CN113030616A - Sensitive load identification method based on voltage sag monitoring data - Google Patents
Sensitive load identification method based on voltage sag monitoring data Download PDFInfo
- Publication number
- CN113030616A CN113030616A CN202110233750.3A CN202110233750A CN113030616A CN 113030616 A CN113030616 A CN 113030616A CN 202110233750 A CN202110233750 A CN 202110233750A CN 113030616 A CN113030616 A CN 113030616A
- Authority
- CN
- China
- Prior art keywords
- power
- active power
- sensitive
- voltage sag
- sensitive load
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R19/00—Arrangements for measuring currents or voltages or for indicating presence or sign thereof
Abstract
The invention provides a sensitive load identification method based on voltage sag monitoring data. The actual measurement shows that the method has better identification capability on the sensitive load.
Description
Technical Field
The invention relates to the field of electric energy quality and the technical field of voltage sag, in particular to a sensitive load identification method based on voltage sag monitoring data.
Background
Today, with the rapid development of economy and the rapid innovation of science and technology, the change of industrial structure leads more high-end manufacturing industries to occupy the position of great significance. In the industries of semiconductors, precision machinery processing, power electronic equipment production and the like, a large number of sensitive devices exist, and when voltage sag occurs, the sensitive devices may stop operating, which may cause interruption of the production process. IEEE defines the voltage sag as a short-term disturbance phenomenon that the root mean square value of the voltage suddenly drops to 90% -10% of the rated voltage and returns to normal after lasting for 0.5-1 min. Voltage sag is one of the important power quality problems, and is widely concerned by domestic and foreign scholars and industries. In order to reduce the impact of voltage sag on these sensitive devices and reduce the economic loss of power consumers, power grid companies need to provide effective voltage sag management schemes to deal with complaints in time. Meanwhile, aiming at the characteristic that sensitive equipment in the industry is more, differential sag treatment value-added service can be provided, the reform of the power selling side in the power industry is promoted, and the business transformation of a power grid company is promoted.
In order to be able to specifically provide a temporary remedy, it is necessary to identify sensitive users. Not all power consumers need high-quality power supply, and the accurate judgment of the sensitive load ratio of the power consumers on the production line and the influence degree on the power consumers when voltage sag occurs is favorable for finding out a customer group needing high-quality power supply.
The identification of sensitive users based on voltage sag monitoring data is essentially to analyze whether sensitive loads exist in a certain substation or a certain power line access point and the proportion of the sensitive loads, and the existing identification method can know which users are the sensitive loads according to the complaint condition of the users and the experience summary of a power company by inquiring information. However, the method lacks mathematical basis, the identification method of the sensitive user is fuzzy, and the proportion of the sensitive load in the total load cannot be obtained.
Disclosure of Invention
The sensitive user identification based on the voltage sag monitoring data is essentially to analyze whether a sensitive load exists under a certain substation or a certain power line access point and the proportion of the sensitive load, and the traditional identification method is used for summarizing and knowing which users are the sensitive loads according to the complaint condition of the users and the experience of a power company through inquiring information. However, the method lacks mathematical basis, the identification method of the sensitive user is fuzzy, and the proportion of the sensitive load in the total load cannot be obtained.
Aiming at the defects and shortcomings in the prior art, the influence of voltage sag on high-end enterprises containing sensitive loads is considered, and the quality of power supply affects the condition of good products produced by the enterprises, however, the method for identifying the sensitive loads is researched subjectively at present. In order to qualitatively analyze the occupation ratio of the sensitive load, the invention provides a sensitive load identification method based on voltage sag monitoring data. And (4) selecting power grid voltage sag data and a simulink model for simulation verification. Experiments show that the method has better identification capability on sensitive loads.
The invention specifically adopts the following technical scheme:
a sensitive load identification method based on voltage sag monitoring data is characterized in that: according to the active power change and the power recovery condition of the line before and after the sag, an active power segmentation method based on singular value decomposition is adopted to accurately position the active power key time node, and the occupation ratio of different types of sensitive loads on the line is calculated.
