CN108734599A - Power grid island effect detecting system and method - Google Patents
Power grid island effect detecting system and method Download PDFInfo
- Publication number
- CN108734599A CN108734599A CN201710251309.1A CN201710251309A CN108734599A CN 108734599 A CN108734599 A CN 108734599A CN 201710251309 A CN201710251309 A CN 201710251309A CN 108734599 A CN108734599 A CN 108734599A
- Authority
- CN
- China
- Prior art keywords
- data
- matrix
- island effect
- phasor
- common connection
- 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
- 230000000694 effects Effects 0.000 title claims abstract description 63
- 238000000034 method Methods 0.000 title claims abstract description 61
- 238000001514 detection method Methods 0.000 claims abstract description 62
- 239000011159 matrix material Substances 0.000 claims description 61
- 230000005611 electricity Effects 0.000 claims description 25
- 239000013598 vector Substances 0.000 claims description 16
- 238000004891 communication Methods 0.000 claims description 11
- 238000009826 distribution Methods 0.000 claims description 11
- 238000000354 decomposition reaction Methods 0.000 claims description 10
- 230000000007 visual effect Effects 0.000 claims description 10
- 230000008569 process Effects 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000007781 pre-processing Methods 0.000 claims description 8
- 238000012549 training Methods 0.000 claims description 7
- 238000005259 measurement Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 7
- 238000012360 testing method Methods 0.000 description 6
- 230000005540 biological transmission Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000006378 damage Effects 0.000 description 3
- 238000009434 installation Methods 0.000 description 3
- 238000012423 maintenance Methods 0.000 description 3
- 238000000513 principal component analysis Methods 0.000 description 3
- 238000013459 approach Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 235000013399 edible fruits Nutrition 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000004744 fabric Substances 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000010248 power generation Methods 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001052 transient effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/381—Dispersed generators
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J2203/00—Indexing scheme relating to details of circuit arrangements for AC mains or AC distribution networks
- H02J2203/20—Simulating, e g planning, reliability check, modelling or computer assisted design [CAD]
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02J—CIRCUIT ARRANGEMENTS OR SYSTEMS FOR SUPPLYING OR DISTRIBUTING ELECTRIC POWER; SYSTEMS FOR STORING ELECTRIC ENERGY
- H02J3/00—Circuit arrangements for ac mains or ac distribution networks
- H02J3/38—Arrangements for parallely feeding a single network by two or more generators, converters or transformers
- H02J3/388—Islanding, i.e. disconnection of local power supply from the network
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Economics (AREA)
- Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Geometry (AREA)
- Evolutionary Computation (AREA)
- Computer Hardware Design (AREA)
- Public Health (AREA)
- Water Supply & Treatment (AREA)
- Power Engineering (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Supply And Distribution Of Alternating Current (AREA)
Abstract
The present invention relates to a kind of power grid island effect detection methods, including：A. the phasor data at corresponding points of common connection is acquired by each phase amount detection unit；B. the data of above-mentioned acquisition are pre-processed；C. it according to above-mentioned pretreated data, is modeled using pca method；D. on the basis of pca method models, real-time data to be tested are detected, to determine whether to generate island effect.The invention further relates to a kind of power grid island effect detecting systems.Data used in the present invention more comprehensively, precision higher, and improve accuracy in detection.
Description
Technical field
The present invention relates to a kind of power grid island effect detecting system and methods.
Background technology
In distributed power generation grid-connected system, when main body power grid is due to electric fault, interruption maintenance or other human factors
When interruption of power supply, each grid-connected system does not detect that power failure cuts off itself, and the isolated state of maintenance power supply,
The island phenomenon that uncontrollable self-energizing is formd with load around, is referred to as island effect.When island effect occurs, electricity
Net voltage or frequency will appear fluctuation, be easy to cause equipment damage；In addition, partial line road is still charged, it can serious prestige
Coerce staff's personal safety.Effective island effect detection method is to safeguard distributed power generation grid-connected system and bulk power grid just
The effective protection measure often run.
Existing island effect detection scheme is broadly divided into remote detection and local detection method.Remote detection is adopted due to data
The use for collecting equipment and communication equipment, increases the complexity and installation maintenance cost of system, comes to the expansion bands of power grid certain
Limitation.In local detection method, active method introduces interference signal to power grid, when being incorporated into the power networks, due to clamping down on for bulk power grid, disturbance
Influence very little of the signal to inverter；And island effect is when occurring, the parameter (voltage, phase, frequency) of inverter output end by
Disturbing signal effect deviates normal value and detects island effect when beyond given threshold.And passive means are to pass through detection
Whether the terminal voltage amplitude, phase, frequency of points of common connection, harmonic wave occur abnormal to determine whether generating lonely when grid cut-off
Island, since it does not introduce any interference signal to power grid, do not damage power quality and by the support of numerous researchers.
