CN107238778A - A kind of method and system for recognizing DC Traction Network fault current - Google Patents
A kind of method and system for recognizing DC Traction Network fault current Download PDFInfo
- Publication number
- CN107238778A CN107238778A CN201610573047.6A CN201610573047A CN107238778A CN 107238778 A CN107238778 A CN 107238778A CN 201610573047 A CN201610573047 A CN 201610573047A CN 107238778 A CN107238778 A CN 107238778A
- Authority
- CN
- China
- Prior art keywords
- mrow
- frequency
- time
- current signal
- signal
- 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
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
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/088—Aspects of digital computing
-
- 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
- G01R31/08—Locating faults in cables, transmission lines, or networks
- G01R31/081—Locating faults in cables, transmission lines, or networks according to type of conductors
- G01R31/086—Locating faults in cables, transmission lines, or networks according to type of conductors in power transmission or distribution networks, i.e. with interconnected conductors
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Testing Of Short-Circuits, Discontinuities, Leakage, Or Incorrect Line Connections (AREA)
Abstract
The present invention relates to a kind of method and system for recognizing DC Traction Network fault current, methods described includes：The current signal collected is subjected to local mean value decomposition and obtains multiple multiplicative function components；The instantaneous amplitude of all multiplicative function components and instantaneous frequency are combined, the time-frequency distributions of the current signal are obtained；The time-frequency entropy of the current signal is calculated according to the time-frequency distributions；The characteristic quantity of the time-frequency entropy and short-circuit current signal is compared, recognize the current signal whether be DC Traction Network short-circuit current.The present invention program can improve the frequent malfunction problem of DC Traction Network protection for feed line, it is ensured that the safe and stable operation of DC traction power-supply system.
Description
Technical field
The present invention relates to electrical network field, more particularly to a kind of method and system for recognizing DC Traction Network fault current.
Background technology
Urban track traffic is using electric power as traction power, and the electric energy needed for it is directly derived from DC Traction Network.Engineering
Experience have shown that, the fault rate of DC Traction Network is higher and approaching failure is big, therefore, reliable DC Traction Network protection for feed line pair
Ensure that safety of urban transit operation is most important.
At present, the main protection of DC Traction Network feeder line is protected using di/dt- Δs I (current changing rate-current increment), its energy
Effective district separating vehicles starting current and Traction networks short circuit current flow.However, due to urban track traffic carrying capacity lifting, drive a vehicle it is close
Degree increase and many cars such as start at the reason, cause DC Traction Network oscillating current occur, its increment and climbing feature all close to
Traction networks short circuit current flow, easily causes protection for feed line misoperation.Also, it is found that and much shakes in DC Traction Network failure wave-recording
Current waveform is swung, understands that short trouble does not occur for the period DC Traction Network through scene investigation, this explanation oscillating current has been made
Into the frequent misoperation of protection for feed line, and have influence on the normal operation of urban track traffic.
The content of the invention
Based on this, in order to be able to accurately identify DC Traction Network short-circuit current, it is ensured that the peace of DC traction power-supply system
There is provided a kind of method and system for recognizing DC Traction Network fault current for full stable operation.
A kind of method for recognizing DC Traction Network fault current, including：The current signal collected is subjected to local mean value
Decompose (LDM) and obtain multiple multiplicative function components；The instantaneous amplitude of all multiplicative function components and instantaneous frequency are combined,
Obtain the time-frequency distributions of the current signal；The time-frequency entropy of the current signal is calculated according to the time-frequency distributions；When will be described
The characteristic quantity of frequency entropy and short-circuit current signal is compared, and whether recognize the current signal is DC Traction Network short circuit event
Hinder electric current.
One embodiment, carries out local mean value decomposition by the current signal collected and obtains multiple multiplicative functions point wherein
The step of amount, includes：The current signal collected decompose obtaining the first multiplicative function component；Collected described
Current signal subtract the first multiplicative function component and obtain the first residual signal；It regard first residual signal as k signals；Will
The k signals decompose obtaining the second multiplicative function component；The k signals are subtracted into the second multiplicative function component and obtain second
Residual signal；Whether detect second residual signal is monotonic function；If it is not, updating k signals with second residual signal；
Circulation performs previous step, untill the second obtained residual signal is a monotonic function.
One embodiment, the instantaneous amplitude of all multiplicative function components and instantaneous frequency are combined, obtained wherein
The step of time-frequency distributions of the current signal, includes：
In formula, L represents the time-frequency distributions of the current signal, and n is the number of PF components, and f represents the multiplicative function point
The instantaneous frequency of amount, t represents time, ai(t) instantaneous amplitude of i-th of multiplicative function component, f are representedi(t) i-th of product is represented
The instantaneous frequency of function component.