Preferably, the change of the active power of the line before and after the sag and the power recovery condition are depicted by a track of the change of the active power along with the time; the active power critical time node comprises: the power begins to fall, the power falls to the lowest time, the power begins to rise again, and the power rises back to the stable time.
Preferably, let the power start-to-fall time t1At the time t when the power drops to the minimum2At the moment t when the power starts to rise back3The power rises back to the stationary time t4The active power values corresponding to the four moments are respectively marked as P1、P2、P3、P4；
Sensitive load in total load of line is compared with symbol P in temporary drop processSLThis means that there are:
wherein, | P2|/P1Representing the normal duty ratio, P, of the load to be able to operate normally without being affected by a voltage sagSLRepresenting the proportion of the load which is influenced by voltage sag and can not work normally due to power drop in the total load of the line, namely the proportion of sensitive load;
after the voltage sag, one part of active power of the sensitive load in the line can be automatically recovered, and the other part of the active power is seriously influenced by the sag and cannot be recovered; the duty ratio of the load which can normally work after the temporary drop is expressed as | P4|/P1For sensitive duty ratio P which cannot be automatically recovered after temporary dropNARExpressed, then, there is the following formula:
sensitive duty ratio P capable of automatically recovering power after temporary reductionARExpressed, its expression is:
the sensitive load in the line is the sum of the sensitive load capable of being automatically recovered and the sensitive load incapable of being automatically recovered, namely the following formula is established:
PSL＝PNAR+PAR (4)。
preferably, for the waveform of the voltage sag, the following is divided: five time periods including an event front section, a first transition section with reduced voltage, an event middle section, a second transition section with increased voltage and an event rear section;
the active power segmentation method based on singular value decomposition comprises the following steps:
step S1: median filtering;
step S2: difference processing;
step S3: and (5) singular value decomposition.
Preferably, step S1 performs a filtering process on the waveform of the active power by median filtering.
Preferably, in step S2, the waveform processed in step S1 is subjected to a difference process for distinguishing a transition section of significant fluctuation from a smooth pre-dip, post-dip and event section.
Preferably, in step S3, in a certain window, singular value decomposition is performed on the sampling point data series of each detection window to obtain the corresponding singular value size, and the transition start-stop time is obtained by setting a transition triggering threshold.
Preferably, in step S3, singular value decomposition is performed by constructing a Hankel matrix from the sampled signals:
for sampled data: x ═ X1,x2,...,xN]Extracting n elements therein as line phasor [ x1,x2,...,xn]And sequentially moving the matrix to the right by one step length to obtain N-N +1 row phasors to form a Hankel matrix.
Preferably, in step S3, the singular signal is decomposed into a plurality of component signals by singular value decomposition of Hankel matrix, and in each decomposition layer, the place of sharp mutation is the mutation point: when the method is applied to automatic segmentation, the boundary of the transition section is a catastrophe point, and a severe catastrophe position is found through singular value decomposition of the matrix, so that disturbance time is positioned.
The invention and the optimal proposal adopt an active power segmentation method based on singular value decomposition to accurately position the active power falling start-stop moment according to the active power change and the power recovery condition of the line before and after the sag, and calculate the occupation ratio of different types of sensitive loads on the line. And (4) selecting power grid voltage sag data and a simulink model for simulation verification. The actual measurement shows that the method has better identification capability on the sensitive load.
Drawings
The invention is described in further detail below with reference to the following figures and detailed description:
FIG. 1 is a schematic diagram of four exemplary traces of active power in accordance with an embodiment of the present invention;
FIG. 2 is a schematic diagram of active power segmentation according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of a system model with a sensitive load according to an embodiment of the present invention;
FIG. 4 shows a circuit L according to an embodiment of the present invention2Voltage and power waveform diagrams;
FIG. 5 shows a circuit L according to an embodiment of the present invention4Voltage and power waveform diagrams;
FIG. 6 is a waveform diagram illustrating a first line according to an embodiment of the present invention;
FIG. 7 is a waveform diagram illustrating a second circuit according to the embodiment of the present invention;
fig. 8 is a waveform diagram of a third line according to the embodiment of the invention.
