01=2ct0+2ct1Set as electricity stealing subscriber set SsThe cluster center of (a); if the sample user is distant from the electricity stealing user set SsWhen the cluster center is nearest, the sample user is counted as the electricity stealing user set Ss(ii) a If the sample user is far from the faulty user set SbWhen the cluster center is nearest, the sample user is counted as a fault user set Sb(ii) a The residual samples are counted into a user set S to be examinedu。
8. The photovoltaic electricity stealing discovery method based on the temporal feature analysis of claim 1, wherein the specific method for performing the third clustering analysis on the samples by using the HCM algorithm based on the second distance is as follows: will dt23Set to 0 as the normal user set SnCluster center of (d)t23=ct2+ct3Set to failed user set SbCluster center of (d)t23=2ct2+2ct3Set as electricity stealing subscriber set SsThe cluster center of (a); if the sample user is distant from the electricity stealing user set SsWhen the cluster center is nearest, the sample user is counted as the electricity stealing user set Ss(ii) a If the sample user is far from the faulty user set SbWhen the cluster center is nearest, the sample user is counted as a fault user set Sb(ii) a The remaining samples are counted in the normal user set Sn。
9. A photovoltaic electricity stealing discovery system based on time characteristic analysis is characterized by comprising:
a critical time measurement point calculation module: the key time measuring points comprise a time measuring point of output improvement, a time measuring point of output stability and a time measuring point of output reduction in an output curve of a sample user; or the key time measuring points comprise a time measuring point of output improvement, a time measuring point of highest output and a time measuring point of output reduction in the output curve of the sample user;
the daily generated energy change proportion calculation module: calculating the daily power generation change proportion of two adjacent days of the sample user;
HCM algorithm cluster analysis module: and performing cluster analysis on the sample users by using an HCM algorithm according to the daily power generation amount change proportion and the key time measuring point, and confirming the electricity stealing users.
10. The photovoltaic electricity stealing discovery system based on temporal feature analysis of claim 9, wherein the HCM algorithm module comprises:
a first clustering analysis module: judging the relation between the difference between the daily power generation change proportion of the sample user and the clustering center of the sample user and the load change threshold of the sample user, and carrying out first clustering analysis on the sample user;
a second clustering analysis module: set S of users to be examined after first clustering analysisuThe sample user in (1), defining a first distance dt01The following were used:
dt01=|t0+t1-ct0-ct1|
ct0to sunrise time, ct1Performing second clustering analysis on the sample by using an HCM algorithm on the basis of the first distance for the time when the illumination intensity reaches the output limit of the equipment for the first time;
a third clustering analysis module: set S of users to be examined after second clustering analysisuThe sample user in (1), defining a second distance dt23The following were used:
dt23=|t2+t3-ct2-ct3|
ct2the time when the illumination intensity is weaker than the output limit of the equipment for the first time in the middle and later days, ct3Based on the second distance for sunset timeAnd thirdly clustering and analyzing the samples by using an HCM algorithm to confirm the electricity stealing users.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911192256.6A CN110910028A (en) | 2019-11-28 | 2019-11-28 | Photovoltaic electricity stealing discovery method and system based on time characteristic analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201911192256.6A CN110910028A (en) | 2019-11-28 | 2019-11-28 | Photovoltaic electricity stealing discovery method and system based on time characteristic analysis |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110910028A true CN110910028A (en) | 2020-03-24 |
Family
ID=69820175
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201911192256.6A Pending CN110910028A (en) | 2019-11-28 | 2019-11-28 | Photovoltaic electricity stealing discovery method and system based on time characteristic analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110910028A (en) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108764603A (en) * | 2018-03-30 | 2018-11-06 | 广东电网有限责任公司 | A kind of user's electricity stealing discrimination method based on higher-dimension random matrix |