One embodiment, includes according to the step of the time-frequency entropy of the time-frequency distributions calculating current signal wherein：
The time-frequency distributions are divided into the time frequency block of N (N=2,3,4 ...) individual area equation, the energy of every piece of time frequency block is Wi(i=
1 ..., N), to WiIt is normalizedThe energy of wherein whole time-frequency distributions is W, whereinObtain described
The time-frequency entropy of the current signal collected is：
One embodiment, the current signal collected decompose obtaining the first multiplicative function component wherein
Step includes：
Step 1：It is determined that all Local Extremums of the current signal collected；
Step 2：The average value of two neighboring extreme point is calculated as corresponding local mean value, two neighboring extreme point is calculated
Difference absolute value divided by 2 be used as corresponding envelope estimate；
Step 3：All adjacent local mean value straight line connection smoothing processings are obtained into local mean value function；Will be all adjacent
Envelope estimate point straight line connection smoothing processing obtain envelope estimation function；
Or, all adjacent local mean value broken line connection smoothing processings are obtained into local mean value function；Will be all adjacent
Envelope estimate point broken line connection smoothing processing obtain envelope estimation function；
Step 4：The current signal collected is subtracted into the local mean value function and obtains intermediate function；In described
Between function divided by the envelope estimation function obtain FM signal；
Step 5：Using described in FM signal as new current signal, it is determined that all local extremums of new current signal
Point, repeat step 2-4 obtains new envelope estimation function and new FM signal；
Step 6：Judge whether the new current signal is pure FM signal according to new envelope estimation function；If it is not,
Repeat step 5；If so, performing step 7；
Step 7：All envelope estimation functions produced in iterative process are multiplied and obtain envelope signal；
Step 8：The envelope signal is multiplied with the pure FM signal and obtains the first multiplicative function component.
One embodiment, the instantaneous amplitude of all multiplicative function components and instantaneous frequency are combined, obtained wherein
Also include before the step of time-frequency distributions of the current signal：
Calculate multiplicative function component instantaneous frequency be：
Wherein, f1(t) instantaneous frequency of first multiplicative function component, s are represented1n(t) pure FM signal.
One embodiment, judges whether the new current signal is pure frequency modulation according to new envelope estimation function wherein
Signal includes：One iteration error Δ e is set, if the new envelope estimation function and 1 difference (- Δ e ,+Δ e) are interior,
It is pure FM signal then to judge the new current signal.
The present invention also provides a kind of system for recognizing DC Traction Network fault current, including：Local mean value decomposing module, multiply
Product function component combination module, time-frequency entropy computing module, time-frequency entropy and characteristic quantity comparison module；The local mean value decomposing module
The current signal collected is decomposed and obtains multiple multiplicative function components；The multiplicative function component combination module will be described all
Instantaneous amplitude and the instantaneous frequency combination of multiplicative function component, obtain the time-frequency distributions of the current signal；The time-frequency entropy meter
Calculate the time-frequency entropy that module calculates the current signal according to the time-frequency distributions；The time-frequency entropy and characteristic quantity comparison module are by institute
The characteristic quantity for stating time-frequency entropy and short-circuit current signal is compared, and recognizes whether the current signal is DC Traction Network
Short-circuit current.
One embodiment wherein, the multiplicative function component combination module is instantaneous by all multiplicative function components
Amplitude and instantaneous frequency combination, obtaining the time-frequency distributions of the current signal includes：
In formula, L represents the time-frequency distributions of the current signal, and n is the number of multiplicative function component, and f represents the product
The instantaneous frequency of function component, t represents time, ai(t) instantaneous amplitude of i-th of multiplicative function component, f are representedi(t) i-th is represented
The instantaneous frequency of individual multiplicative function component.
One embodiment wherein, the time-frequency entropy computing module calculates the current signal according to the time-frequency distributions
Time-frequency entropy includes：The time-frequency distributions are divided into the time-frequency of N (N=2,3,4 ...) individual area equation by the time-frequency entropy computing module
Block, the energy of every piece of time frequency block is Wi(i=1 ..., N), to WiIt is normalizedWherein W is whole time-frequency distributions
Energy,The time-frequency entropy for obtaining the current signal collected is：
Wherein, the current signal time-frequency entropy collected described in s (q) expressions.
The beneficial effect of this programme：The time-frequency entropy of the current signal gathered is decomposed based on local mean value, is adopted described
The time-frequency entropy of the current signal of collection and the characteristic quantity of short-circuit current signal are compared, and can accurately identify the current signal
Whether be DC Traction Network short-circuit current, it is ensured that the safe and stable operation of DC traction power-supply system.