Detailed Description
In order to make the features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in detail as follows:
the main content of this embodiment is based on the purpose of pertinently performing sag control on a user and pre-judging whether a line needs sag control in advance, and a series of research and analysis are performed, including the following three aspects:
firstly, an active power track of a power consumer is depicted. And judging the track type of the active power according to the voltage and power waveforms obtained by measurement, analyzing the sensitive load ratio, and analyzing the recoverable sensitive load ratio and the unrecoverable sensitive load ratio.
Second, a segmentation method based on singular value decomposition is studied. The waveform of the active power needs to accurately position four key time nodes at the moment when the power starts to fall, the moment when the power falls to the lowest, the moment when the power starts to rise again and the moment when the power rises back to a stable state.
Thirdly, identifying the line sensitive load based on the measured data, building a circuit model containing sensitive equipment, and verifying the correctness of the method provided by the embodiment.
1. Sensitive load identification method
(1) Active power trajectory classification
The embodiment calculates the occupation ratios of different types of sensitive loads according to the change of active power on the line before and after the sag and the power recovery condition. As shown in fig. 1, the active power trajectory after the voltage sag event occurs exhibits the following four types.
After a voltage sag event occurs on a subscriber line, if no sensitive load exists on the whole line, the active power on the line in an ideal state is not influenced and basically does not fluctuate, and the track of the active power is shown as track 1; if partial sensitive load exists on the whole line, the active power on the line also drops when the voltage drops, the active power is restored to a normal value after the voltage is restored, and the track of the active power is shown as track 2; if some sensitive load exists on the whole line, the active power on the line also drops when the voltage drops, but the active power cannot be recovered to a normal value after the voltage is recovered, and the track is shown as track 3; if there is a large amount of sensitive loads on the whole line, the active power on the line also drops to a large extent when the voltage drops, and then the active power does not change with the rise of the voltage any more, and the active power change waveform is shown as a trace 4.
(2) Feature analysis of power trajectories
Voltage sag waveforms caused by different operations are generally classified into three types, namely, the sag amplitudes of three-phase voltages are the same, the start and stop times of sag are consistent, the sag depths are the same, and any one-phase voltage effective value of the voltage sag waveforms is only needed to be selected for analysis; secondly, only one phase has voltage drop, the other two phases have small fluctuation or the voltage does not drop and rise reversely, and the drop phase selection and analysis are carried out on the temporary drop waves; and the third type is that two phases have voltage dip with different degrees, the voltage of the third phase does not drop or rise, and the waveform selects the voltage waveform with deeper dip to participate in analysis. Because the lower the voltage drop, the deeper the sag, the more severe the impact on the user equipment, which may result in reduced power and even stalling of the equipment, and thus, economically, serious losses.
It is found from the above four active power trace graphs that the division of the trace needs to be started from four key points, namely the power starting falling time t1At the time t when the power drops to the minimum2At the moment t when the power starts to rise back3The power rises back to the stationary time t4The active power values corresponding to the four moments are respectively marked as P1、P2、P3、P4。
In the process of temporary drop, if the sensitive load in the total load of the line is more than the symbol PSLIndicates that there is
Wherein, | P2|/P1Representing the normal duty ratio, P, of the load to be able to operate normally without being affected by a voltage sagSLRepresenting the proportion of the load which is influenced by voltage sag and can not work normally due to power drop in the total load of the line, namely the proportion of sensitive load.
