CN109581268A - A kind of optical fiber current mutual inductor method for diagnosing faults and device - Google Patents
A kind of optical fiber current mutual inductor method for diagnosing faults and device Download PDFInfo
- Publication number
- CN109581268A CN109581268A CN201811528452.1A CN201811528452A CN109581268A CN 109581268 A CN109581268 A CN 109581268A CN 201811528452 A CN201811528452 A CN 201811528452A CN 109581268 A CN109581268 A CN 109581268A
- Authority
- CN
- China
- Prior art keywords
- value
- data
- mutual inductor
- noise component
- optical fiber
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R35/00—Testing or calibrating of apparatus covered by the other groups of this subclass
- G01R35/02—Testing or calibrating of apparatus covered by the other groups of this subclass of auxiliary devices, e.g. of instrument transformers according to prescribed transformation ratio, phase angle, or wattage rating
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Testing Electric Properties And Detecting Electric Faults (AREA)
- Protection Of Transformers (AREA)
Abstract
A kind of optical fiber current mutual inductor method for diagnosing faults and device; the original optical path of optical fiber current mutual inductor and circuit structure are remained unchanged; mutual inductor output data quantity ' protection 1 data of channel ', ' protection 2 data of channel ', ' measurement channel data ' and quantity of state ' light power meter numerical value ', ' voltage value ', ' temperature value ' are subjected to Correlation method for data processing respectively; and it is compared with the threshold value of setting; start dependent diagnostic program if beyond threshold range, makes early warning and alarm prompt.Integrated, synthesization the diagnostic method of the present invention accurate, real-time detection mutual inductor working condition and can carry out in time early warning and alarm to fault condition.
Description
Technical field
The present invention relates to a kind of optical fiber current mutual inductor method for diagnosing faults and devices, belong to optical fiber current mutual inductor neck
Domain.
Background technique
Mutual inductor is the important equipment for monitoring operation of power networks state, and measurement, monitoring and protection control rely on it in substation
Come information such as electric current, voltages needed for obtaining measurement, metering, protection.There are electric insulation weaknesses, body for conventional electromagnetic mutual inductor
The problems such as product is heavy, dynamic range is small, ferro-resonance over-voltage, it is difficult in a wide range of interior holding certain precision and the linearity.
As the continuous development of digital transformer substation smart grid promotes, it is mutual that optical fiber current mutual inductor relies on it to be different from conventional electromagnetic
The unique advantage of sensor, the favor increasingly built by intelligent power network.But it is further to restrict optical fiber current mutual inductor at present
The problem of principal element of large-scale application is its poor longtime running stability, high failure rate, so research fibre-optic current mutual inductance
The method for diagnosing faults of device has a major impact its performance boost.
Currently, both at home and abroad to the research of optical fiber current mutual inductor method for diagnosing faults be concentrated mainly on based on optical path improve and
Based on two aspect of signal processing.Based on the improved method of optical path such as Publication No. CN105467188A and Publication No.
The Chinese patent of CN107328986A, additionally increase on the basis of optical fiber current mutual inductor routine light channel structure photodetector and
AD sample circuit is monitored and diagnoses for the state to entire optical path, and this method not only makes optical fiber current mutual inductor whole
Body increased costs more exacerbate the complexity of system.In method such as Publication No. CN107784170A based on signal processing
State's patent, using the noise profile of Allan variance research current transformer, to speculate its fault state, this method due to
Allan variance is difficult to analyze the unstable signal of dynamic to accurately identify failure in practical applications and be accurately positioned failure
Source.The Chinese patent of Publication No. CN108627793A is using the method for waveform inversion to the electronics based on Rogowski coil principle
Formula current transformer carries out fault identification, has certain limitation for the common gradual failure diagnosis of mutual inductor.
Recently, it is received using the method that mutual inductor on site operation data carries out monitoring and fault diagnosis more and more
Researcher's note that the mutual inductor internal state quantity monitoring method that the Chinese patent of such as Publication No. CN106093824A proposes
The method for carrying out System For Fault Diagnosis of Instrument Transformers using wavelet transformation described in Chinese patent with Publication No. CN107885179A, though
So be all based on mutual inductor on site operation data it be monitored and diagnosed, but method it is single cannot in real time accurate device carry out therefore
Hinder early warning and alarm, and cannot effectively distinguish electric network fault or mutual inductor faults itself.If rear end protective device is caused to miss
Movement, it will bring huge security risk and economic loss.
Summary of the invention
The technical problem to be solved by the present invention is having overcome the deficiencies of the prior art and provide a kind of optical fiber current mutual inductor
Method for diagnosing faults and device remain unchanged the original optical path of optical fiber current mutual inductor and circuit structure, mutual inductor are exported
Data volume ' protection 1 data of channel ', ' protection 2 data of channel ', ' measurement channel data ' and quantity of state ' light power meter numerical value ',
' voltage value ', ' temperature value ' carry out Correlation method for data processing respectively, and are compared with the threshold value of setting, if beyond threshold range
Start dependent diagnostic program, makes early warning and alarm prompt.Integrated, synthesization the diagnostic method of the present invention can be accurate, real
When detection mutual inductor working condition and early warning and alarm are carried out in time to fault condition.