Brief description of the drawings
Fig. 1 is the indicative flowchart of the method for the identification DC Traction Network fault current of an embodiment；
Fig. 2 is the indicative flowchart of the method for the identification DC Traction Network fault current of another embodiment；
Fig. 3 is the indicative flowchart of the method for the identification DC Traction Network fault current of another embodiment；
Fig. 4 is the schematic diagram of the system of the identification DC Traction Network fault current of an embodiment；
Fig. 5 is the short-circuit current oscillogram of an embodiment；
Fig. 6 is the LMD time frequency distribution maps of an embodiment；
Fig. 7 is the oscillating current oscillogram of an embodiment；
Fig. 8 is the LMD time frequency distribution maps of another embodiment.
Embodiment
In order to further illustrate the effect of the technological means of the invention taken and acquirement, below in conjunction with the accompanying drawings and preferably
Embodiment, to technical scheme, carries out clear and complete description.
Fig. 1 is the indicative flowchart of the method for the identification DC Traction Network fault current of an embodiment.
As shown in figure 1, a kind of method for recognizing DC Traction Network fault current, including：
S101, carries out local mean value decomposition (LMD) by the current signal x (t) collected, obtains multiple multiplicative function components
PFi(t)。
S102, by all multiplicative function component PFi(t) instantaneous amplitude a1(t) with instantaneous frequency f1(t) combine, obtain institute
State current signal x (t) time-frequency distributions L (f, t)；
In the present embodiment, the time-frequency distributions L (f, t) of the current signal x (t) definition is：
In formula, n is the number of PF components, and f represents the instantaneous frequency of the multiplicative function component.
S103, the time-frequency entropy s (q) of the current signal is calculated according to the time-frequency distributions L (f, t)；
In the present embodiment, the time-frequency distributions are divided into the time frequency block of N (N=2,3,4 ...) individual area equation, every piece
The energy of time frequency block is Wi(i=1 ..., N), the energy of whole time-frequency distributions is W, to WiIt is normalizedWhereinThe time-frequency entropy for obtaining the current signal collected is：
Wherein, the current signal time-frequency entropy collected described in s (q) expressions.
S104, the characteristic quantity of the time-frequency entropy s (q) and the short-circuit current signal of DC Traction Network is compared,
Whether recognize the current signal is DC Traction Network short-circuit current.
In the present embodiment, the characteristic quantity of the time-frequency entropy and short-circuit current signal is compared, identification is described
Current signal whether be DC Traction Network short-circuit current, it is to avoid the frequent misoperation of protection for feed line, it is ensured that DC traction supply
The safe and stable operation of electric system.Such as, the short-circuit current signal of 200ms before protection as shown in Figure 5 starts is gathered, it is right
It carries out LMD and decomposes the time-frequency distributions obtained as described in Figure 6, and the time-frequency distributions calculated as described in Figure 6 obtain time-frequency entropy and are
2.78.The characteristic quantity of other 5 short-circuit current signals, the time-frequency of the current signal by comparing collection are listed in table 1
Entropy and characteristic value, it can be determined that the current signal of collection is short-circuit current signal.Collection protection as shown in Figure 7 starts
Preceding 200ms current oscillation signal, LMD is carried out to it and decomposes the time-frequency distributions obtained as described in Figure 8, is calculated as described in Figure 8
It is 0.96 that time-frequency distributions, which obtain time-frequency entropy,.The characteristic quantity of its 5 short-circuit current signal with being listed in table 1 is compared,
The current signal that may determine that collection is not short-circuit current signal；The spy of other 5 current oscillation signals is listed in table 2
The amount of levying, passes through the time-frequency entropy and characteristic value of the current signal that compares collection, it can be determined that the current signal of collection is vibration electricity
Flow signal.
The characteristic quantity of the short-circuit current of table 1
The characteristic quantity of the oscillating current of table 2
The beneficial effect of the present embodiment includes：Based on the time-frequency entropy of the LMD current signals gathered, by the collection
The time-frequency entropy of current signal and the characteristic quantity of short-circuit current signal are compared, and whether can accurately identify the current signal
For the short-circuit current of DC Traction Network, it is ensured that the safe and stable operation of DC traction power-supply system.
Fig. 2 is the indicative flowchart of the method for the identification DC Traction Network fault current of another embodiment.
Multiple multiplicative function components are obtained as shown in Fig. 2 the current signal x (t) collected is carried out into local mean value and decomposed
The step of include：
S201, the current signal x (t) collected decompose obtaining the first multiplicative function component PF1(t)；
S202, subtracts the first multiplicative function component PF1 by the current signal x (t) collected and obtains the first remaining letter
Number u1(t)；It regard first residual signal as k signals；
S203, the k signals decompose to obtain the second multiplicative function component；
S204, subtracts the second multiplicative function component by the k signals and obtains the second residual signal；
S205, detects whether second residual signal is monotonic function；
S206, if second residual signal is not a monotonic function, k signals are updated with second residual signal；
S203, S204, S205 are repeated, untill the second residual signal is a monotonic function；
S207, the current signal x (t) of collection, which is decomposed, to be finished, and obtains all multiplicative function components of the current signal.