After the voltage sag, a part of the active power of the sensitive load in the line can be automatically recovered, and another part of the active power is seriously affected by the sag and cannot be recovered, as shown in a trace 3. The duty ratio of the load that can normally work after the temporary drop can be expressed as | P4|/P1For sensitive duty ratio P which cannot be automatically recovered after temporary dropNARExpressed, then, there is the following formula:
sensitive duty ratio P capable of automatically recovering power after temporary reductionARExpressed, its expression is:
the sensitive load in the line is the sum of the sensitive load capable of being automatically recovered and the sensitive load incapable of being automatically recovered, namely the following formula is established:
PSL＝PNAR+PAR (4)
the characteristics of the four different active power traces in fig. 1 that can be obtained from the above-defined characteristic quantities are shown in table 1 below.
TABLE 1 active Power trace characteristics
After the acquired voltage sag data are subjected to data processing to obtain an active power waveform, which type of track the voltage sag data belong to can be analyzed according to waveform characteristics, and different sag treatment schemes are proposed according to different track types. The influence of voltage sag on users with power waveforms belonging to the track 1 is small, and the power waveforms can be ignored during the sag control; sensitive loads on a power line of the users belonging to the track 2 are all of a recoverable type, and the influence of voltage sag on the power users is general, so that the management is convenient; the active power waveform belongs to a user of a track 3, sensitive load on a power line is not completely recoverable, and the influence of voltage sag on the user is serious; and the power waveform is such users of the track 4, the equipment accessed on the power line is extremely sensitive, and the existing sensitive load can not be automatically recovered, which can seriously affect the whole production line of the user, bring great economic loss and is very necessary to carry out sag treatment on the user.
2. Singular value decomposition segmentation method
For the waveform of the voltage sag, it can be divided into 5 periods: before the temporary drop occurs, a transition section (voltage drop), a temporary drop section, a transition section (voltage rise) and five time periods after the temporary drop occurs. Similarly, we can obtain the segmentation form of the active power, and find out the time t when the active power starts to fall1At the moment t when the active power decreases to the minimum2The moment t when the active power starts to rise back3Active power rising to a stable time t4The four time nodes divide the waveform of the active power changing along with the time into five sections. I.e. dividing the waveform into the front part of the event (0-t)1) Transition section 1 (t)1-t2) Middle of event (t)2-t3) Transition section 2 (t)3-t4) And a post-event segment (after t 4), a schematic diagram of which is shown in fig. 2.
To accurately obtain t1-t4And P1-P4The optimal segmentation result is obtained, and the active power can be segmented by adopting an automatic segmentation method based on singular value decomposition.
As shown in fig. 2, there are two transition sections, which are the transition process between the two states before and after the occurrence of the system fault. The signal change amplitude of the transition section is large, the speed is high, the change trend is complex and the signal is often accompanied with the oscillation. In order to enable the segment of the singular signal to realize adaptivity, the automatic segmentation can start from three steps, which are respectively: median filtering, differential processing and singular value decomposition.
(1) Median filtering principle
The median filtering technique actually replaces the average value of all the points in the surrounding neighborhood by the value size of one sampling point, and is a nonlinear data processing technique. And effectively suppressing the noise based on the ordering statistical theory. The method can make the numerical values in the neighborhood tend to the value of one sampling point, thereby eliminating the influence of the abrupt change point on the whole waveform without influencing the authenticity of the whole waveform.
Because the voltage sag waveform has noise, and the sag event section may be accompanied by other power quality disturbance events, so that the waveform disturbance of the event section is large, the difference between the event section and the transition section is reduced, and in order to reduce the influence of the change on the detection algorithm, the waveform of the active power is filtered through median filtering.
For the acquired actual voltage waveform, usually, there is a small voltage fluctuation when the sag occurs, rather than always maintaining stability, so that a filtering process is required to optimize the waveform and accurately reflect the waveform characteristics.
(2) Differential processing
Because the waveform has severe fluctuation at a singular point, the singular point usually corresponds to the starting and stopping time of a sag transition section, the singular value is amplified to be beneficial to the decomposition of subsequent singular values and the influence of a threshold value on the performance of a detection algorithm can be reduced, so that the monitored waveform signals are subjected to differential processing, and the transition section with obvious fluctuation is distinguished from the transition section before stable sag, after sag and an event section.