The object of the invention is achieved by the following technical programs:
A kind of optical fiber current mutual inductor method for diagnosing faults, includes the following steps:
Step 1: judging temperature value, voltage value, the light power meter numerical value in grid operation data, if wherein
When any value is more than corresponding threshold value, start mulual inductor malfunction warning diagnostic program, no person is transferred to step 2；
Step 2: to the measurement channel data in grid operation data and 2 data of channel is protected to carry out wavelet transformation respectively,
Then the data after wavelet transformation are carried out with singular values standard form respectively and obtains corresponding singular value；If the metering after wavelet transformation is logical
Transient current Sudden Changing Rate is more than corresponding threshold value in track data and protection 2 data of channel, and, the singular value of measurement channel data
It is identical to occur the moment with the singular value of protection 2 data of channel, and, temperature value described in step 1, voltage value, optical power count
When value is less than corresponding threshold value, start power grid diagnostic program；
Based on 1 data of protection channel in grid operation data, is calculated using dynamic Allan Variance Method and obtain noise point
The noise coefficient of amount starts mulual inductor malfunction early warning program if the noise coefficient of noise component(s) is more than corresponding threshold value.
Above-mentioned optical fiber current mutual inductor method for diagnosing faults, the temperature value in the step 1 includes acquisition unit temperature value
With sensing unit temperature value；The measurement point position of the acquisition unit temperature value is in indoor or outdoor air-conditioning cabinet, acquisition unit temperature
The normality threshold of angle value is -40 DEG C~65 DEG C, and, the daily range of temperature of acquisition unit temperature value is more than or equal to 0.5 DEG C
And it is less than or equal to 30 DEG C；
The measurement point position of the sensing unit temperature value is in an outdoor high-pressure side, the normality threshold of sensing unit temperature value
It is -60 DEG C~70 DEG C, and, the daily range of temperature of sensing unit temperature value is more than or equal to 0.5 DEG C and is less than or equal to 30 DEG C.
Above-mentioned optical fiber current mutual inductor method for diagnosing faults, the voltage value in the step 1 include the driving of SLD light source
Voltage value and circuit board power supply value；The normality threshold of the driving voltage value of the SLD light source is 2V~3V；The circuit board power supply
The normality threshold of value is 4V~6V.
Above-mentioned optical fiber current mutual inductor method for diagnosing faults, the normality threshold of light power meter numerical value is to visit in the step 1
It surveys effective optical power that device receives and accounts for the 40%~100% of total optical power.
Above-mentioned optical fiber current mutual inductor method for diagnosing faults, the noise component(s) in the step 2 include quantizing noise point
Amount, angle random walk noise component(s), zero bias unstability noise component(s), angular speed random walk noise component(s), rate ramp
Noise component(s) and sinusoidal noise component；If the noise coefficient of any noise component(s) is more than corresponding threshold value, starting is mutual
Sensor fault pre-alarming diagnostic program.
A kind of optical fiber current mutual inductor trouble-shooter, including the diagnosis of the first diagnostic module, second opinion module, third
Module；
First diagnostic module judges temperature value, voltage value, the light power meter numerical value in grid operation data,
If any numerical value is more than corresponding threshold value, start mulual inductor malfunction warning diagnostic program；
The second opinion module is to the measurement channel data in grid operation data and 2 data of channel is protected to carry out respectively
Then wavelet transformation carries out singular values standard form respectively to the data after wavelet transformation and obtains corresponding singular value；If wavelet transformation
Transient current Sudden Changing Rate is more than corresponding threshold value in measurement channel data and protection 2 data of channel afterwards, and, measurement channel number
According to singular value with protection 2 data of channel singular value occur the moment it is identical, and, temperature value described in the first diagnostic module,
When voltage value, light power meter numerical value are less than corresponding threshold value, start power grid diagnostic program；
The third diagnostic module utilizes the variance side dynamic Allan based on 1 data of protection channel in grid operation data
Method calculates the noise coefficient for obtaining noise component(s), if the noise coefficient of noise component(s) is more than corresponding threshold value, starts mutual inductance
Device fault pre-alarming diagnostic program.
Above-mentioned optical fiber current mutual inductor trouble-shooter, the temperature value in first diagnostic module includes acquisition unit
Temperature value and sensing unit temperature value；The measurement point position of the acquisition unit temperature value is in indoor or outdoor air-conditioning cabinet, acquisition
The normality threshold of cell temperature value is -40 DEG C~65 DEG C, and, the daily range of temperature of acquisition unit temperature value is more than or equal to
0.5 DEG C and be less than or equal to 30 DEG C；
The measurement point position of the sensing unit temperature value is in an outdoor high-pressure side, the normality threshold of sensing unit temperature value
It is -60 DEG C~70 DEG C, and, the daily range of temperature of sensing unit temperature value is more than or equal to 0.5 DEG C and is less than or equal to 30 DEG C.