In the present embodiment, the first multiplicative function component PF1(t) highest of the current signal x (t) collected is included
Frequency content, be the AM/FM amplitude modulation/frequency modulation signal of a simple component；The current signal x (t) collected is subtracted into the first product
Function component PF1(t) the first residual signal u is obtained1(t), generally, the first residual signal u1(t) still containing relatively more in
Useful frequency content, so, by the first residual signal u1(t) decomposed as new signal.Until residual signal uk(t)
Untill being a monotonically increasing function or monotonic decreasing function.Now, the current signal x (t) collected is broken down into many
Individual multiplicative function component PFi(t), the current signal x (t) collected can be decomposed obtained all multiplicative function components
With residual signal uk(t) reconstruct, i.e.,
Wherein, PFi(t) it is i-th of multiplicative function component, k is the number of multiplicative function component, uk(t) be monotonic increase or
The residual signal of monotone decreasing.
The beneficial effect of the present embodiment includes：Complicated non-linear, non-stationary current signal x (t) are resolved into limited
Instantaneous frequency has the PF components of physical significance, by the way that the instantaneous amplitude of all PF components and instantaneous frequency are combined, obtains institute
The time-frequency distributions of current signal are stated, the time-frequency entropy of the current signal can be calculated according to time-frequency distributions.
Fig. 3 is the indicative flowchart of the method for the identification DC Traction Network fault current of another embodiment.
The first multiplicative function point is obtained as shown in figure 3, by the current signal collected decompose in the present embodiment
The step of amount, includes：
S301：It is determined that all Local Extremum n of the current signal x (t) collectedi。
S302：The average value of two neighboring extreme point is calculated as corresponding local mean value mi, calculate two neighboring extreme value
The absolute value of the difference of point divided by 2 it is used as corresponding envelope estimate ai, i.e.,：
Wherein, ni、ni+1It is two two adjacent extreme points, miIt is local mean value, aiIt is envelope estimate.
S303：All adjacent local mean value straight line connection smoothing processings are obtained into local mean value function m11(t)；Will be all
Adjacent envelope estimate point straight line connection smoothing processing obtains envelope estimation function a11(t)；
Or, all adjacent local mean value broken line connection smoothing processings are obtained into local mean value function m11(t)；Will be all
Adjacent envelope estimate point broken line connection smoothing processing obtains envelope estimation function a11(t)；
S304：The current signal x (t) collected is subtracted into the local mean value function m11(t) intermediate function is obtained
h11(t), i.e.,：
h11(t)=x (t)-m11(t)；
Then, by the intermediate function h11Divided by the envelope estimation function a (t)11(t) FM signal s is obtained11(t),
I.e.：
S305：By FM signal s11(t) it is described as new current signal, it is determined that all local poles of new current signal
It is worth point, repeat step S302-S304 obtains new envelope estimation function and new FM signal.
S306：Judge whether new current signal is pure FM signal s according to new envelope estimation function1n(t)；If it is not,
Repeat S305；If so, performing S307.
It is preferred that, an iteration error Δ e is set, if the new envelope estimation function and 1 difference are in (- Δ e ,+Δ
E) whether in, then it is pure FM signal s to be judged as new current signal1n(t)。
In the present embodiment, the instantaneous frequency of calculating multiplicative function component is：
S307：All envelope estimation functions produced in iterative process are multiplied and obtain envelope signal a1(t), i.e.,：
Wherein, a11(t)、a12(t)、a1n(t) it is the envelope estimation function that produces in iterative process.
S308：By the envelope signal a1(t) with pure FM signal s1n(t) it is multiplied and obtains the first multiplicative function component PF1,
I.e.：
PF1(t)=a1(t)s1n(t)。
The beneficial effect of the present embodiment：It make use of part mean decomposition method complicated non-linear, non-flat by one
Steady current signal, which adaptively resolves into limited instantaneous frequency, has the PF components of physical significance, with empirical mode decomposition method
Compare, the end effect of part mean decomposition method has obtained certain suppression.
Based on the method identical thought with the identification DC Traction Network fault current in above-described embodiment, the present invention is also carried
System for recognizing DC Traction Network fault current, the system can be used for the side for performing above-mentioned identification DC Traction Network fault current
Method.For convenience of description, in the structural representation of identification DC Traction Network fault current system embodiment, it illustrate only and this
The related part of inventive embodiments, can be with it will be understood by those skilled in the art that the restriction of schematic structure not structure paired systems
Including than illustrating more or less parts, either combining some parts or different parts arrangement.