Existing island effect detection technique avoids the problem of conventional monitoring methods may cause to a certain extent, such as
It avoids by introducing the shortcomings that interference is to destroy power quality into power grid, but still has following technical problem：One
It is that the training of neural network needs a large amount of partial data, could improves accuracy of detection, and the SCADA that current power grid relies primarily on
Device data amount is simultaneously little, and data acquisition when mainly for stable state, lacks the high-precision note of real-time transient data
Record；Second is that the process object of wavelet transformation and neural network can only be synchronization that is single, can not being carried out at the same time in wide scope
Detection, and rapid identification is carried out to island effect scene.
Invention content
In view of this, it is necessary to provide a kind of power grid island effect detecting system and method, can utilize multiple dimensioned
Image information and vision significance obtain the information of image multi-dimensional degree, and carry out power grid island effect detection.
The present invention provides a kind of power grid island effect detecting system, including：Main network system, main circuit breaker, secondary breaker,
Distributed grid-connected electricity generation system, local load demand and major network load demand, which is characterized in that the system further includes：With not
Less than the high speed communication equipment that a phase amount detection unit is electrically connected, and be electrically connected respectively with the high speed communication equipment
Global positioning system and data acquisition applications analysis center, data acquisition applications analysis center include acquisition module, pre- place
Module, modeling module and detection module are managed, wherein：The acquisition module is used to acquire by each phase amount detection unit corresponding public
Phasor data at tie point altogether；The preprocessing module is for pre-processing the data of above-mentioned acquisition；The modeling mould
Block is used to, according to above-mentioned pretreated data, model using pca method；The detection module is used in pivot analysis
On the basis of Method Modeling, real-time data to be tested are detected, to determine whether to generate island effect.
Wherein, the phasor data includes：Voltage magnitude, current amplitude and voltage phase angle.
The points of common connection is the public company of main network system, distributed grid-connected electricity generation system and major network load demand
Contact.
The modeling module is specifically used for：
c1：By the main network system of above-mentioned acquisition, distributed grid-connected electricity generation system and major network load demand points of common connection
The voltage magnitude at place, the phasor data of current amplitude and voltage phase angle, use phasor dataIt indicates；
c2：The PMU data composition matrix in N number of place is handled, data matrix X ∈ R are obtainedN×m：
c3：The eigenvalue matrix of normalized treated data matrix X：
c4：Singular value decomposition is carried out by the variance matrix to data matrix to obtain：
X=U Λ0VT,
Wherein, Λ ∈ RN×NBe diagonal matrix, contain the characteristic value of the variance matrix, and by the variance matrix into
Row is following to be decomposed：
c5：Raw data matrix is decomposed into load vectors p by pca methodiAnd score vector tiProduct it
With along with residual error E：
Wherein, k indicates the pivot number chosen, for representing the maximum principal component space of variance in data；
c6：Calculate two statistical indicators that pca method is used for detection model：
T2Index：
And Q indexs：
c7：Detection control limit is calculated on the basis of the processing of normal data, due to T2Statistical value meets F distributions, presses
According to confidence alpha, Statisti-cal control limitIt is calculated by following formula：
Wherein, k is pivot number, and m is the variable number of training data, and F (k, m-k, α) is indicated when data fit F distributions
When, the probability value when degree of freedom is (m-k), and confidence level is α；
c8：Calculate the Statisti-cal control limit of Q values：
Wherein, θiIndicate that all remaining characteristic values correspond to the sum of 1,2,3 different power powers：
h0For intermediate result variable：
The detection module is specifically used for：By normal operation and containing island effect event real-time data to be tested
Calculate new statistical indicator T2With Q values, then by new value of statistical indicant and Statisti-cal control limit be compared, generate Visual retrieval
Chart detects the island effect of power grid according to the Visual retrieval chart；It is further confirmed that specifically by calculating contribution rate
The phase amount detection unit of island effect occurs.
The present invention also provides a kind of power grid island effect detection methods, including：A. it is acquired by each phase amount detection unit
Phasor data at corresponding points of common connection；B. the data of above-mentioned acquisition are pre-processed；C. according to above-mentioned pretreated
Data are modeled using pca method；D. on the basis of pca method models, real-time data to be tested are examined
It surveys, to determine whether to generate island effect.
Wherein, the phasor data includes：Voltage magnitude, current amplitude and voltage phase angle.
The points of common connection is the public company of main network system, distributed grid-connected electricity generation system and major network load demand
Contact.