Above-mentioned optical fiber current mutual inductor trouble-shooter, the voltage value in first diagnostic module includes SLD light source
Driving voltage value and circuit board power supply value；The normality threshold of the driving voltage value of the SLD light source is 2V~3V；The circuit
The normality threshold of plate power supply value is 4V~6V.
Above-mentioned optical fiber current mutual inductor trouble-shooter, the normal threshold of light power meter numerical value in first diagnostic module
Value is that the effective optical power received accounts for the 40%~100% of total optical power.
Above-mentioned optical fiber current mutual inductor trouble-shooter, the noise component(s) in the second opinion module include that quantization is made an uproar
Sound component, angle random walk noise component(s), zero bias unstability noise component(s), angular speed random walk noise component(s), rate
Slope noise component(s) and sinusoidal noise component；If the noise coefficient of any noise component(s) is more than corresponding threshold value, open
Dynamic mulual inductor malfunction early warning diagnostic program.
The present invention has the following beneficial effects: compared with the prior art
(1) it identifies that fineness is not high, diagnostic method is single the present invention overcomes existing fiber current mulual inductor malfunction to ask
Topic.By in optical fiber current mutual inductor field operational data data volume and quantity of state analyze respectively, may be implemented to it not
With the positioning of fault type, that is, temperature sensor fault, fault, optical path failure；
(2) present invention can utilize optical fiber current mutual inductor on the basis of not increasing optical path and circuit structure complexity
Live real-time running data carries out the fault self-diagnosis that analysis can be more convenient, efficiently realize optical fiber current mutual inductor；
(3) present invention makes full use of the method advantage of wavelet transformation, the failure of sign mutation when accurately extracting electric network fault
Feature can identify and distinguish between electric network fault and optical fiber current mutual inductor faults itself；
(4) different types of fault location not only may be implemented in the present invention, while utilizing dynamic Allan variance (DAVAR)
The advantage of analysis accomplishes fault pre-alarming when deterioration occurs for optical fiber current mutual inductor performance in time.
Detailed description of the invention
Fig. 1 is the step flow chart of the method for the present invention；
Fig. 2 is the step flow chart of the embodiment of the present invention；
Fig. 3 is the flow chart that Wavelet Transformation Algorithm of the embodiment of the present invention carries out fault diagnosis；
Fig. 4 is that the embodiment of the present invention utilizes dynamic Allan variance analysis sample data structural schematic diagram.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to implementation of the invention
Mode is described in further detail.
A kind of optical fiber current mutual inductor method for diagnosing faults, as shown in Figure 1, including the following steps:
Step 101, the temperature value in grid operation data, voltage value, light power meter numerical value judge, if wherein
When any value is more than corresponding threshold value, start mulual inductor malfunction warning diagnostic program, no person is transferred to step 2.
Temperature value in the step 1 includes acquisition unit temperature value and sensing unit temperature value；The acquisition unit temperature
For the measurement point position of angle value in indoor or outdoor air-conditioning cabinet, the normality threshold of acquisition unit temperature value is -40 DEG C~65 DEG C, and,
The daily range of temperature of acquisition unit temperature value is more than or equal to 0.5 DEG C and is less than or equal to 30 DEG C；The sensing unit temperature
The measurement point position of value is -60 DEG C~70 DEG C in an outdoor high-pressure side, the normality threshold of sensing unit temperature value, and, sensing is single
The daily range of temperature of first temperature value is more than or equal to 0.5 DEG C and is less than or equal to 30 DEG C.Voltage value packet in the step 1
Include the driving voltage value and circuit board power supply value of SLD light source；The normality threshold of the driving voltage value of the SLD light source be 2V~
3V；The normality threshold of the circuit board power supply value is 4V~6V.The normality threshold of light power meter numerical value is to connect in the step 1
The effective optical power received accounts for the 40%~100% of total optical power.
Step 102, in grid operation data measurement channel data and protection 2 data of channel carry out small echo change respectively
It changes, then the data after wavelet transformation is carried out with singular values standard form respectively and obtains corresponding singular value；If the meter after wavelet transformation
Measuring transient current Sudden Changing Rate in channel data and protection 2 data of channel is more than corresponding threshold value, and, the surprise of measurement channel data
The singular value generation moment of different value and protection 2 data of channel is identical, and, temperature value described in step 101, voltage value, light function
When rate count value is less than corresponding threshold value, start power grid diagnostic program；
Based on 1 data of protection channel in grid operation data, is calculated using dynamic Allan Variance Method and obtain noise point
The noise coefficient of amount, the noise component(s) include that quantization noise components, angle random walk noise component(s), zero bias unstability are made an uproar
Sound component, angular speed random walk noise component(s), rate ramp noise component(s) and sinusoidal noise component；If any noise
When the noise coefficient of component is more than corresponding threshold value, start mulual inductor malfunction early warning diagnostic program.