Fig. 4 is the schematic diagram of the system of the identification DC Traction Network fault current of an embodiment.
As shown in figure 4, a kind of system for recognizing DC Traction Network fault current, including：Local mean value decomposing module 101,
Multiplicative function component combination module 102, time-frequency entropy computing module 103, time-frequency entropy and characteristic quantity comparison module 104；
Local mean value decomposing module 101, multiple multiplicative function components are obtained for the current signal collected to be decomposed；
Multiplicative function component combination module 102, for by the instantaneous amplitude of all multiplicative function components and instantaneous frequency
Rate is combined, and obtains the time-frequency distributions of the current signal；
Time-frequency entropy computing module 103, the electric current is calculated for the time-frequency entropy computing module according to the time-frequency distributions
The time-frequency entropy of signal；
Time-frequency entropy and characteristic quantity comparison module 104, the time-frequency entropy and characteristic quantity comparison module are by the time-frequency entropy and short
The characteristic quantity of road fault-current signal is compared, recognize the current signal whether be DC Traction Network short trouble electricity
Stream.
In the present embodiment, the multiplicative function component combination module 102 is by the instantaneous of all multiplicative function components
Amplitude and instantaneous frequency combination, obtaining the time-frequency distributions of the current signal includes：
In formula, L represents the time-frequency distributions of the current signal, and n is the number of multiplicative function component, and f represents the product
The instantaneous frequency of function component, t represents time, ai(t) instantaneous amplitude of i-th of multiplicative function component, f are representedi(t) i-th is represented
The instantaneous frequency of individual multiplicative function component.
In the present embodiment, the time-frequency entropy computing module calculates the time-frequency of the current signal according to the time-frequency distributions
Entropy includes：The time-frequency distributions are divided into the time frequency block of N (N=2,3,4 ...) individual area equation by the time-frequency entropy computing module, often
The energy of block time frequency block is Wi(i=1 ..., N), to WiIt is normalizedWherein W is the energy of whole time-frequency distributions
Amount,The time-frequency entropy for obtaining the current signal collected is：
Wherein, the current signal time-frequency entropy collected described in s (q) expressions.
The beneficial effect of the present embodiment：The time-frequency entropy of the current signal gathered based on local mean value decomposing module, is led to
Time-frequency entropy and characteristic quantity comparison module are crossed by the feature of the time-frequency entropy of the current signal of the collection and short-circuit current signal
Amount is compared, and it is the short-circuit current of DC Traction Network to recognize the current signal, it is ensured that DC traction power-supply system
Safe and stable operation.
Each technical characteristic of embodiment described above can be combined arbitrarily, to make description succinct, not to above-mentioned reality
Apply all possible combination of each technical characteristic in example to be all described, as long as however, the combination of these technical characteristics is not deposited
In contradiction, the scope of this specification record is all considered to be.
Embodiment described above only expresses the several embodiments of the present invention, and it describes more specific and detailed, but simultaneously
Can not therefore it be construed as limiting the scope of the patent.It should be pointed out that coming for one of ordinary skill in the art
Say, without departing from the inventive concept of the premise, various modifications and improvements can be made, these belong to the protection of the present invention
Scope.Therefore, the protection domain of patent of the present invention should be determined by the appended claims.
Claims (10)
1. a kind of method for recognizing DC Traction Network fault current, it is characterised in that including：
The current signal collected is subjected to local mean value decomposition and obtains multiple multiplicative function components；
The instantaneous amplitude of all multiplicative function components and instantaneous frequency are combined, the when frequency division of the current signal is obtained
Cloth；
The time-frequency entropy of the current signal is calculated according to the time-frequency distributions；
The characteristic quantity of the time-frequency entropy and the short-circuit current signal of DC Traction Network is compared, the electric current letter is recognized
Number whether it is DC Traction Network short-circuit current.
2. the method for identification DC Traction Network fault current according to claim 1, it is characterised in that by the electricity collected
Stream signal, which carries out the step of local mean value decomposition obtains multiple multiplicative function components, to be included：
The current signal collected decompose obtaining the first multiplicative function component；By the current signal collected
Subtract the first multiplicative function component and obtain the first residual signal；It regard first residual signal as k signals；
The k signals decompose to obtain the second multiplicative function component；The k signals are subtracted into the second multiplicative function component to obtain
To the second residual signal；Whether detect second residual signal is monotonic function；If it is not, being updated with second residual signal
K signals；
Circulation performs previous step, untill the second obtained residual signal is a monotonic function.