The step c detailed processes include：
c1：By the main network system of above-mentioned acquisition, distributed grid-connected electricity generation system and major network load demand points of common connection
The voltage magnitude at place, the phasor data of current amplitude and voltage phase angle, use phasor dataIt indicates；
c2：The PMU data composition matrix in N number of place is handled, data matrix X ∈ R are obtainedN×m：
c3：The eigenvalue matrix of normalized treated data matrix X：
c4：Singular value decomposition is carried out by the variance matrix to data matrix to obtain：
X=U Λ0VT,
Wherein, Λ ∈ RN×NBe diagonal matrix, contain the characteristic value of the variance matrix, and by the variance matrix into
Row is following to be decomposed：
c5：Raw data matrix is decomposed into load vectors p by pca methodiAnd score vector tiProduct it
With along with residual error E：
Wherein, k indicates the pivot number chosen, for representing the maximum principal component space of variance in data；
c6：Calculate two statistical indicators that pca method is used for detection model：
T2Index：
And Q indexs：
c7：Detection control limit is calculated on the basis of the processing of normal data, due to T2Statistical value meets F distributions, presses
According to confidence alpha, Statisti-cal control limitIt is calculated by following formula：
Wherein, k is pivot number, and m is the variable number of training data, and F (k, m-k, α) is indicated when data fit F distributions
When, the probability value when degree of freedom is (m-k), and confidence level is α；
c8：Calculate the Statisti-cal control limit of Q values：
Wherein, θiIndicate that all remaining characteristic values correspond to the sum of 1,2,3 different power powers：
h0For intermediate result variable：
The step d is specifically included：Normal operation and containing island effect event real-time data to be tested are pressed
New statistical indicator T is calculated according to step c1 to step c62With Q values, then by new value of statistical indicant and above-mentioned steps c7 and step
Statisti-cal control limit described in c8 is compared, and is generated Visual retrieval chart and is detected according to the Visual retrieval chart
The island effect of power grid；The specific phase amount detection unit that island effect occurs is further confirmed that by calculating contribution rate.
Compared with prior art, the present invention mainly has the advantage that：
(1) present invention is the PMU data based on Wide Area Measurement System, and compared with prior art, data used are more
Add comprehensive, precision higher, and since PMU technologies introduce the temporal information of the whole network synchronization, can include more fully dynamically
Information improves the accuracy of island effect detection；
(2) pca method that uses of the present invention, is a kind of common process statistics method, be usually used in image procossing,
The artificial aptitude area such as recognition of face, the at present application in power grid islet operation detection field or an innovation direction.It is main
Element method can simultaneously be handled multiplely compared with the methods of the wavelet analysis that the prior art mostly uses greatly, neural network
Quickly identification island effect scene is calculated according to pivot contribution rate in the real time data of point.It avoids at single object
The limitation of reason and analysis, improves treatment effeciency from power grid overall operation, improves accuracy in detection.
Description of the drawings
Fig. 1 is the implementation environment schematic diagram of power grid island effect detecting system of the present invention；
Fig. 2 is the operation process chart of power grid island effect detection method preferred embodiment of the present invention；
Fig. 3 is the testing result schematic diagram that one embodiment of the invention pca method handles normal operation data；
Fig. 4 is that one embodiment of the invention pca method handles the testing result signal containing island effect event data
Figure；
Fig. 5 is that one embodiment of the invention island effect scene judges result schematic diagram.
Specific implementation mode
Below in conjunction with the accompanying drawings and specific embodiment the present invention is described in further detail.
As shown in fig.1, being the implementation environment schematic diagram of power grid island effect detecting system of the present invention.
Power grid island effect detecting system of the present invention is established on the power supply system including distributed energy, packet
It includes：Main network system 1, main circuit breaker 2, secondary breaker 31,32,33, distributed grid-connected electricity generation system (is distinguished in the present embodiment
Also can be other kinds of for other distributed energies 43 such as wind generator system 41, solar power system 42, marine geothermal
Distributed grid-connected electricity generation system), local load demand 51,52 and major network load demand 6；Further include：With phase amount detection unit
(Phasor Measurement Unit, PMU) 71,72,73 high speed communication equipment 9 being electrically connected, and respectively with it is described
The global positioning system (GPS) 8 and data acquisition applications analysis center 10 that high speed communication equipment 9 is electrically connected.Wherein：
The phase amount detection unit 71,72,73, be separately mounted to main network system 1 and distributed grid-connected electricity generation system 41,
42, at 43 points of common connection, for acquiring the phasor data at corresponding points of common connection.
The high speed communication equipment 9 is used to the phasor data of above-mentioned acquisition being transmitted to data acquisition applications analysis center
10。
The GPS8 is used to carry out the time synchronization of the whole network.