The difference processing is actually a commonly used numerical solution, with many properties similar to differentiation, and the principle is as follows: let y be f (x), the derivative of y over x is:
(3) singular value decomposition
Singular Value Decomposition (SVD) is essentially a matrix Decomposition method, is very important in linear algebra, and is widely applied in application fields such as solution optimization, least squares, multivariate statistical analysis, and the like. And in a certain window, carrying out singular value decomposition on the sampling point data series of each detection window to obtain the corresponding singular value, and reasonably setting a transition section trigger threshold to obtain the transition section start-stop moment.
Assuming that the matrix Y is an m × n dimensional matrix and the rank is r, the singular value decomposition of the matrix Y can be expressed as:
where U and V are orthogonal matrices of m × m and n × n dimensions, respectively, and Σ is a diagonal matrix of r × r dimensions. Its diagonal elements are the non-zero singular values e of the matrix Y, while e1≥e2≥...≥eiAnd 0 is a zero element matrix.
Removing the zero singular value of Y from equation (6) yields the following equation:
wherein: u. ofi、viThe ith column vector for U and V, respectively.
When voltage sag occurs, the voltage amplitude and the phase of the transition section can change rapidly, and if the starting and stopping moments of rapid change can be detected, the boundary of the transition section can be determined, and the waveform segmentation is realized. Therefore, the determination of the boundary of the transition section is the core for realizing the sag segmentation, and the positioning of the boundary of the transition section is carried out by adopting the principle of a singular value decomposition method.
For any real matrix A ∈ Rm×nThere must be two orthogonal matrices U and V as follows:
U＝[u1,u2,…um]∈Rm×m，UTU＝I (8)
V＝[v1,v2,…vn]∈Rn×n，VTV＝I (9)
the singular value decomposition of the a matrix is then expressed as:
A＝USVT (10)
where S is called the singular value diagonal matrix and S ∈ Rm×nThe expression is as follows:
where 0 is a zero matrix, p ═ min (m, n), and σ1≥σ2≥...≥σp≥0，σi(i ═ 1, 2.., p) is the singular values of the a matrix.
The singular value decomposition of a can also be further expressed as:
Ai＝σiuivi T (12)
the accurate positioning of the disturbance signal requires that an appropriate a matrix is constructed according to the sampling signal, and in the embodiment, the Hankel matrix is constructed by the original signal to decompose the singular value.
For sampled data: x ═ X1,x2,...,xN]Extracting n elements therein as line phasor [ x1,x2,...,xn]And sequentially moving the matrix to the right by one step length to obtain N-N +1 row phasors, so that a Hankel matrix is formed:
wherein, if m is N-N +1, then A is in the same size as Rm×nAnd the formula is developed as follows:
let the first row element of the A matrix be Pi,1I.e. Pi,1[x1,x2,...,xn]Remember that the nth column phasor of the first row element is Hi,nAnd satisfies the following conditions:
Piis the signal of the i-th layer component. The original signal X can be represented as:
wherein q is the number of decomposed layers. From (16) the known signal can be obtained by linear superposition of a series of component signals by singular value decomposition. The voltage sag signal and the power sag signal are singular signals, the singular signals are decomposed into a plurality of component signals through singular value decomposition of a Hankel matrix, and in each decomposition layer, the place where severe mutation can be seen is a mutation point. When the method is applied to automatic segmentation, the boundary of the transition section is also a mutation point, and a severe mutation position is found through singular value decomposition of the matrix, so that the disturbance time is positioned.
To verify the accuracy and validity of the proposed method, this example builds a model in MATLAB/Simulink as shown in fig. 3. Four lines L are connected to 10kV bus1The voltage is reduced by a transformer and then is connected with a load containing sensitive equipment; line L2Is connected with a common load through a step-down transformer; line L3A common load is connected, and the circuit is mainly used for short-circuit fault setting; line L4The medium load is a mixed load, 30% of sensitive load and 70% of common load are connected, and the two are in parallel relation.