A kind of optical fiber current mutual inductor trouble-shooter, including the diagnosis of the first diagnostic module, second opinion module, third
Module；
First diagnostic module judges temperature value, voltage value, the light power meter numerical value in grid operation data,
If any numerical value is more than corresponding threshold value, start power grid diagnostic program.Temperature value in first diagnostic module includes
Acquisition unit temperature value and sensing unit temperature value；The measurement point position of the acquisition unit temperature value is in indoor or outdoor air-conditioning cabinet
In, the normality threshold of acquisition unit temperature value is -40 DEG C~65 DEG C, and, the daily range of temperature of acquisition unit temperature value
More than or equal to 0.5 DEG C and it is less than or equal to 30 DEG C；The measurement point position of the sensing unit temperature value is in an outdoor high-pressure side, sensing
The normality threshold of cell temperature value is -60 DEG C~70 DEG C, and, the daily range of temperature of sensing unit temperature value is more than or equal to
0.5 DEG C and be less than or equal to 30 DEG C.Voltage value in first diagnostic module includes driving voltage value and the circuit board power supply of SLD light source
Value；The normality threshold of the driving voltage value of the SLD light source is 2V~3V；The normality threshold of the circuit board power supply value be 4V~
6V.In first diagnostic module the normality threshold of light power meter numerical value be the effective optical power received account for total optical power 40%~
100%.
The second opinion module is to the measurement channel data in grid operation data and 2 data of channel is protected to carry out respectively
Then wavelet transformation carries out singular values standard form respectively to the data after wavelet transformation and obtains corresponding singular value；If wavelet transformation
Transient current Sudden Changing Rate is more than corresponding threshold value in measurement channel data and protection 2 data of channel afterwards, and, measurement channel number
According to singular value with protection 2 data of channel singular value occur the moment it is identical, and, temperature value described in the first diagnostic module,
When voltage value, light power meter numerical value are less than corresponding threshold value, start power grid diagnostic program；
The third diagnostic module utilizes the variance side dynamic Allan based on 1 data of protection channel in grid operation data
Method calculates the noise coefficient for obtaining noise component(s), if the noise coefficient of noise component(s) is more than corresponding threshold value, starts mutual inductance
Device fault pre-alarming diagnostic program.
Noise component(s) in second opinion module include quantization noise components, angle random walk noise component(s), zero bias not
Stability noise component(s), angular speed random walk noise component(s), rate ramp noise component(s) and sinusoidal noise component；If wherein
When the noise coefficient of any noise component(s) is more than corresponding threshold value, start mulual inductor malfunction early warning diagnostic program.
Embodiment:
A kind of optical fiber current mutual inductor method for diagnosing faults based on field operational data analysis of the invention, such as Fig. 2 institute
Show, detailed description are as follows:
In the present embodiment, quantity of state ' temperature value ' is defaulted as 4 bytes, the first two byte representation secondary acquisition device temperature
Data, latter two byte representation primary transducer temperature data.Constructed method for diagnosing faults, specific as follows:
It is hexadecimal data due to generally exporting, data processing first is carried out to ' temperature value ', acquisition will be acquired every time
Temperature data be separately converted to metric temperature value T1And T2。
Set temperature judgment threshold range is respectively to gatherWith
(1) ifOrThen directly initiate diagnostic program, displays temperature sensor fault；
Temperature change in every 24 hours is poor, i.e. temperature maximum T1max、T2maxWith minimum value T1min、T2minDifference note
Make △ T1With △ T2,
△T1=T1max-T1min
△T2=T2max-T2min
Set temperature minimum change and maximum variable quantity threshold value are respectively tmaxAnd tmin,
(2) if △ T1>tmaxOr △ T1<tminOr △ T2<tminOr △ T2>tmax, then diagnostic program is directly initiated, is shown
Temperature sensor fault.
(3) if above-mentioned condition (1) is invalid, i.e., temperature value within the set threshold range, but has in condition (2) any
It sets up, i.e., temperature change is unsatisfactory for the setting upper limit or lower limit, then equally directly initiates diagnostic program, the event of displays temperature sensor
Barrier.
(4) (1) more than, (2), (3) return to progress temperature data processing next time and threshold value if all invalid
Determine, until startup separator alert program.
In the present embodiment, quantity of state ' voltage value ' default is also 4 bytes, the first two byte representation circuit board power supply electricity
Press data, latter two byte representation SLD light source drive voltage data.Constructed method for diagnosing faults, specific as follows:
It is hexadecimal data due to generally exporting, data processing first is carried out to ' voltage value ', acquisition will be acquired every time
Voltage data be separately converted to metric voltage value V1And V2。
Setting voltage judgment threshold range is respectively to gather { ψ1And { ψ2,
(1) ifOrThen directly initiate diagnostic program, display circuit failure；
(2) if above-mentioned condition is invalid, decision circuitry is normal, comes back for voltage data processing next time and threshold value
Determine, until startup separator alert program.
In the present embodiment, quantity of state ' light power meter numerical value ' is defaulted as 2 bytes, and expression is that detector can receive
Effective interference light intensity ratio, constructed method for diagnosing faults is specific as follows:
It is hexadecimal data due to generally exporting, data processing first is carried out to ' light power meter numerical value ', will be adopted every time
The optical power enumeration data that collection obtains is converted into metric light power meter numerical value p.