Data acquisition applications analysis center 10 is used to carry out analyzing processing to the data that high speed communication equipment 9 transmits,
Realize the detection to power grid island effect.Data acquisition applications analysis center 10 includes the acquisition module being electrically connected successively
101, preprocessing module 102, modeling module 103 and detection module 104.Wherein：
The acquisition module 101 is for passing through each phase amount detection unit (Phasor Measurement Unit, PMU)
Phasor data at the corresponding points of common connection of acquisition.The phasor data includes：Voltage magnitude, current amplitude and voltage-phase
Angle.Specifically：
The phase amount detection unit acquires voltage magnitude, current amplitude and voltage phase angle at each points of common connection
Phasor data, the phasor data by global positioning system (GPS) 8 carry out the whole network time synchronization, then pass through high speed
Communication equipment 9 carries out data transmission, and analyzing processing is carried out in data acquisition applications analysis center 10.
Further, main network system 1, distributed grid-connected electricity generation system 41 and the major network load that PMU71 is acquired in real time are used
The phasor data of voltage magnitude, current amplitude and voltage phase angle at 6 points of common connection of family；The main electricity that PMU72 is acquired in real time
Voltage magnitude, current amplitude at 6 points of common connection of net system 1, distributed grid-connected electricity generation system 42 and major network load demand and
The phasor data of voltage phase angle；Main network system 1, distributed grid-connected electricity generation system 43 and the major network that PMU73 is acquired in real time are negative
The phasor data of voltage magnitude, current amplitude and voltage phase angle at load 6 points of common connection of user.
The preprocessing module 102 is for pre-processing the data of above-mentioned acquisition.Specifically：
The pretreatment includes：
Incomplete raw data file is removed from the data of above-mentioned acquisition, the incomplete raw data file has
May be transmission fault, PMU is not keyed up or operation exception causes；
The normal reading PMU data at all points of common connection on the same day is chosen, imports analysis software MATLAB one by one
In, and remove the data row for mark in PMU data, access time, voltage magnitude, current amplitude and voltage-phase
Angle；
The time of selection is converted, is converted from general universal time coordinated (UTC, Universal Time Coordinated)
It (is made within 24 hours) for the local time；
The voltage phase angle of selection is pre-processed, the vector form of voltage phase angle is converted into angle value, is convenient for
Matrix calculates.
The modeling module 103 is used to, according to above-mentioned pretreated data, utilize pivot analysis (Principal
Component Analysis, PCA) Method Modeling.It specifically includes：
Main network system 1, distributed grid-connected electricity generation system 41 and the major network load that phasor measurement unit 71 is acquired in real time
The phasor data of voltage magnitude, current amplitude and voltage phase angle at 6 points of common connection of user, with vectorIndicate m time series sampled point of PMU71 continuous acquisitions here；
Likewise, the data of PMU72 devices acquisition can be expressed asIt can similarly obtain, other i-th
The PMU of a place installation can obtain the phasor data of same format
According to the calculation features of pca method, the PMU data composition matrix in N number of place is handled, is counted
According to matrix X ∈ RN×m：
Pca method is handled, the eigenvalue matrix of normalized treated data matrix X：
By to the variance matrix of data matrix carry out singular value decomposition (Singular Value Decomposition,
SVD it) obtains：
X=U Λ0VT,
Wherein, Λ ∈ RN×NIt is diagonal matrix, contains the characteristic value of the variance matrix；And by the variance matrix into
Row is following to be decomposed：
Raw data matrix is decomposed into load vectors p by pca methodiAnd score vector tiThe sum of products again
In addition residual error E：
Wherein, k indicates the pivot number chosen, for representing the maximum principal component space of variance in data.Under normal conditions,
The 95% of data population variance can be included by choosing each pivot numbers of k；
Calculate two important statistical indicators that PCA methods are used for detection model：Hotelling's T2,
With Q indexs：
By PCA approach applications when process detects, detection control limit is calculated on the basis of the processing of normal data, due to
T2Statistical value meets F distributions, and according to confidence degree α, (in the present embodiment, the confidence alpha is 95%) statistics control
System limitIt is calculated by following formula：
Wherein, k is pivot number, and m is the variable number of training data, and F (k, m-k, α) is indicated when data fit F distributions
When, the probability value when degree of freedom is (m-k), and confidence level is α；
Calculate the Statisti-cal control limit of Q values：
Wherein, θiIndicate that all remaining characteristic values correspond to the sum of 1,2,3 different power powers：
h0For intermediate result variable：
The detection module 104 is used on the basis of pca method models, and is examined to real-time data to be tested
It surveys, to determine whether to generate island effect.Specifically：
On the basis of pca method models, by normal operation and contain the real-time to be detected of island effect event
Data calculate new statistical indicator T by modeling module 1032It, then will be in new value of statistical indicant and modeling module 103 with Q values
The Statisti-cal control limit is compared, and generates Visual retrieval chart, as shown in Figure 3 and Figure 4：
Fig. 3 shows the testing result to normal data, the T of new data2The control that statistical indicator is all indicated in dotted line
Under limit, and Q statistics is substantially under control limit；
Fig. 4 show Principal Component Analysis Model to have recorded island effect generation PMU data be detected as a result, when
Carve 16:51:In 41 oval regions shown in solid after the saltus step of first time, T2With Q simultaneously surmounted detection limit, and continue to when
Between window terminate, successfully detect island effect；
In order to further confirm that the specific PMU that island effect occurs, knot as shown in Figure 5 can get by calculating contribution rate
Fruit：
As can be seen that the PMU of site number the 4th contributes maximum to failure in Fig. 5, it is believed that be point of the positions PMU
There is island effect in cloth electricity generation system, provides quick and easy and accurate testing result for Systems Operator, and assist determining
Plan.