On the line L3The short-circuit fault of the A phase is set, the amplitude is temporarily reduced by 85 percent, the duration is 0.075s, and the line L is observed2The power waveform is shown in fig. 4. Due to the line L2The upper is normal load, is not influenced by voltage sag, so L2The active power on is substantially unchanged. FIG. 5 shows a line L4Voltage and power waveform diagrams, it can be known that when a voltage sag occurs, the line active power is reduced to 70%, and 30% of the load is affected; after the temporary reduction is finished, 17% of the load is recovered to supply power, and 13% of the load is unrecoverable, so that the active power is stabilized at 87% after the temporary reduction is finished, and the simulation setting condition is met.
And (3) actual measurement data verification:
and selecting voltage sag actual measurement data of different power lines of the power grid for actual measurement, acquiring an active power waveform, and dividing the track type according to the waveform characteristics to obtain the sensitive load ratio on the line. And (3) adopting an MATLAB to write a program, carrying out voltage sag pretreatment and obtaining a simulation experiment oscillogram. The voltage effective values and the three-phase active power waveforms of the three different lines after the sag occurs are shown in fig. 6-8.
The active power waveform of the line shown in fig. 6 belongs to trace 3, and after the sag occurs, the active power first drops to 21% of the rated value, and after 0.245s, the active power starts to recover, and finally recovers to 74% of the rated value. That is, in this sag, 53% of the load of the line is recoverable sensitive load, and 26% of the line is unrecoverable sensitive load. The active power waveform of the line shown in fig. 7 belongs to trace 4, and after the sag occurs, the active power drops to 20.5% of the rated value, and the active power still remains 20.5% after the sag is finished. And the temporary drop loss is proved to be unrecoverable. The active power waveform of the line shown in fig. 8 belongs to trace 2, and after the sag occurs, the active power drops to 73.8% of the rated value, and returns to the rated value after the end of the transition section 2. The sensitive load of the sag loss can be recovered.
The present invention is not limited to the above preferred embodiments, and other various sensitive load identification methods based on voltage sag monitoring data can be derived by anyone in light of the present invention.
Claims (9)
1. A sensitive load identification method based on voltage sag monitoring data is characterized in that: according to the active power change and the power recovery condition of the line before and after the sag, an active power segmentation method based on singular value decomposition is adopted to accurately position the active power key time node, and the occupation ratio of different types of sensitive loads on the line is calculated.
2. The voltage sag monitoring data-based sensitive load identification method according to claim 1, wherein: the change of the active power of the line before and after the sag and the power recovery condition are depicted by a track of the change of the active power along with the time; the active power critical time node comprises: the power begins to fall, the power falls to the lowest time, the power begins to rise again, and the power rises back to the stable time.
3. The voltage sag monitoring data-based sensitive load identification method according to claim 2, wherein:
let the power start to fall time t1At the time t when the power drops to the minimum2At the moment t when the power starts to rise back3The power rises back to the stationary time t4The active power values corresponding to the four moments are respectively marked as P1、P2、P3、P4；
Sensitive load in total load of line is compared with symbol P in temporary drop processSLThis means that there are:
wherein, | P2|/P1Representing the normal duty ratio, P, of the load to be able to operate normally without being affected by a voltage sagSLRepresenting the proportion of the load which is influenced by voltage sag and can not work normally due to power drop in the total load of the line, namely the proportion of sensitive load;
after the voltage sag, one part of active power of the sensitive load in the line can be automatically recovered, and the other part of the active power is seriously influenced by the sag and cannot be recovered; the duty ratio of the load which can normally work after the temporary drop is expressed as | P4|/P1For sensitive duty ratio P which cannot be automatically recovered after temporary dropNARExpressed, then, there is the following formula:
sensitive duty ratio P capable of automatically recovering power after temporary reductionARExpressed, its expression is:
the sensitive load in the line is the sum of the sensitive load capable of being automatically recovered and the sensitive load incapable of being automatically recovered, namely the following formula is established:
PSL＝PNAR+PAR (4)。
4. the voltage sag monitoring data-based sensitive load identification method according to claim 3, wherein: for the waveform of the voltage sag, the following is divided: five time periods including an event front section, a first transition section with reduced voltage, an event middle section, a second transition section with increased voltage and an event rear section;
the active power segmentation method based on singular value decomposition comprises the following steps:
step S1: median filtering;
step S2: difference processing;
step S3: and (5) singular value decomposition.