Set light power meter numerical value judgment threshold range as gather { P },
(1) ifDiagnostic program is then directly initiated, shows optical path failure；
(2) if above-mentioned condition is invalid, judge that optical path is normal, come back for optical power next time and count data processing
And threshold determination, until startup separator alert program.
In the present embodiment, data volume ' protection 2 data of channel ' and ' measurement channel data ' default are all 4 bytes, table
The primary current value digital signal for showing optical fiber current mutual inductor output, is denoted as S respectivelyb2(t),Sj(t).It is constructed based on small echo
The method for diagnosing faults of transformation, as shown in figure 3, specific as follows:
(1) wavelet transformation
By signal Sb2(t) db4 wavelet transformation, the Coefficients of Approximation matrix W a after obtaining wavelet transformation are carried outb2And detail coefficients
Matrix W db2；
By signal Sj(t) db4 wavelet transformation, the Coefficients of Approximation matrix W a after obtaining wavelet transformation are carried outjWith detail coefficients square
Battle array Wdj。
(2) modulus maximum and threshold value comparison
Wavelet transformation data window is set as n, then k-th of wavelet transformation detail coefficients in m group detail coefficients are Wdb2m
[k]、WdjM [k '], wherein k, k ' ∈ [1, n]
If met | Wdb2M [k] | ≠ 0, | WdjM [k '] | ≠ 0, then it is modulus maximum point, this value Wd at k and k ' pointb2m
[k]、WdjM [k '] is modulus maximum.
By the threshold value △ W of obtained modulus maximum and settingdb2With △ WdjCompare, if met
Wdb2m[k]>△Wdb2
Wdjm[k’]>△Wdj
Wherein, △ Wdb2=1.5*max | Wdb2m[k]|,
△Wdj=1.5*max | Wdjm[k']|；
(3) compare modulus maximum and correspond to the moment
Compare whether corresponding modulus maximum after this two paths of data carries out wavelet transformation occurs in synchronization, i.e. acquirement mould
The Wd of maximumb2m[k]、WdjM [k '] respectively corresponds moment T1 and T2, if T1=T2, demonstrating the point is signal surprise
Dissimilarity, while recording the unusual point moment.
(4) if above-mentioned condition (1) and (2) meet simultaneously, and quantity of state at this time ' light power meter numerical value ' does not start light
Road fault warning, that is, returning the result is that normal, which starts electric network fault alert program.
(5) if above-mentioned condition (1) and (2) have one to be unsatisfactory for, return re-starts wavelet transformation and condition judgement,
Until startup separator alert program.
Require to need to have to meet this fault diagnosis functions: event does not occur simultaneously for power grid and optical fiber current mutual inductor
Barrier, i.e. hypothesis electric network fault and mulual inductor malfunction are independent from each other event.
In the present embodiment, data volume ' protection 1 data of channel ' is 4 bytes, indicates optical fiber current mutual inductor output one
Secondary side current value signals.The constructed fault diagnosis model based on dynamic Allan variance (DAVAR), as shown in figure 4, specifically
It is as follows:
Allan variance method is a kind of important hand of measurement and evaluation optical fiber current mutual inductor all kinds of errors and noise characteristic
Section, but it is primarily adapted for use in and analyzes stationary signal, and one group of noise coefficient in specific time can be calculated.But
It is that optical fiber current mutual inductor by substation's complex electromagnetic environment, vibration and temperature and humidity etc. due to being influenced for a long time, actual error
Signal then contains many instability features.Dynamic Allan variance (DAVAR) method is the extension of Allan variance method and complete
Kind, essence is the Allan variance for calculating separately signal in different periods, more accurately reflects that each noise component(s) of signal exists
Dynamic change characterization in entire time domain.
Allan variance method can in the time domain calculate optical fiber current mutual inductor data Frequency-domain Stability, and in log
τ-logστThe model i.e. parameter of every random error is recognized on double logarithmic chart using the slope of curve.Allan variance can be with
It is written as:
Wherein, Q is quantizing noise coefficient；N is angle random walk noise coefficient；B is bias unstability noise coefficient；
K is rate random walk noise coefficient；R is rate ramp noise coefficient.The above each type of noise error is in the side Allan
Different slopes is corresponded on difference-correlation time log-log graph, respectively -1, -0.5,0,0.5,1.
It is carried out curve fitting with least square method, the noise coefficient of available above each noise item is as follows:
Allan variance specifically calculates that steps are as follows:
(1) current data is obtained.' protection is logical for output data quantity with fixed sample frequency f to optical fiber current mutual inductor
1 data of road ' it is acquired, L point is sampled altogether, obtains the current sample set [ω that length is L1, ω2..., ωL]；
(2) independent array is generated.By m every in above-mentioned current sample set (m=1,2 ..., M, M < L/2) a current data
It is divided into one group, obtains J independent arrays, J=[L/m], [] indicates to be rounded downwards；
(3) average current data.Every group of initial data is averaged, that is, asks group mean aver (ω (m)), obtains element
For the stochastic variable set of group mean:
In formula, k=1,2 ... J
(4) variance is calculated.Define the duration τ of each arraym=mTs is correlation time, and Allan variance can be as the following formula
It calculates:
Choose different τ, the distribution of available difference noise.