3. the method for identification DC Traction Network fault current according to claim 1, it is characterised in that all multiply described
The step of instantaneous amplitude and the instantaneous frequency combination of Product function component, time-frequency distributions for obtaining the current signal, includes：
<mrow>
<mi>L</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>,</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>a</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mi>c</mi>
<mi>o</mi>
<mi>s</mi>
<mo>&lsqb;</mo>
<mn>2</mn>
<mi>&pi;</mi>
<mo>&Integral;</mo>
<msub>
<mi>f</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mi>d</mi>
<mi>t</mi>
<mo>&rsqb;</mo>
<mo>;</mo>
</mrow>
In formula, L represents the time-frequency distributions of the current signal, and n is the number of multiplicative function component, and f represents the multiplicative function
The instantaneous frequency of component, t represents time, ai(t) instantaneous amplitude of i-th of multiplicative function component, f are representedi(t) represent to multiply for i-th
The instantaneous frequency of Product function component.
4. the method for identification DC Traction Network fault current according to claim 1, it is characterised in that according to the time-frequency
The step of distribution calculates the time-frequency entropy of the current signal includes：
The time-frequency distributions are divided into the time frequency block of N (N=2,3,4 ...) individual area equation, the energy of every piece of time frequency block is Wi(i=
1 ..., N), to WiIt is normalizedWherein W is the energy of whole time-frequency distributions,Obtain the collection
To the time-frequency entropy of current signal be：
<mrow>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>q</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mo>-</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<msub>
<mi>q</mi>
<mi>i</mi>
</msub>
<mi>ln</mi>
<mi> </mi>
<msub>
<mi>q</mi>
<mi>i</mi>
</msub>
<mo>;</mo>
</mrow>
Wherein, the current signal time-frequency entropy collected described in s (q) expressions.
5. the method for identification DC Traction Network fault current according to claim 2, it is characterised in that collected described
Current signal decompose the step of obtaining the first multiplicative function component and include：
Step 1：It is determined that all Local Extremums of the current signal collected；
Step 2：The average value of two neighboring extreme point is calculated as corresponding local mean value, the difference of two neighboring extreme point is calculated
Absolute value divided by 2 be used as corresponding envelope estimate；
Step 3：All adjacent local mean value straight line connection smoothing processings are obtained into local mean value function；By all adjacent bags
Network estimate point straight line connection smoothing processing obtains envelope estimation function；
Or, all adjacent local mean value broken line connection smoothing processings are obtained into local mean value function；By all adjacent bags
Network estimate point broken line connection smoothing processing obtains envelope estimation function；
Step 4：The current signal collected is subtracted into the local mean value function and obtains intermediate function；By the middle letter
Several divided by described envelope estimation function obtains FM signal；
Step 5：Using described in FM signal as new current signal, it is determined that all Local Extremums of new current signal, weight
Multiple step 2-4, obtains new envelope estimation function and new FM signal；
Step 6：Judge whether the new current signal is pure FM signal according to new envelope estimation function；If it is not, repeating
Step 5；If so, performing step 7；
Step 7：All envelope estimation functions produced in iterative process are multiplied and obtain envelope signal；
Step 8：The envelope signal is multiplied with the pure FM signal and obtains the first multiplicative function component.
6. the method for identification DC Traction Network fault current according to claim 5, it is characterised in that all multiply described
Also wrapped before instantaneous amplitude and the instantaneous frequency combination of Product function component, the step of obtaining the time-frequency distributions of the current signal
Include：
Calculate multiplicative function component instantaneous frequency be：
<mrow>
<msub>
<mi>f</mi>
<mn>1</mn>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mn>2</mn>
<mi>&pi;</mi>
</mrow>
</mfrac>
<mfrac>
<mrow>
<mi>d</mi>
<mi> </mi>
<mi>arccos</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>s</mi>
<mrow>
<mn>1</mn>
<mi>n</mi>
</mrow>
</msub>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mi>d</mi>
<mi>t</mi>
</mrow>
</mfrac>
<mo>;</mo>
</mrow>
Wherein, f1(t) instantaneous frequency of first multiplicative function component, s are represented1n(t) pure FM signal.
7. the method for identification DC Traction Network fault current according to claim 5, it is characterised in that according to new envelope
Estimation function judges whether new current signal is that pure FM signal includes：
One iteration error Δ e is set, if the new envelope estimation function and 1 difference (in-Δ e ,+Δ e), are then judging
New current signal is pure FM signal.
8. a kind of system for recognizing DC Traction Network fault current, it is characterised in that including：Local mean value decomposing module, product
Function component composite module, time-frequency entropy computing module, time-frequency entropy and characteristic quantity comparison module；
The current signal collected is decomposed and obtains multiple multiplicative function components by the local mean value decomposing module；
The multiplicative function component combination module combines the instantaneous amplitude of all multiplicative function components and instantaneous frequency, obtains
To the time-frequency distributions of the current signal；
The time-frequency entropy computing module calculates the time-frequency entropy of the current signal according to the time-frequency distributions；
The characteristic quantity of the time-frequency entropy and short-circuit current signal is compared by the time-frequency entropy and characteristic quantity comparison module,
Recognize the current signal whether be DC Traction Network short-circuit current.
9. the system of identification DC Traction Network fault current according to claim 8, it is characterised in that the multiplicative function
Component combination module combines the instantaneous amplitude of all multiplicative function components and instantaneous frequency, obtains the current signal
Time-frequency distributions include：
<mrow>
<mi>L</mi>
<mrow>
<mo>(</mo>
<mi>f</mi>
<mo>,</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>n</mi>
</munderover>
<msub>
<mi>a</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mi>c</mi>
<mi>o</mi>
<mi>s</mi>
<mo>&lsqb;</mo>
<mn>2</mn>
<mi>&pi;</mi>
<mo>&Integral;</mo>
<msub>
<mi>f</mi>
<mi>i</mi>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mi>d</mi>
<mi>t</mi>
<mo>&rsqb;</mo>
<mo>;</mo>
</mrow>
In formula, L represents the time-frequency distributions of the current signal, and n is the number of multiplicative function component, and f represents the multiplicative function
The instantaneous frequency of component, t represents time, ai(t) instantaneous amplitude of i-th of multiplicative function component, f are representedi(t) represent to multiply for i-th
The instantaneous frequency of Product function component.
10. the system of identification DC Traction Network fault current according to claim 8, it is characterised in that the time-frequency entropy
The time-frequency entropy that computing module calculates the current signal according to the time-frequency distributions includes：
The time-frequency distributions are divided into the time frequency block of N (N=2,3,4 ...) individual area equation, every piece by the time-frequency entropy computing module
The energy of time frequency block is Wi(i=1 ..., N), to WiIt is normalizedWherein W is the energy of whole time-frequency distributions,The time-frequency entropy for obtaining the current signal collected is：
<mrow>
<mi>s</mi>
<mrow>
<mo>(</mo>
<mi>q</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<mo>-</mo>
<munderover>
<mo>&Sigma;</mo>
<mrow>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
<mi>N</mi>
</munderover>
<msub>
<mi>q</mi>
<mi>i</mi>
</msub>
<mi>ln</mi>
<mi> </mi>
<msub>
<mi>q</mi>
<mi>i</mi>
</msub>
<mo>;</mo>
</mrow>
Wherein, the current signal time-frequency entropy collected described in s (q) expressions.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610573047.6A CN107238778A (en) | 2016-07-18 | 2016-07-18 | A kind of method and system for recognizing DC Traction Network fault current |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610573047.6A CN107238778A (en) | 2016-07-18 | 2016-07-18 | A kind of method and system for recognizing DC Traction Network fault current |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107238778A true CN107238778A (en) | 2017-10-10 |
Family
ID=59983443
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610573047.6A Pending CN107238778A (en) | 2016-07-18 | 2016-07-18 | A kind of method and system for recognizing DC Traction Network fault current |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107238778A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108321776A (en) * | 2018-02-06 | 2018-07-24 | 上海交通大学 | UHVDC Transmission Lines guard method based on special frequency channel electric current |
| CN108664936A (en) * | 2018-05-14 | 2018-10-16 | 浙江师范大学 | A kind of diagnostic method and system based on mechanical disorder |
| CN111856324A (en) * | 2020-07-30 | 2020-10-30 | 中国联合网络通信集团有限公司 | Fault detection method and device for traction network feeder line |
| CN112255501A (en) * | 2020-10-12 | 2021-01-22 | 成都交大许继电气有限责任公司 | Method for accurately extracting fault current during traction network fault |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1847867A (en) * | 2006-03-24 | 2006-10-18 | 西南交通大学 | Post-wavelet analysis treating method and device for electric power transient signal |
| CN102866027A (en) * | 2012-08-13 | 2013-01-09 | 燕山大学 | Rotary machinery fault feature extracting method based on local mean decomposition (LMD) and local time-frequency entropy |
| EP2752674A1 (en) * | 2013-01-03 | 2014-07-09 | ABB Technology AG | A detection method of a ground fault in an electric power distribution network |
| CN104897403A (en) * | 2015-06-24 | 2015-09-09 | 北京航空航天大学 | Self-adaption fault diagnosis method based on permutation entropy (PE) and manifold-based dynamic time warping (MDTW) |
-
2016
- 2016-07-18 CN CN201610573047.6A patent/CN107238778A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1847867A (en) * | 2006-03-24 | 2006-10-18 | 西南交通大学 | Post-wavelet analysis treating method and device for electric power transient signal |
| CN102866027A (en) * | 2012-08-13 | 2013-01-09 | 燕山大学 | Rotary machinery fault feature extracting method based on local mean decomposition (LMD) and local time-frequency entropy |
| EP2752674A1 (en) * | 2013-01-03 | 2014-07-09 | ABB Technology AG | A detection method of a ground fault in an electric power distribution network |
| CN104897403A (en) * | 2015-06-24 | 2015-09-09 | 北京航空航天大学 | Self-adaption fault diagnosis method based on permutation entropy (PE) and manifold-based dynamic time warping (MDTW) |
Non-Patent Citations (2)
| Title |
|---|
| 冯璐: "基于LMD能量熵和GK聚类的电能质量扰动识别方法研究", 《中国优秀硕士论文全文数据库》 * |
| 程军圣等: "基于局部均值分解的循环频率和能量谱在齿轮故障诊断中的应用", 《振动工程学报》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108321776A (en) * | 2018-02-06 | 2018-07-24 | 上海交通大学 | UHVDC Transmission Lines guard method based on special frequency channel electric current |
| CN108664936A (en) * | 2018-05-14 | 2018-10-16 | 浙江师范大学 | A kind of diagnostic method and system based on mechanical disorder |
| CN108664936B (en) * | 2018-05-14 | 2020-09-01 | 浙江师范大学 | Diagnosis method and system based on machine fault |
| CN111856324A (en) * | 2020-07-30 | 2020-10-30 | 中国联合网络通信集团有限公司 | Fault detection method and device for traction network feeder line |
| CN112255501A (en) * | 2020-10-12 | 2021-01-22 | 成都交大许继电气有限责任公司 | Method for accurately extracting fault current during traction network fault |
| CN112255501B (en) * | 2020-10-12 | 2023-09-26 | 成都交大许继电气有限责任公司 | Method for accurately extracting fault current during traction network fault |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103308292B (en) | Based on the vacuum breaker mechanical state detection method of analysis of vibration signal | |
| CN107238778A (en) | A kind of method and system for recognizing DC Traction Network fault current | |
| CN102841251B (en) | Electric car charging station harmonic wave detection method in microgrid | |
| Li et al. | Fault detection and classification in medium voltage DC shipboard power systems with wavelets and artificial neural networks | |
| Zand et al. | Fault locating transmission lines with thyristor-controlled series capacitors By fuzzy logic method | |
| CN105116208B (en) | A kind of extra-high voltage DC transmission system commutation failure method for diagnosing faults | |
| CN104836232B (en) | Frequency wide-range self-adaptive repetition control method of active power filter | |
| JP2007202393A (en) | A kind of spectrum for analyzing system circuit stabilization | |
| CN107546769B (en) | Method for obtaining transient stability of grid-connected inverter type distributed power supply | |
| CN106328120A (en) | Public place abnormal sound characteristic extraction method | |
| CN104796031B (en) | For a kind of control method of new control system of track traffic subordinate inverter | |
| CN103872690A (en) | Method for controlling dynamic voltage restorer based on HHT detection method and PFC | |
| Cui et al. | High‐frequency resonance suppression of high‐speed railways in China | |
| CN103560491B (en) | A kind of phase current balanced protection circuit of automobile frequency converter and guard method | |
| CN107036709A (en) | A kind of transformer station's noise matching separation method | |
| CN109088425A (en) | A kind of commutation failure prevention method based on ac bus voltage disturbance amount | |
| CN104236933A (en) | Hidden fault danger warning method for train hauling system | |
| CN102222162A (en) | Prevention control and emergent state aid decision-making method for multiple safety and stability constrains | |
| CN107181460A (en) | A kind of photovoltaic system method for detecting arc | |
| Nunes et al. | Insulation coordination considering the switching overvoltage waveshape—Part I: Methodology | |
| Zhang et al. | VCT-AOC comprehensive method to suppress high-frequency resonance and low-frequency oscillation in railway traction power supply system | |
| CN104465233A (en) | Configuration method with voltage dip character of low-voltage releasing devices taken into consideration | |
| Gupta et al. | Wavelet Based Enhanced Fault Detection Scheme for A Distribution System Embedded with Electric Vehicle Charging Station | |
| Yang et al. | Feature-based solution to harmonics interference on track circuit in electrified heavy haul railway | |
| Naik et al. | Analysis of power quality disturbances using wavelet packet transform |
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: 20171010 |
|
| RJ01 | Rejection of invention patent application after publication |