5. The voltage sag monitoring data-based sensitive load identification method according to claim 4, wherein: step S1 performs filtering processing on the waveform of the active power by median filtering.
6. The voltage sag monitoring data-based sensitive load identification method according to claim 5, wherein: in step S2, the waveform processed in step S1 is subjected to a difference process for distinguishing a transition section of significant fluctuation from a stationary pre-dip, post-dip, and event section.
7. The voltage sag monitoring data-based sensitive load identification method according to claim 6, wherein: in step S3, singular value decomposition is performed on the sampling point data series of each detection window within a certain window to obtain the corresponding singular value size, and the transition start-stop time is obtained by setting a transition triggering threshold.
8. The voltage sag monitoring data-based sensitive load identification method according to claim 7, wherein: in step S3, singular value decomposition is performed by constructing a Hankel matrix from the sampled signals:
for sampled data: x ═ X1,x2,...,xN]Extracting n elements therein as line phasor [ x1,x2,...,xn]And sequentially moving the matrix to the right by one step length to obtain N-N +1 row phasors to form a Hankel matrix.
9. The voltage sag monitoring data-based sensitive load identification method according to claim 8, wherein: in step S3, the singular signal is decomposed into a plurality of component signals by singular value decomposition of the Hankel matrix, and in each decomposition layer, the place of the sharp mutation is the mutation point: when the method is applied to automatic segmentation, the boundary of the transition section is a catastrophe point, and a severe catastrophe position is found through singular value decomposition of the matrix, so that disturbance time is positioned.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110233750.3A CN113030616A (en) | 2021-03-03 | 2021-03-03 | Sensitive load identification method based on voltage sag monitoring data |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110233750.3A CN113030616A (en) | 2021-03-03 | 2021-03-03 | Sensitive load identification method based on voltage sag monitoring data |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113030616A true CN113030616A (en) | 2021-06-25 |
Family
ID=76465619
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110233750.3A Pending CN113030616A (en) | 2021-03-03 | 2021-03-03 | Sensitive load identification method based on voltage sag monitoring data |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113030616A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113484689A (en) * | 2021-07-22 | 2021-10-08 | 云南电网有限责任公司昆明供电局 | Distribution network fault studying and judging method and system based on feeder line current sudden reduction degree and switch load ratio probability density |
| EP4296697A1 (en) * | 2022-06-24 | 2023-12-27 | Eaton Intelligent Power Limited | Method and system of detection of dropped loads resulting from an electrical power quality event |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5886429A (en) * | 1997-12-11 | 1999-03-23 | Board Of Regents, The University Of Texas System | Voltage sag/swell testing station |
| CN107462764A (en) * | 2017-09-25 | 2017-12-12 | 南京灿能电力自动化股份有限公司 | A kind of voltage dip detection and the automatic segmentation method portrayed |
| CN107979086A (en) * | 2017-11-14 | 2018-05-01 | 国网江苏省电力公司电力科学研究院 | Voltage sag reason recognition methods based on EM algorithms and gradient boosted tree |
| CN109975636A (en) * | 2019-03-25 | 2019-07-05 | 深圳供电局有限公司 | A kind of voltage sag sensitivity appraisal procedure and system |
| CN111680879A (en) * | 2020-05-11 | 2020-09-18 | 国家电网有限公司 | Power distribution network operation toughness evaluation method and device considering sensitive load failure |
-
2021
- 2021-03-03 CN CN202110233750.3A patent/CN113030616A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5886429A (en) * | 1997-12-11 | 1999-03-23 | Board Of Regents, The University Of Texas System | Voltage sag/swell testing station |
| CN107462764A (en) * | 2017-09-25 | 2017-12-12 | 南京灿能电力自动化股份有限公司 | A kind of voltage dip detection and the automatic segmentation method portrayed |
| CN107979086A (en) * | 2017-11-14 | 2018-05-01 | 国网江苏省电力公司电力科学研究院 | Voltage sag reason recognition methods based on EM algorithms and gradient boosted tree |
| CN109975636A (en) * | 2019-03-25 | 2019-07-05 | 深圳供电局有限公司 | A kind of voltage sag sensitivity appraisal procedure and system |
| CN111680879A (en) * | 2020-05-11 | 2020-09-18 | 国家电网有限公司 | Power distribution network operation toughness evaluation method and device considering sensitive load failure |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113484689A (en) * | 2021-07-22 | 2021-10-08 | 云南电网有限责任公司昆明供电局 | Distribution network fault studying and judging method and system based on feeder line current sudden reduction degree and switch load ratio probability density |
| CN113484689B (en) * | 2021-07-22 | 2023-08-29 | 云南电网有限责任公司昆明供电局 | Distribution network fault studying and judging method and system based on load sudden reduction degree and switch load duty ratio probability density |
| EP4296697A1 (en) * | 2022-06-24 | 2023-12-27 | Eaton Intelligent Power Limited | Method and system of detection of dropped loads resulting from an electrical power quality event |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109193650B (en) | Power grid weak point evaluation method based on high-dimensional random matrix theory | |
| CN113030616A (en) | Sensitive load identification method based on voltage sag monitoring data | |
| Wang et al. | A low-rank matrix approach for the analysis of large amounts of power system synchrophasor data | |
| CN111679158A (en) | Power distribution network fault identification method based on synchronous measurement data similarity | |
| Li et al. | Fast event identification through subspace characterization of PMU data in power systems | |
| CN111007359A (en) | Power distribution network single-phase earth fault identification starting method and system | |
| Macii et al. | Rapid voltage change detection: Limits of the IEC standard approach and possible solutions | |
| CN112396007A (en) | Non-invasive three-threshold detection method and system for load sudden change event of residents | |
| CN111291918B (en) | Rotating machine degradation trend prediction method based on stationary subspace exogenous vector autoregression | |
| CN109635430B (en) | Power grid transmission line transient signal monitoring method and system | |
| CN113162037B (en) | Power system transient voltage stability self-adaptive evaluation method and system | |
| CN115905835B (en) | Low-voltage alternating current arc fault diagnosis method integrating multidimensional features | |
| Devadasu et al. | Identification of voltage quality problems under different types of Sag/Swell faults with Fast Fourier Transform analysis | |
| CN110514953B (en) | Power angle and voltage aliasing-based power grid fault simulation identification method and system | |
| Shalalfeh et al. | Modeling of PMU data using ARFIMA models | |
| CN105116323A (en) | Motor fault detection method based on RBF | |
| CN109241874A (en) | Power signal filtering method in Energy Decomposition | |
| CN111368933A (en) | Power distribution network transient process fault classification method and system based on Softmax regression | |
| CN110702981A (en) | Load switch event detection method and system using classification tree | |
| CN112034232A (en) | Power supply system voltage sag detection method | |
| Du et al. | Sensitive load identification method based on voltage sag monitoring data | |
| Feng et al. | A method for identifying major disturbance sources in a regional grid | |
| CN112345876A (en) | Fault positioning method and system suitable for interval DTU | |
| CN110187167A (en) | A kind of detection method and device of the load switch event based on manifold classification | |
| CN110514884B (en) | Power signal filtering method and system based on delay vector |
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: 20210625 |
|
| RJ01 | Rejection of invention patent application after publication |