Dynamic Allan variance method (DAVAR) calculates Allan variance, obtaining by setting time t and observation interval τ repeatedly
The whole Allan variances obtained, which gather together, to be analyzed, and the statistics that each noise component(s) of signal changes over time can be obtained
Characteristic.
Seeking DAVAR, specific step is as follows:
(1) an analysis time point t is randomly choosed1To get t=t1；
It (2) is t with central point1, length be W window function be truncated random signal x (t)；
(3) Allan variance is calculated
(4) another analysis time point t is selected2To get t=t2.Need to meet the window function truncated data at the time point simultaneously
It should be with previous time point t1Truncated data is overlapping, then repeatedly step (2)；
(5) set of Allan variance is obtainedCorresponding different analysis
Time point tkWith different observation interval τ, the stability of random signal x (t) real-time measurement is characterized.
(6) it is acquired respectively using what is obtained in each statistical time period tmThe maximum value Q of interior each noise coefficientmmax、
Nmmax、Bmmax、Kmmax、Rmmax；
(7) by the noise coefficient maximum value obtained every time above respectively with the corresponding threshold value △ Q that respectively setsm、△Nm、△
Bm、△KmWith △ RmIt is compared, if there is wherein any two or more exceed threshold value simultaneously, then startup separator early warning program,
Display ' pays attention to maintenance '.
Output number of the method that the present invention uses dynamic Allan variance and wavelet transformation to combine to optical fiber current mutual inductor
According to being analyzed, make full use of dynamic Allan variance and the outstanding time frequency analysis feature of wavelet transformation, can effectively identify therefore
Hinder feature, makes early warning and police instruction.
The present invention is reasonably sentenced using analyzing the quantity of state in optical fiber current mutual inductor output data by setting
Other condition can accurately determine and identify ' optical path fault warning ', ' fault alarm ' and ' temperature sensor fault alarm '.
The content that description in the present invention is not described in detail belongs to the well-known technique of those skilled in the art.
Claims (10)
1. a kind of optical fiber current mutual inductor method for diagnosing faults, characterized by the following steps:
Step 1: judging temperature value, voltage value, the light power meter numerical value in grid operation data, if any
When numerical value is more than corresponding threshold value, start mulual inductor malfunction warning diagnostic program, no person is transferred to step 2；
Step 2: to the measurement channel data in grid operation data and 2 data of channel is protected to carry out wavelet transformation respectively, then
Data after wavelet transformation are carried out with singular values standard form respectively and obtains corresponding singular value；If the measurement channel number after wavelet transformation
According to protection 2 data of channel in transient current Sudden Changing Rate be more than corresponding threshold value, and, the singular value of measurement channel data and guarantor
Protecting the singular values of 2 data of channel, to occur the moment identical, and, temperature value described in step 1, voltage value, light power meter numerical value are equal
When being less than corresponding threshold value, start power grid diagnostic program；
Based on 1 data of protection channel in grid operation data, is calculated using dynamic Allan Variance Method and obtain noise component(s)
Noise coefficient starts mulual inductor malfunction early warning program if the noise coefficient of noise component(s) is more than corresponding threshold value.
2. a kind of optical fiber current mutual inductor method for diagnosing faults according to claim 1, it is characterised in that: the step 1
In temperature value include acquisition unit temperature value and sensing unit temperature value；The measurement point position of the acquisition unit temperature value is in family
In interior or outdoor air-conditioning cabinet, the normality threshold of acquisition unit temperature value is -40 DEG C~65 DEG C, and, acquisition unit temperature value is daily
Range of temperature be more than or equal to 0.5 DEG C and be less than or equal to 30 DEG C；
The measurement point position of the sensing unit temperature value in an outdoor high-pressure side, the normality threshold of sensing unit temperature value is-
60 DEG C~70 DEG C, and, the daily range of temperature of sensing unit temperature value is more than or equal to 0.5 DEG C and is less than or equal to 30 DEG C.
3. a kind of optical fiber current mutual inductor method for diagnosing faults according to claim 1, it is characterised in that: the step 1
In voltage value include SLD light source driving voltage value and circuit board power supply value；The driving voltage value of the SLD light source it is normal
Threshold value is 2V~3V；The normality threshold of the circuit board power supply value is 4V~6V.
4. a kind of optical fiber current mutual inductor method for diagnosing faults according to claim 1, it is characterised in that: the step 1
The normality threshold of middle light power meter numerical value is that effective optical power that detector receives accounts for the 40%~100% of total optical power.
5. a kind of optical fiber current mutual inductor method for diagnosing faults according to claim 1, it is characterised in that: the step 2
In noise component(s) include quantization noise components, angle random walk noise component(s), zero bias unstability noise component(s), angular speed
Random walk noise component(s), rate ramp noise component(s) and sinusoidal noise component；If the noise system of any noise component(s)
When number is more than corresponding threshold value, start mulual inductor malfunction early warning diagnostic program.