As shown in fig.2, being the operation process chart of power grid island effect detection method preferred embodiment of the present invention.
Step S1, it is public by each phase amount detection unit (Phasor Measurement Unit, PMU) acquisition correspondence
Phasor data at tie point.The phasor data includes：Voltage magnitude, current amplitude and voltage phase angle.Specifically：
The phase amount detection unit acquires voltage magnitude, current amplitude and voltage phase angle at each points of common connection
Phasor data, the phasor data by global positioning system (GPS) 8 carry out the whole network time synchronization, then pass through high speed
Communication equipment 9 carries out data transmission, and analyzing processing is carried out in data acquisition applications analysis center 10.
Further, main network system 1, distributed grid-connected electricity generation system 41 and the major network load that PMU71 is acquired in real time are used
The phasor data of voltage magnitude, current amplitude and voltage phase angle at 6 points of common connection of family；The main electricity that PMU72 is acquired in real time
Voltage magnitude, current amplitude at 6 points of common connection of net system 1, distributed grid-connected electricity generation system 42 and major network load demand and
The phasor data of voltage phase angle；Main network system 1, distributed grid-connected electricity generation system 43 and the major network that PMU73 is acquired in real time are negative
The phasor data of voltage magnitude, current amplitude and voltage phase angle at load 6 points of common connection of user.
Step S2 pre-processes the data of above-mentioned acquisition.Specifically：
The pretreatment includes：
Incomplete raw data file is removed from the data of above-mentioned acquisition, the incomplete raw data file has
May be transmission fault, PMU is not keyed up or operation exception causes；
The normal reading PMU data at all points of common connection on the same day is chosen, imports analysis software MATLAB one by one
In, and remove the data row for mark in PMU data, access time, voltage magnitude, current amplitude and voltage-phase
Angle；
The time of selection is converted, is converted from general universal time coordinated (UTC, Universal Time Coordinated)
It (is made within 24 hours) for the local time；
The voltage phase angle of selection is pre-processed, the vector form of voltage phase angle is converted into angle value, is convenient for
Matrix calculates.
Step S3 utilizes pivot analysis (Principal Component according to above-mentioned pretreated data
Analysis, PCA) Method Modeling.Specifically comprise the following steps：
Step S31, the main network system 1 that phasor measurement unit 71 is acquired in real time, 41 and of distributed grid-connected electricity generation system
The phasor data of voltage magnitude, current amplitude and voltage phase angle at 6 points of common connection of major network load demand, with vectorIndicate m time series sampled point of PMU71 continuous acquisitions here；
Step S32, likewise, the data of PMU72 devices acquisition can be expressed asSimilarly may be used
, the PMU of other i-th of places installation can obtain the phasor data of same format
Step S33, according to the calculation features of pca method, at the PMU data composition matrix in N number of place
Reason, obtains data matrix X ∈ RN×m：
Step S34, handles pca method, the characteristic value square of normalized treated data matrix X
Battle array：
Step S35 carries out singular value decomposition (Singular Value by the variance matrix to data matrix
Decomposition, SVD) it obtains：
X=U Λ0VT,
Wherein, Λ ∈ RN×NIt is diagonal matrix, contains the characteristic value of the variance matrix；And by the variance matrix into
Row is following to be decomposed：
Raw data matrix is decomposed into load vectors p by step S36 by pca methodiAnd score vector tiMultiply
The sum of product adds residual error E：
Wherein, k indicates the pivot number chosen, for representing the maximum principal component space of variance in data, it is generally the case that
The 95% of data population variance can be included by choosing each pivot numbers of k；
Step S37 calculates two important statistical indicators that PCA methods are used for detection model：Hotelling's T2,
With Q indexs：
Step S38 will calculate detection control by PCA approach applications when process detects on the basis of the processing of normal data
System limit, due to T2Statistical value meets F distributions, and according to confidence degree α, (in the present embodiment, the confidence alpha is
95%), Statisti-cal control limitsIt is calculated by following formula：
Wherein, k is pivot number, and m is the variable number of training data, and F (k, m-k, α) is indicated when data fit F distributions
When, the probability value when degree of freedom is (m-k), and confidence level is α；
Step S39 calculates the Statisti-cal control limit of Q values：
Wherein, θiIndicate that all remaining characteristic values correspond to the sum of 1,2,3 different power powers：
h0For intermediate result variable：
Step S4：On the basis of pca method models, real-time data to be tested are detected, to determine whether
Generate island effect.Specifically：
On the basis of pca method models, by normal operation and contain the real-time to be detected of island effect event
Data calculate new statistical indicator T according to step S31 to step S372With Q values, then by new value of statistical indicant and above-mentioned steps
Statisti-cal control limit described in S38 and step S39 is compared, and generates Visual retrieval chart, as shown in Figure 3 and Figure 4：
Fig. 3 shows the testing result to normal data, the T of new data2The control that statistical indicator is all indicated in dotted line
Under limit, and Q statistics is substantially under control limit；
Fig. 4 show Principal Component Analysis Model to have recorded island effect generation PMU data be detected as a result, when
Carve 16:51:In 41 oval regions shown in solid after the saltus step of first time, T2With Q simultaneously surmounted detection limit, and continue to when
Between window terminate, successfully detect island effect；
In order to further confirm that the specific PMU that island effect occurs, knot as shown in Figure 5 can get by calculating contribution rate
Fruit：
As can be seen that the PMU of site number the 4th contributes maximum to failure in Fig. 5, it is believed that be point of the positions PMU
There is island effect in cloth electricity generation system, provides quick and easy and accurate testing result for Systems Operator, and assist determining
Plan.
Island effect detection method proposed by the present invention utilizes master by the PMU data acquired based on Wide Area Measurement System
Element method carries out pivot analysis to the data under normal operational condition and establishes model, realizes the isolated island effect of online real time data
It should quickly detect, carrying out easy failure contribution rate for the data limited more than monitoring and controlling calculates, and can determine island effect
Scene.This invention takes full advantage of the Dynamic High-accuracy real time data based on Wide Area Measurement System, helps to improve
Island effect accuracy in detection eliminates destruction of the interference of active detecting method introducing to power quality；Using pivot analysis side
Method can simultaneously be detected the PMU data in multiple places, and the erroneous judgement that may be brought when avoiding single object handles helps
Automation and the level of IT application are protected in improving smart grid distribution.
Although the present invention is described with reference to current better embodiment, those skilled in the art should be able to manage
Solution, for above-mentioned better embodiment only for illustrating the present invention, protection domain not for the purpose of limiting the invention is any in the present invention
Spirit and spirit within, any modification, equivalence replacement, improvement for being done etc., should be included in the present invention right protect
Within the scope of shield.
Claims (10)
1. a kind of power grid island effect detecting system, including：Main network system, main circuit breaker, secondary breaker, distributed grid-connected hair
Electric system, local load demand and major network load demand, which is characterized in that the system further includes：With a no less than phasor
The high speed communication equipment that detection unit is electrically connected, and the global positioning system with high speed communication equipment electric connection respectively
With data acquisition applications analysis center, data acquisition applications analysis center includes acquisition module, preprocessing module, modeling mould
Block and detection module, wherein：
The acquisition module is used to acquire the phasor data at corresponding points of common connection by each phase amount detection unit；
The preprocessing module is for pre-processing the data of above-mentioned acquisition；
The modeling module is used to, according to above-mentioned pretreated data, model using pca method；
The detection module is used on the basis of pca method models, and is detected to real-time data to be tested, with true
It is fixed whether to generate island effect.
2. the system as claimed in claim 1, which is characterized in that the phasor data includes：Voltage magnitude, current amplitude and
Voltage phase angle.
3. system as claimed in claim 2, which is characterized in that the points of common connection be main network system, it is distributed simultaneously
The points of common connection of net electricity generation system and major network load demand.
4. system as claimed in claim 3, which is characterized in that the modeling module is specifically used for：
c1：At the main network system of above-mentioned acquisition, distributed grid-connected electricity generation system and major network load demand points of common connection
The phasor data of voltage magnitude, current amplitude and voltage phase angle, uses phasor dataIt indicates；
c2：The PMU data composition matrix in N number of place is handled, data matrix X ∈ R are obtainedN×m：
c3：The eigenvalue matrix of normalized treated data matrix X：
c4：Singular value decomposition is carried out by the variance matrix to data matrix to obtain：
X=U Λ0VT,
Wherein, Λ ∈ RN×NIt is diagonal matrix, contains the characteristic value of the variance matrix, and the variance matrix is carried out such as
Lower decomposition：
c5：Raw data matrix is decomposed into load vectors p by pca methodiAnd score vector tiThe sum of products again
In addition residual error E：
Wherein, k indicates the pivot number chosen, for representing the maximum principal component space of variance in data；
c6：Calculate two statistical indicators that pca method is used for detection model：
T2Index：
And Q indexs：
c7：Detection control limit is calculated on the basis of the processing of normal data, due to T2Statistical value meets F distributions, according to confidence
Spend α, Statisti-cal control limitIt is calculated by following formula：
Wherein, k is pivot number, and m is the variable number of training data, and F (k, m-k, α) is indicated when data fit F is distributed,
Degree of freedom is (m-k), probability value when confidence level is α；
c8：Calculate the Statisti-cal control limit of Q values：
Wherein, θiIndicate that all remaining characteristic values correspond to the sum of 1,2,3 different power powers：
h0For intermediate result variable：
5. system as claimed in claim 4, which is characterized in that the detection module is specifically used for：
Normal operation and containing island effect event real-time data to be tested are calculated into new statistical indicator T2With Q values, then general
New value of statistical indicant and Statisti-cal control limit are compared, and generate Visual retrieval chart, according to the Visual retrieval chart,
Detect the island effect of power grid；
The specific phase amount detection unit that island effect occurs is further confirmed that by calculating contribution rate.
6. a kind of power grid island effect detection method, which is characterized in that this method comprises the following steps：
A. the phasor data at corresponding points of common connection is acquired by each phase amount detection unit；
B. the data of above-mentioned acquisition are pre-processed；
C. it according to above-mentioned pretreated data, is modeled using pca method；
D. on the basis of pca method models, real-time data to be tested are detected, to determine whether to generate isolated island
Effect.
7. method as claimed in claim 6, which is characterized in that the phasor data includes：Voltage magnitude, current amplitude and
Voltage phase angle.
8. the method for claim 7, which is characterized in that the points of common connection be main network system, it is distributed simultaneously
The points of common connection of net electricity generation system and major network load demand.
9. method as claimed in claim 8, which is characterized in that the step c detailed processes include：
c1：At the main network system of above-mentioned acquisition, distributed grid-connected electricity generation system and major network load demand points of common connection
The phasor data of voltage magnitude, current amplitude and voltage phase angle, uses phasor dataIt indicates；
c2：The PMU data composition matrix in N number of place is handled, data matrix X ∈ R are obtainedN×m：
c3：The eigenvalue matrix of normalized treated data matrix X：
c4：Singular value decomposition is carried out by the variance matrix to data matrix to obtain：
X=U Λ0VT,
Wherein, Λ ∈ RN×NIt is diagonal matrix, contains the characteristic value of the variance matrix, and the variance matrix is carried out such as
Lower decomposition：
c5：Raw data matrix is decomposed into load vectors p by pca methodiAnd score vector tiThe sum of products again
In addition residual error E：
Wherein, k indicates the pivot number chosen, for representing the maximum principal component space of variance in data；
c6：Calculate two statistical indicators that pca method is used for detection model：
T2Index：
And Q indexs：
c7：Detection control limit is calculated on the basis of the processing of normal data, due to T2Statistical value meets F distributions, according to confidence
Spend α, Statisti-cal control limitIt is calculated by following formula：
Wherein, k is pivot number, and m is the variable number of training data, and F (k, m-k, α) is indicated when data fit F is distributed,
Degree of freedom is (m-k), probability value when confidence level is α；
c8：Calculate the Statisti-cal control limit of Q values：
Wherein, θiIndicate that all remaining characteristic values correspond to the sum of 1,2,3 different power powers：
h0For intermediate result variable：
10. method as claimed in claim 9, which is characterized in that the step d is specifically included：
Normal operation and containing island effect event real-time data to be tested are calculated newly according to step c1 to step c6
Statistical indicator T2With Q values, then the Statisti-cal control described in new value of statistical indicant and above-mentioned steps c7 and step c8 is limited and is carried out
Compare, generates Visual retrieval chart and the island effect of power grid is detected according to the Visual retrieval chart；
The specific phase amount detection unit that island effect occurs is further confirmed that by calculating contribution rate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710251309.1A CN108734599A (en) | 2017-04-18 | 2017-04-18 | Power grid island effect detecting system and method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710251309.1A CN108734599A (en) | 2017-04-18 | 2017-04-18 | Power grid island effect detecting system and method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108734599A true CN108734599A (en) | 2018-11-02 |
Family
ID=63923935
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710251309.1A Pending CN108734599A (en) | 2017-04-18 | 2017-04-18 | Power grid island effect detecting system and method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108734599A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111678679A (en) * | 2020-05-06 | 2020-09-18 | 内蒙古电力(集团)有限责任公司电力调度控制分公司 | Circuit breaker fault diagnosis method based on PCA-BPNN |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090057665A1 (en) * | 2007-08-29 | 2009-03-05 | Broadcom Corporation | Power managing semiconductor die with reduced power consumption |
| CN104321945A (en) * | 2012-03-30 | 2015-01-28 | Abb技术有限公司 | Islanding detection in electricity distribution network |
| CN105470986A (en) * | 2015-12-17 | 2016-04-06 | 国家电网公司 | Power system partitioning method |
-
2017
- 2017-04-18 CN CN201710251309.1A patent/CN108734599A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090057665A1 (en) * | 2007-08-29 | 2009-03-05 | Broadcom Corporation | Power managing semiconductor die with reduced power consumption |
| CN104321945A (en) * | 2012-03-30 | 2015-01-28 | Abb技术有限公司 | Islanding detection in electricity distribution network |
| CN105470986A (en) * | 2015-12-17 | 2016-04-06 | 国家电网公司 | Power system partitioning method |
Non-Patent Citations (4)
| Title |
|---|
| 孙健: "智能电网暂态故障检测和电流过载预防控制研究", 《中国博士学位论文全文数据库工程科技Ⅱ辑》 * |
| 张颖伟等: "《复杂工业过程的故障诊断》", 31 December 2007, 东北大学出版社 * |
| 陶维青等: "SVD和神经网络在孤岛检测中的应用", 《电力系统保护与控制》 * |
| 霍明亮: "《小型分布式风电单相并网逆变电路设计、仿真与实现》", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111678679A (en) * | 2020-05-06 | 2020-09-18 | 内蒙古电力(集团)有限责任公司电力调度控制分公司 | Circuit breaker fault diagnosis method based on PCA-BPNN |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Data-driven event detection of power systems based on unequal-interval reduction of PMU data and local outlier factor | |
| CN108490284B (en) | New energy data acquisition device, system and method for multiple application scenes | |
| CN105403811B (en) | Wind power plant power grid fault diagnosis method and device | |
| CN101900777B (en) | Monitoring method and device of power system | |
| Hajnoroozi et al. | Generating unit model validation and calibration through synchrophasor measurements | |
| CN107153150A (en) | A kind of power distribution network over-voltage fault type recognition method and device | |
| CN102680799A (en) | Transformer substation capacitor on-line monitoring method and device based on wireless mode | |
| CN111665406A (en) | Charging station electric energy quality monitoring system | |
| CN107290592A (en) | Grounded screen electric resistance on-line monitoring system | |
| CN104793082A (en) | Harmonic correlation analysis based electricity system harmonic source recognition device | |
| Bu et al. | Stability and dynamics of active distribution networks (ADNs) with D-PMU technology: A review | |
| CN108418302A (en) | A kind of power distribution network platform based on big data analysis | |
| CN111551887A (en) | Multidimensional identification voltage transformer metering performance online monitoring platform | |
| CN202929117U (en) | Positive-sequence impedance parameter measure apparatus for transmission line | |
| Damala et al. | A simple decision tree-based disturbance monitoring system for VSC-based HVDC transmission link integrating a DFIG wind farm | |
| CN102467556A (en) | Power quality intelligent information system and method | |
| Guo et al. | A statistical process control approach for automatic anti-islanding detection using synchrophasors | |
| CN202903892U (en) | A transformer station capacitance on-line monitoring device based on a wireless mode | |
| CN108734599A (en) | Power grid island effect detecting system and method | |
| Saitoh | GPS synchronized measurement applications in Japan | |
| CN111638401A (en) | Capacitor online monitoring system and method | |
| CN116593926A (en) | Portable information system power supply comprehensive performance detection method and platform | |
| CN109283385A (en) | A kind of arrester monitoring data on-line processing method and system | |
| Moloi et al. | Power Quality Assessment of A Wind Power-Integrated System into the Power Grid | |
| JP2023545098A (en) | High-resolution electrical measurement data processing |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181102 |
|
| RJ01 | Rejection of invention patent application after publication |