CN109190916A (en) * | 2018-08-09 | 2019-01-11 | 国网浙江桐庐县供电有限公司 | Method of opposing electricity-stealing based on big data analysis |
CN109583680A (en) * | 2018-09-30 | 2019-04-05 | 国网浙江长兴县供电有限公司 | A kind of stealing discrimination method based on support vector machines |
CN110147871A (en) * | 2019-04-17 | 2019-08-20 | 中国电力科学研究院有限公司 | A kind of stealing detection method and system based on SOM neural network Yu K- mean cluster |
CN110288039A (en) * | 2019-06-29 | 2019-09-27 | 河南工业大学 | Based on user power utilization load characteristic stealing detection method |
-
2019
- 2019-11-28 CN CN201911192256.6A patent/CN110910028A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108764603A (en) * | 2018-03-30 | 2018-11-06 | 广东电网有限责任公司 | A kind of user's electricity stealing discrimination method based on higher-dimension random matrix |
CN109190916A (en) * | 2018-08-09 | 2019-01-11 | 国网浙江桐庐县供电有限公司 | Method of opposing electricity-stealing based on big data analysis |
CN109583680A (en) * | 2018-09-30 | 2019-04-05 | 国网浙江长兴县供电有限公司 | A kind of stealing discrimination method based on support vector machines |
CN110147871A (en) * | 2019-04-17 | 2019-08-20 | 中国电力科学研究院有限公司 | A kind of stealing detection method and system based on SOM neural network Yu K- mean cluster |
CN110288039A (en) * | 2019-06-29 | 2019-09-27 | 河南工业大学 | Based on user power utilization load characteristic stealing detection method |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2016008263A1 (en) | Method for distinguishing power distribution area from phase by voltage characteristics | |
Kayal et al. | An analytical approach for allocation and sizing of distributed generations in radial distribution network | |
CN111103565B (en) | Data transformation method and system based on intelligent electric energy meter metering error analysis | |
CN112633924B (en) | District electric energy substitution demand analysis method based on load decomposition | |
Bracale et al. | Analysis of the Italian distribution system evolution through reference networks | |
CN111008510A (en) | Method for identifying ammeter in transformer area and method for judging transformer area to which ammeter belongs | |
CN108336741A (en) | A kind of Contingency screening method and system of overall process Voltage stability analysis | |
CN107832949A (en) | A kind of big data method of servicing towards power consumer | |
CN107545365B (en) | Reliability evaluation method suitable for power distribution network containing high-permeability distributed power supply | |
CN110231503B (en) | High-loss platform area electricity stealing user identification and positioning method based on Glandum causal test | |
CN110768256A (en) | Transformer area topology identification method, device and system based on voltage harmonic atlas | |
CN107732902B (en) | Power distribution network economic operation monitoring and evaluation method | |
CN105701265A (en) | Double-fed wind generator modeling method and apparatus | |
CN110910028A (en) | Photovoltaic electricity stealing discovery method and system based on time characteristic analysis | |
CN110969539A (en) | Photovoltaic electricity stealing discovery method and system based on curve morphological analysis | |
Li et al. | Large‐scale source‐grid‐load friendly interactive system introduction and real load shedding verification test technology | |
Song et al. | New line‐loss management technology of distribution area based on electricity information collection system | |
Alyami et al. | Overvoltage risk analysis in distribution networks with high penetration of PVs | |
CN110854922A (en) | System and method for evaluating new energy accepting capability of regional power grid based on ant colony algorithm | |
Nose et al. | A LRT control method using sensor information in distribution systems with a large amount of PVs and EVs | |
CN107818414B (en) | Method for generating N-2 expected accident set of large-scale alternating current-direct current hybrid power grid | |
Thayer et al. | Improving end-use load modeling using machine learning and smart meter data | |
Athanasiadis et al. | A Benchmarking Testbed for Low-Voltage Active Distribution Network Studies | |
CN115809761B (en) | Voltage quality analysis method and system based on low-voltage transformer area | |
WU et al. | Frequency characteristics of East China Power Grid after bipolar locking of ultra‐high voltage direct current (UHVDC) |
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 | ||
AD01 | Patent right deemed abandoned | ||
AD01 | Patent right deemed abandoned |
Effective date of abandoning: 20230707 |