6. a kind of optical fiber current mutual inductor trouble-shooter, it is characterised in that: including the first diagnostic module, second opinion mould
Block, third diagnostic module；
First diagnostic module judges temperature value, voltage value, the light power meter numerical value in grid operation data, if
When any numerical value is more than corresponding threshold value, start mulual inductor malfunction warning diagnostic program；
The second opinion module is to the measurement channel data in grid operation data and 2 data of channel is protected to carry out small echo respectively
Then transformation carries out singular values standard form respectively to the data after wavelet transformation and obtains corresponding singular value；If after wavelet transformation
Transient current Sudden Changing Rate is more than corresponding threshold value in measurement channel data and protection 2 data of channel, and, measurement channel data
Singular value is identical with the singular value generation moment of protection 2 data of channel, and, temperature value described in the first diagnostic module, voltage
When value, light power meter numerical value are less than corresponding threshold value, start power grid diagnostic program；
The third diagnostic module utilizes dynamic Allan Variance Method meter based on 1 data of protection channel in grid operation data
The noise coefficient for obtaining noise component(s) is calculated, if the noise coefficient of noise component(s) is more than corresponding threshold value, starting mutual inductor event
Hinder early warning diagnostic program.
7. a kind of optical fiber current mutual inductor trouble-shooter according to claim 6, it is characterised in that: described first examines
Temperature value in disconnected module includes acquisition unit temperature value and sensing unit temperature value；The measurement point of the acquisition unit temperature value
In indoor or outdoor air-conditioning cabinet, the normality threshold of acquisition unit temperature value is -40 DEG C~65 DEG C, and, acquisition unit temperature
It is worth daily range of temperature more than or equal to 0.5 DEG C and is less than or equal to 30 DEG C；
The measurement point position of the sensing unit temperature value in an outdoor high-pressure side, the normality threshold of sensing unit temperature value is-
60 DEG C~70 DEG C, and, the daily range of temperature of sensing unit temperature value is more than or equal to 0.5 DEG C and is less than or equal to 30 DEG C.
8. a kind of optical fiber current mutual inductor trouble-shooter according to claim 6, it is characterised in that: described first examines
Voltage value in disconnected module includes the driving voltage value and circuit board power supply value of SLD light source；The driving voltage value of the SLD light source
Normality threshold be 2V~3V；The normality threshold of the circuit board power supply value is 4V~6V.
9. a kind of optical fiber current mutual inductor trouble-shooter according to claim 6, it is characterised in that: described first examines
The normality threshold of light power meter numerical value is that the effective optical power received accounts for the 40%~100% of total optical power in disconnected module.
10. a kind of optical fiber current mutual inductor trouble-shooter according to claim 6, it is characterised in that: described second
Noise component(s) in diagnostic module includes quantization noise components, angle random walk noise component(s), zero bias unstability noise point
Amount, angular speed random walk noise component(s), rate ramp noise component(s) and sinusoidal noise component；If any noise component(s)
Noise coefficient be more than corresponding threshold value when, start mulual inductor malfunction early warning diagnostic program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811528452.1A CN109581268B (en) | 2018-12-13 | 2018-12-13 | Fault diagnosis method and device for optical fiber current transformer |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811528452.1A CN109581268B (en) | 2018-12-13 | 2018-12-13 | Fault diagnosis method and device for optical fiber current transformer |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109581268A true CN109581268A (en) | 2019-04-05 |
| CN109581268B CN109581268B (en) | 2021-06-11 |
Family
ID=65928526
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811528452.1A Active CN109581268B (en) | 2018-12-13 | 2018-12-13 | Fault diagnosis method and device for optical fiber current transformer |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109581268B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619727A (en) * | 2019-09-10 | 2019-12-27 | 珠海市联电科技有限公司 | Underground power cable external damage prevention early warning system based on optical fiber vibration sensing technology |
| CN110988431A (en) * | 2019-11-28 | 2020-04-10 | 北京自动化控制设备研究所 | Optical fiber current transformer with high integration level |
| CN112180315A (en) * | 2020-09-27 | 2021-01-05 | 国网江苏省电力有限公司电力科学研究院 | Fault feature extraction method, device and system for optical fiber current transformer |
| CN112557991A (en) * | 2020-11-12 | 2021-03-26 | 珠海一多监测科技有限公司 | Current transformer fault diagnosis method based on mole number and temperature |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101666841A (en) * | 2009-09-18 | 2010-03-10 | 重庆大学 | Fault diagnosis method of digitalizing transformer substation electronic transducer |
| CN106093824A (en) * | 2016-05-31 | 2016-11-09 | 全球能源互联网研究院 | A kind of internal state quantity monitoring method of electronic mutual inductor |
| CN106597347A (en) * | 2016-12-08 | 2017-04-26 | 中国电力科学研究院 | Method and system for online monitoring state of fiber-optical current transducer and self-diagnosis |
| CN107885179A (en) * | 2016-09-30 | 2018-04-06 | 中国电力科学研究院 | A kind of electronic mutual inductor exception discriminating method |
-
2018
- 2018-12-13 CN CN201811528452.1A patent/CN109581268B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101666841A (en) * | 2009-09-18 | 2010-03-10 | 重庆大学 | Fault diagnosis method of digitalizing transformer substation electronic transducer |
| CN106093824A (en) * | 2016-05-31 | 2016-11-09 | 全球能源互联网研究院 | A kind of internal state quantity monitoring method of electronic mutual inductor |
| CN107885179A (en) * | 2016-09-30 | 2018-04-06 | 中国电力科学研究院 | A kind of electronic mutual inductor exception discriminating method |
| CN106597347A (en) * | 2016-12-08 | 2017-04-26 | 中国电力科学研究院 | Method and system for online monitoring state of fiber-optical current transducer and self-diagnosis |
Non-Patent Citations (6)
| Title |
|---|
| LIHUIWANG等: ""Methods of Fault Diagnosis in Fiber Optic Current Transducer Based on Allan Variance"", 《HINDAWI PUBLISHING CORPORATION》 * |
| 周刚 等: ""基于Allan方差理论的光纤电流互感器运行状态诊断_周刚"", 《高压电器》 * |
| 熊小伏 等: ""基于小波变换的数字化变电站电子式互感器突变性故障诊断方法"", 《电网技术》 * |
| 王梓: ""光纤电流互感器信号处理方法的研究"", 《中国优秀硕士学位论文全文数据库 工程科技II辑》 * |
| 胡蓓 等: ""全光纤电流互感器关键状态量及其监控方法"", 《高电压技术》 * |
| 赵双双 等: ""电子式互感器状态评价及可靠性分析"", 《理论与方法》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110619727A (en) * | 2019-09-10 | 2019-12-27 | 珠海市联电科技有限公司 | Underground power cable external damage prevention early warning system based on optical fiber vibration sensing technology |
| CN110988431A (en) * | 2019-11-28 | 2020-04-10 | 北京自动化控制设备研究所 | Optical fiber current transformer with high integration level |
| CN112180315A (en) * | 2020-09-27 | 2021-01-05 | 国网江苏省电力有限公司电力科学研究院 | Fault feature extraction method, device and system for optical fiber current transformer |
| CN112557991A (en) * | 2020-11-12 | 2021-03-26 | 珠海一多监测科技有限公司 | Current transformer fault diagnosis method based on mole number and temperature |
| CN112557991B (en) * | 2020-11-12 | 2024-05-28 | 珠海一多监测科技有限公司 | Current transformer fault diagnosis method based on mole number and temperature |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109581268B (en) | 2021-06-11 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109581268A (en) | A kind of optical fiber current mutual inductor method for diagnosing faults and device | |
| CN110275094B (en) | High-voltage cable partial discharge acousto-optic combined detection method and system | |
| CN100561246C (en) | Load behavior is discerned the method for generator rotor interturn short-circuit fault down automatically | |
| CN202066932U (en) | Potable partial-discharge ultrasonic cloud detection device | |
| CN105512474A (en) | Transformer state monitoring data anomaly detection method | |
| CN102298107A (en) | Portable ultrasonic wave and cloud detection apparatus for partial discharge | |
| Yu et al. | Nonintrusive appliance load monitoring for smart homes: Recent advances and future issues | |
| CN105606975A (en) | Localizable ultrahigh-frequency cable partial discharge detection method and device | |
| CN106124939A (en) | Distributed high tension cable partial discharge monitoring and alignment system | |
| CN108828406A (en) | The fault recognition method and its system of non-intrusion type user power utilization | |
| CN106646034B (en) | A kind of arrester structural health on-line monitoring analysis system | |
| CN105093059B (en) | A kind of transformer winding working state detecting method based on normalization singular spectrum entropy | |
| CN110085006A (en) | Ship monitoring method, device, system and storage medium | |
| CN106241538A (en) | A kind of elevator energy consumption on-line monitoring system based on Internet of Things | |
| CN103954815A (en) | Electricity larceny prevention check main machine | |
| CN102353877A (en) | Partial discharge detection method for realizing phase resolved partial discharge (PRPD) pattern and time resolved partial discharge (TRPD) pattern | |
| CN106093679A (en) | A kind of cable on-line checking and fault diagnosis system | |
| CN104391501B (en) | A kind of condition monitoring system and method for Supervisory Computer Control System For Hydroelectic Power Plant | |
| CN214375062U (en) | Lightning arrester operation performance on-line monitoring system | |
| CN112381467A (en) | Artificial intelligence line loss analysis system based on big data of distribution station district total collection | |
| CN115618205A (en) | Portable voiceprint fault detection system and method | |
| CN117394311B (en) | Power distribution network toughness assessment and emergency control method based on multi-source information fusion | |
| CN117607784A (en) | Ammeter operation error real-time monitoring system | |
| CN117420393A (en) | Method, system, equipment and storage medium for identifying partial discharge defect of switch cabinet | |
| CN107015184A (en) | Superfrequency Partial Discharge Detection performance judgment method based on data analysis |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |