CN110824344A - High-voltage circuit breaker state evaluation method based on vibration signal short-time energy-entropy ratio and DTW - Google Patents
High-voltage circuit breaker state evaluation method based on vibration signal short-time energy-entropy ratio and DTW Download PDFInfo
- Publication number
- CN110824344A CN110824344A CN201911008524.4A CN201911008524A CN110824344A CN 110824344 A CN110824344 A CN 110824344A CN 201911008524 A CN201911008524 A CN 201911008524A CN 110824344 A CN110824344 A CN 110824344A
- Authority
- CN
- China
- Prior art keywords
- short
- circuit breaker
- dtw
- frame
- time energy
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01R—MEASURING ELECTRIC VARIABLES; MEASURING MAGNETIC VARIABLES
- G01R31/00—Arrangements for testing electric properties; Arrangements for locating electric faults; Arrangements for electrical testing characterised by what is being tested not provided for elsewhere
- G01R31/327—Testing of circuit interrupters, switches or circuit-breakers
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01M—TESTING STATIC OR DYNAMIC BALANCE OF MACHINES OR STRUCTURES; TESTING OF STRUCTURES OR APPARATUS, NOT OTHERWISE PROVIDED FOR
- G01M13/00—Testing of machine parts
Abstract
The invention discloses a high-voltage circuit breaker state evaluation method based on a vibration signal short-time energy-entropy ratio and a DTW (dynamic time warping), and relates to a state evaluation method. At present, the state evaluation of the high-voltage circuit breaker cannot be quick, accurate and intuitive. The invention comprises the following steps: firstly, respectively framing vibration signals under normal working conditions and vibration signals under unknown working conditions of the circuit breaker, and sequentially calculating the short-time energy-entropy ratio of the two groups of vibration signals; then, taking the short-time energy entropy ratio obtained by normal working condition signal calculation as a reference vector, taking the short-time energy entropy ratio obtained by unknown working condition signal calculation as a test vector input DTW, and obtaining an optimal matching path of two input vectors; and finally, judging the working state of the breaker according to the change curve of the matching path. According to the fault diagnosis method using the short-time energy-entropy ratio as the DTW input vector, the operating state of the circuit breaker can be evaluated and judged quickly, accurately and visually only by using a group of vibration signals under normal working conditions as a reference.
Description
Technical Field
The invention relates to a state evaluation method, in particular to a high-voltage circuit breaker state evaluation method based on a vibration signal short-time energy-entropy ratio and a DTW.
Background
As one of important devices in power equipment, a high voltage circuit breaker controls on and off of current in a power system, and reliability thereof affects stability of operation of the entire power system. According to relevant investigation statistics, the fault of the circuit breaker operating mechanism machinery and the auxiliary control circuit thereof accounts for about 60% of all faults in the faults of various high-voltage circuit breakers, and the fault trends rise year by year. Therefore, state monitoring and fault diagnosis research on the circuit breaker operating mechanism has become a research hotspot in recent years. The vibration generated in the opening and closing process of the high-voltage circuit breaker comprises state information of each component of the circuit breaker operating mechanism during working, and the fault characteristics of the circuit breaker operating mechanism can be extracted by carrying out analysis on the vibration signal. On the other hand, the waveform of the opening and closing coil current can reflect the states of the electromagnet core movement, the control loop, the tripping pawl and the auxiliary contact, so that the operating state of the circuit breaker can be evaluated by extracting and analyzing the change characteristics of the opening and closing coil current waveform. Therefore, most researchers have studied circuit breakers as objects of study, such as vibration signals of circuit breaker operating mechanisms and current signals of opening and closing coils.
At present, machine learning algorithms such as artificial neural networks, fuzzy C-means algorithms (FCMs), support vector machines and the like are usually used for evaluating and judging the operating conditions of the circuit breakers, but the algorithms all need a large amount of sample data to perform model training to ensure the accuracy of fault identification. However, due to the restriction of the working characteristics of the circuit breaker and the wide variety of the circuit breaker, the data collected under the actual working condition is limited, and the number of test samples is small. Dynamic Time Warping (DTW) is used as a pattern recognition algorithm widely applied to the field of fault diagnosis and the like, the state evaluation of industrial equipment can be completed only by one group of reference signals and one group of signals to be detected, and a large amount of sample data is not needed. In other documents, it is mentioned that DTW is applied to fault diagnosis of a high-voltage circuit breaker, a dynamic programming idea is introduced to process a vibration signal of the circuit breaker, and an operation state of the circuit breaker is judged by comparing a signal to be detected with a reference signal to match a change of a path. Therefore, in the subsequent research, some scholars apply DTW to research and analysis of current signals of switching coils of circuit breakers, and determine faults of the circuit breakers according to changes of the current signals, but the switching coil current exists only in the early stage of the action of a circuit breaker operating mechanism, and cannot include characteristics in the whole action process of the circuit breaker.
Disclosure of Invention
The technical problem to be solved and the technical task provided by the invention are to perfect and improve the prior technical scheme, and provide a high-voltage circuit breaker state evaluation method based on the vibration signal short-time energy-entropy ratio and the DTW, so as to quickly, accurately and intuitively evaluate and judge the operating state of the circuit breaker. Therefore, the invention adopts the following technical scheme.
The high-voltage circuit breaker state evaluation method based on the vibration signal short-time energy-entropy ratio and the DTW comprises the following steps of:
1) selecting proper frame length wlen, frame shift inc and window function omega (m), respectively aligning reference signal R and to-be-detected signalThe signal T is measured and is subjected to framing to obtain a corresponding characteristic parameter matrix R with the size of mxnfAnd Tf(ii) a Wherein, each column of the matrix is called a "frame" characteristic parameter time sequence, and the dimension m of each "frame" characteristic parameter time sequence is wlen;
2) calculating the short-time energy-entropy ratio of each frame characteristic parameter time sequence and combining the short-time energy-entropy ratios to obtain a short-time energy-entropy ratio sequence EERRAnd EERT；
3) Will EERRAnd EERTInputting DTW as input parameter to obtain optimal matching path W*(ii) a According to the obtained optimal matching path W*And evaluating and diagnosing the operating state of the circuit breaker.
As a preferable technical means: in step 1), the window function ω (m) is
Rectangular window: ω (n) ═ 1 (1)
Or: haining window: ω (n) ═ 0.5- (1-cos (2 π n/(L-1))) (2)
Or: hamming window: ω (n) ═ 0.54-0.46cos (2 π n/(L-1)) (3)
Wherein L is the window length, and n is more than or equal to 0 and less than or equal to L-1.
As a preferable technical means: in step 2), the calculation of the short-time energy entropy comprises the following steps:
201) the method includes the steps of setting a time sequence x (N), wherein N is 1,2, and N, firstly, eliminating a direct current component and normalizing an amplitude value of x (N), and defining an ith frame vibration signal obtained by framing a window function omega (m) as yi(m) then yi(m) satisfies:
where ω (m) is a window function, yi(m) is a frame number, wlen is a frame length, inc is a frame shift length, and fn is a total frame number after signal framing; for yi(m) Fourier transform of the k-th spectral frequency component fkHas an energy spectrum of Yi(k) Then the normalized spectral probability density function p for each frequency componenti(k) Is defined as:
in the formula, pi(k) For the kth frequency component f of the ith framekCorresponding probability density, N is FFT length;
202) spectral entropy of frame i HiComprises the following steps:
203) the energy of the ith frame is:
204) the energy-entropy ratio of the ith frame is:
as a preferable technical means: in step 3), the processing of DTW comprises the steps of:
301) constructing a cost matrix d with the size of m × n, wherein the matrix elements d (i, j) are Euclidean distances of x (i) and y (j), namely:
302) under the preset boundary condition and constraint rule, a path W which can minimize the accumulative cost matrix D among the elements of the matrix D can be found according to the dynamic programming idea*I.e. the optimal regular path;
W＝(w1,w2,...,wk),wk＝(i,j)k(10)
wherein: x and y are time series, m and n are lengths of the time series, and i and j are indexes of the sequence x and y respectively; the boundary conditions are as follows: the regular path starts from the index (1,1) element of the cost matrix and ends at the index (m, n) element, i.e., w1＝(1,1)1,wk＝(m,n)k(ii) a The constraint rules include local continuous constraints: the locally continuous constraint mode is realized by defining a matching point before each position in the regular path.
As a preferable technical means: the expression of the local constraint mode is as follows:
。
as a preferable technical means: the window function ω (m) employs a hamming window.
As a preferable technical means: in the step 1), the frame length is less than 8ms and more than or equal to 3 ms.
As a preferable technical means: the frame length is 6ms, the sampling frequency is set to be 10kHz, and when the frame length is converted into points, the frame length is 60; the frame shift inc is 20.
Has the advantages that: 1) the technical scheme adopts a short-time energy-entropy ratio, which can effectively improve the signal-to-noise ratio of a signal when processing a vibration signal of a circuit breaker, and can enhance the micro-impact characteristics in an original signal, reduce the difference between the micro-impact event and a strong-impact event, and more retain the characteristic information contained in the original signal, wherein the characteristic information contained in a characteristic sequence is richer than the short-time energy;
2) the short-time energy entropy ratio effectively extracts the characteristics contained in the original signal, removes redundant components in the original signal, reduces the complexity of input characteristics, replaces the original signal with the short-time energy entropy ratio as an input parameter of the DTW algorithm, improves the matching performance of the DTW algorithm, reduces the performance requirement on a computer, and accelerates the calculation speed;
3) according to the curve change trend of the regular path calculated by the DTW algorithm, although the operation condition of the circuit breaker can be intuitively judged, the fault type and the fault degree cannot be directly and accurately judged, and the regular path change curve can be further explored and researched subsequently by combining the action condition of the circuit breaker operating mechanism parts in the closing process;
4) due to the complexity of the structure of the circuit breaker, various faults can occur in the operation process of the circuit breaker, and the occurrence of some faults can influence the mechanical characteristics of the circuit breaker and can be reflected on a vibration signal; some faults may be reflected very weakly on the vibration signal and are difficult to distinguish through the vibration signal, and the proposed method may not distinguish the faults which are weakly related to the mechanical characteristics of the circuit breaker;
5) according to the technical scheme, the vibration signals under a group of normal working conditions are only needed to be used as a reference, so that the state evaluation and fault diagnosis can be accurately, quickly and visually carried out on the operating conditions of the circuit breaker, and the method can be applied to the on-line monitoring of the high-voltage circuit breaker.
Drawings
FIG. 1 is a flow chart of the present invention.
Figure 2 is a DTW schematic of the present invention.
Fig. 3(a) and 3(b) are views of DTW local constraint patterns according to the present invention.
Fig. 4(a) and 4(b) are the corresponding relationship of two time series under the DTW algorithm and the regular path diagram thereof.
FIG. 5 is a diagram of the cumulative cost matrix of the present invention.
Fig. 6(a), 6(b), 6(c), and 6(d) are graphs of vibration signals of the breaker beam in a normal state, a fatigue state of a closing spring, a loose state of a base bolt, and an abnormal state of a control circuit voltage, respectively.
Fig. 7(a) and 7(b) show the short-term entropy ratios of the lower signal when the frame lengths are 3ms and 6ms, respectively.
Fig. 8 is a short-time analysis profile of the vibration signal of the present invention.
Fig. 9(a), 9(b), 9(c), and 9(d) are DTW rule path diagrams in a normal state, a fatigue state of a closing spring, a loosening state of a base bolt, and an abnormal state of a control circuit voltage, respectively.
Fig. 10 is a normalized path diagram under a fatigue failure condition of a closing spring.
Fig. 11(a), 11(b), and 11(c) are respectively a graph of the original signal DTW, the short-time energy DTW, and the short-time energy-entropy ratio DTW.
Detailed Description
The technical scheme of the invention is further explained in detail by combining the drawings in the specification.
First, as shown in fig. 1, the present invention comprises the steps of:
1) selecting a proper frame length wlen, a frame shift inc and a window function omega (m), respectively framing the reference signal R and the signal T to be detected, and obtaining a corresponding characteristic parameter matrix R with the size of m multiplied by nfAnd Tf(ii) a Wherein, each column of the matrix is called a "frame" characteristic parameter time sequence, and the dimension m of each "frame" characteristic parameter time sequence is wlen;
2) calculating the short-time energy-entropy ratio of each frame characteristic parameter time sequence and combining the short-time energy-entropy ratios to obtain a short-time energy-entropy ratio sequence EERRAnd EERT；
3) Will EERRAnd EERTInputting DTW as input parameter to obtain optimal matching path W*(ii) a According to the obtained optimal matching path W*And evaluating and diagnosing the operating state of the circuit breaker.
Secondly, some contents are explained in detail below
2.1 short-term energy-to-entropy ratio
In the short-time analysis of signals, firstly, a proper window function omega (n) needs to be selected, the frame length and the frame shift are used for framing the analyzed signals to obtain a series of characteristic parameter time sequences taking frames as units, and then the characteristic parameter time sequences are subjected to analysis processing on a time domain or a frequency domain. Some commonly used window functions are shown in equations (1) to (3).
Rectangular window: ω (n) ═ 1 (1)
Haining window: ω (n) ═ 0.5- (1-cos (2 π n/(L-1))) (2)
Hamming window: ω (n) ═ 0.54-0.46cos (2 π n/(L-1)) (3)
Wherein L is the window length, and n is more than or equal to 0 and less than or equal to L-1.
The short-time energy-to-entropy ratio is a time domain analysis algorithm, can obviously improve the signal-to-noise ratio of a signal, effectively enhances the characteristics of a micro impact component in a vibration signal, and highlights a vibration event, and the calculation process is as follows:
1) the method includes the steps of setting a time sequence x (N), wherein N is 1,2, and N, firstly, eliminating a direct current component and normalizing an amplitude value of x (N), and defining an ith frame vibration signal obtained by framing a window function omega (m) as yi(m) then yi(m) satisfies:
where ω (m) is a window function, yi(m) is the number of a frame, wlen is the frame length, inc is the frame shift length, and fn is the total number of frames after the signal is framed. For yi(m) Fourier transform of the k-th spectral frequency component fkHas an energy spectrum of Yi(k) Then the normalized spectral probability density function p for each frequency componenti(k) Is defined as:
in the formula, pi(k) For the kth frequency component f of the ith framekCorresponding probability density, N is the FFT length.
2) Spectral entropy of frame i HiComprises the following steps:
3) the energy of the ith frame is:
4) the energy-entropy ratio of the ith frame is:
2.2 dynamic time warping
As shown in fig. 2, DTW is a flexible pattern matching algorithm on the idea of establishing dynamic programming, and the goal is to construct a cumulative cost matrix of two input sequences, perform global or local expansion, compression or deformation on the two input sequences under a certain constraint condition, and find out an optimal matching path of the two input sequences. Inputting the characteristics of the vibration signal under the normal working condition of the circuit breaker into DTW as a reference vector, inputting the characteristics of the vibration signal under the current unknown working condition into DTW as a test vector, and if the obtained regular path shows a straight line of 45 degrees, indicating that the working state of the circuit breaker under the current working condition is good; if the obtained regular path and the 45-degree straight line find obvious deviation, the working state of the circuit breaker under the current working condition is changed, and a fault exists. The DTW basic process is as follows.
As shown in fig. 2, time series x and y are provided, with lengths m and n, respectively, defining indices for i and j, respectively, for the series x and y.
First, a cost matrix d with the size of m × n is constructed, and matrix elements d (i, j) are euclidean distances of x (i) and y (j), that is:
then, under the preset boundary condition and constraint rule, a path W which minimizes the accumulative cost matrix D among the elements of the matrix D can be found out according to the dynamic planning idea*I.e. the optimal regular path.
W＝(w1,w2,...,wk),wk＝(i,j)k(10)
According to the characteristics of the circuit breaker vibration signal, the technical scheme defines the following boundary conditions and constraint rules.
1) Boundary conditions: the generated vibration signal of the circuit breaker operating mechanism is an impact signal formed by overlapping a plurality of impact events in sequence, and has the advantages of time sequence and continuity. Thus, defining a canonical path starts with the index (1,1) element of the cost matrix and ends with the index (m, n) element, i.e., w1＝(1,1)1,wk＝(m,n)k。
2) Local continuous constraint: as shown in fig. 3(a) and 3(b), in the local path, in order to limit the excessive compression or expansion of the regular path, there are two typical constraint manners, which are realized by defining the matching point before each position in the regular path.
Because the collected circuit breaker vibration signal contains a large amount of interference and noise, abrupt points or abrupt paths are inevitable to exist in the regular path when the circuit breaker vibration signal is subjected to DTW (digital time warping) regulation. When the local constraint mode of fig. 3(b) is adopted for path regularization, the mutation points can be effectively avoided, so that the influence of the mutation points on the path regularization is reduced.
Fig. 4(a) and 4(b) show the correspondence between two time sequences under the DTW algorithm and their warping paths, and it can be seen that when DTW warping is performed in the "27-45-63" local constraint manner, the warping paths effectively skip the mutation points or mutation paths, so the present technical solution adopts this local constraint. At this time, the recursive relationship and the recursive relationship of the cumulative cost matrix D (i, j) are shown in fig. 5 and equation (11).
Finally, the minimum accumulated cost D of the two time sequences x and y can be obtained*D (m, n) and its optimal regular path W*。
Third, experimental verification
3.1 Experimental model
In order to verify the effectiveness of the proposed method, the technical scheme takes an LW42A-40.5 outdoor SF6 high-voltage circuit breaker produced by yunnan yun open electrical products limited company as a research object, and in terms of signal acquisition and storage, in the embodiment, a DH5922N dynamic signal test analysis system and A1 a102E type acceleration sensor produced by Jiangsu Donghua test technologies limited company are selected. The range of the acceleration sensor is 0-5000 m.s-2The frequency range is 1-10000 Hz. Considering the vibration source position, the sensor is installed on a beam of the circuit breaker in a magnetic type. In the whole closing process of the circuit breaker, the action time of the operating mechanism is about 80 +/-15 ms. Therefore, in order to ensureAnd the vibration signals are fully collected, the signal collection time is set to be 400ms, and the sampling frequency is set to be 10 kHz. 3.2 feature extraction
Under the no-load state of the circuit breaker, the technical scheme collects vibration signals of the circuit breaker under normal working conditions and three simulated fault working conditions of fatigue of a closing spring, looseness of a base bolt and abnormal voltage of a control loop. The fatigue fault of the closing spring is realized by adjusting the fastening bolt of the spring to reduce the pretightening force of the fastening bolt, the bolt loosening fault of the base is realized by loosening the bolt of the base, and the abnormal fault of the voltage of the control loop is realized by adjusting the voltage of the control loop of the circuit breaker by the high-voltage switch tester.
As shown in Table 1, according to the technical scheme, two groups of data of the circuit breaker in each running state are selected as experimental samples, and the numbers of the sample data are sequentially 1-8. As shown in fig. 6, a part of sample data shows that a certain offset occurs at the occurrence time of the strongest impact event in the closing process of the circuit breaker under each operating condition.
TABLE 1 sample data
Taking sample data 1 as an example, the short-time energy-entropy ratio of the signal is calculated. When calculating the short-time energy entropy ratio, firstly selecting a proper window function omega (m), a frame length wlen and a frame shift inc to frame the signal, and then calculating the short-time energy entropy ratio of the framed signal. From the perspective of reducing leakage, because the vibration signals of the circuit breaker are distributed on a plurality of frequency bands, and the hamming window can eliminate high-frequency interference relative to the rectangular window and reduce leakage energy, the hamming window is selected as the window function in the technical scheme, as shown in formula (3).
On the other hand, the frame length is a determining factor that determines the temporal entropy ratio to the resolution. When the sampling frequency of the signal is fixed, the smaller the frame length value is, the higher the time resolution of the short-time entropy ratio is, but the too short frame length is not beneficial to exerting the advantage of improving the signal-to-noise ratio of the signal by the short-time entropy ratio. Therefore, the frame length is selected by taking the sampling frequency of the signal, the time resolution requirement, the signal-to-noise ratio requirement and other conditions into comprehensive consideration to select a proper value. Because the noise interference inevitably exists in the acquisition process of the vibration signal, the frame length is not suitable to be set to be short. By observing the collected vibration signals, the vibration signals in 8ms are relatively stable, so the frame length value is less than 8 ms. As shown in fig. 7(a) and 7(b), the frame lengths of 3ms and 6ms are respectively selected to calculate the short-time energy-entropy ratio, and it can be seen that when the frame length is small, the smoothing effect of the short-time analysis processing method is not obvious, and when the frame length is large, the smoothing effect is good, which is beneficial to the DTW algorithm to perform path normalization. In summary, in the present technical solution, the frame length is selected to be 6ms, and the frame length is converted into points, that is, the frame length is 60. The frame shift is generally 25% -50% of the frame length, and the technical scheme is that the frame shift inc is 20.
As shown in fig. 8, the short-term energy-to-short-term entropy ratio of the sample data 1 is obtained through calculation, and it can be seen that the short-term energy feature sequence loses the characteristics of the tiny impact events contained in the original signal, and the short-term energy entropy feature sequence effectively extracts the characteristics of the events with small impact contained in the vibration signal, so that the local features of the signal are enhanced.
3.3 DTW matching
Short-time entropy bit characteristics obtained by calculating sample data 1 (circuit breaker vibration signals collected in a normal state) are used as reference vectors of the DTW, other sample data characteristics are used as test vectors and input into the DTW, and optimal regular paths obtained in all states are shown in FIG. 9.
It can be seen that the matching path (dotted line) obtained by using the normal signal as the test vector is basically consistent with the reference line (solid line), and the matching path obtained by using the fault signal as the test vector has obvious deviation with the reference line, so that the existence of the fault and the change of the current operating state of the circuit breaker are proved. Comparing the optimal regular paths obtained in the same fault state, it is found that the change trends of the matched path curves obtained in the same fault state are the same, as shown in fig. 10. Therefore, in the subsequent research, the fault type and the fault degree of the circuit breaker can be judged according to the change curve of the DTW matching path.
In order to prove the superiority of the proposed method, the technical solution further inputs the original signal and the original short-time energy signature sequence into the DTW, and obtains a DTW regular path, as shown in fig. 11(a), 11(b), and 11 (c). It can be seen that the regular path deviation degree obtained by using the short-time entropy bit characteristic sequence as the DTW input parameter is most obvious, and the change of the operation condition of the circuit breaker can be most obviously indicated. Compared with the original signal feature sequence, the method has the advantages that the short-time energy-entropy ratio extracts feature information contained in each frame of original signal, removes redundant information contained in the original signal and reduces complexity of input features; compared with the short-time energy characteristic sequence, the short-time energy entropy ratio effectively extracts the characteristics of the micro impact events in the original signal, enhances the local characteristics of the original signal, and the sequence contains more abundant characteristic information. From the view of operation efficiency, due to the characteristics of the DTW algorithm, when the algorithm is used for matching path optimization, a multidimensional array with the size corresponding to the dimension of an input vector can be found in a computer memory, the short-time energy entropy ratio of the algorithm to the DTW input vector can undoubtedly reduce the memory occupancy rate, improve the operation efficiency and further accelerate the operation speed of the computer.
The method for evaluating the state of the high-voltage circuit breaker based on the short-time energy-entropy ratio and the DTW shown in fig. 1 is a specific embodiment of the present invention, has embodied the substantial features and the progress of the present invention, and can be modified equivalently in terms of shape, structure and the like according to the practical use requirements and under the teaching of the present invention, and is within the protection scope of the present solution.
Claims (8)
1. The method for evaluating the state of the high-voltage circuit breaker based on the short-time energy-entropy ratio and DTW of the vibration signal is characterized by comprising the following steps of:
1) selecting a proper frame length wlen, a frame shift inc and a window function omega (m), respectively framing the reference signal R and the signal T to be detected, and obtaining a corresponding characteristic parameter matrix R with the size of m multiplied by nfAnd Tf(ii) a Wherein each column of the matrix is called a 'frame' characteristic parameter time sequenceThe dimension m of each "frame" feature parameter time series is wlen;
2) calculating the short-time energy-entropy ratio of each frame characteristic parameter time sequence and combining the short-time energy-entropy ratios to obtain a short-time energy-entropy ratio sequence EERRAnd EERT；
3) Will EERRAnd EERTInputting DTW as input parameter to obtain optimal matching path W*(ii) a According to the obtained optimal matching path W*And evaluating and diagnosing the operating state of the circuit breaker.
2. The method for evaluating the state of a high-voltage circuit breaker based on the short-time energy-entropy ratio and DTW of a vibration signal as claimed in claim 1, wherein: in step 1), the window function ω (m) is
Rectangular window: ω (n) ═ 1 (1)
Or: haining window: ω (n) ═ 0.5- (1-cos (2 π n/(L-1))) (2)
Or: hamming window: ω (n) ═ 0.54-0.46cos (2 π n/(L-1)) (3)
Wherein L is the window length, and n is more than or equal to 0 and less than or equal to L-1.
3. The method for evaluating the state of a high-voltage circuit breaker based on the short-time energy-entropy ratio and DTW of a vibration signal as claimed in claim 2, wherein: in step 2), the calculation of the short-time energy entropy comprises the following steps:
201) the method includes the steps of setting a time sequence x (N), wherein N is 1,2, and N, firstly, eliminating a direct current component and normalizing an amplitude value of x (N), and defining an ith frame vibration signal obtained by framing a window function omega (m) as yi(m) then yi(m) satisfies:
where ω (m) is a window function, yi(m) is a frame number, wlen is a frame length, inc is a frame shift length, and fn is a total frame number after signal framing; for yi(m) Fourier transform of the k-th spectral frequency component fkHas an energy spectrum of Yi(k)Then the normalized spectral probability density function p for each frequency componenti(k) Is defined as:
in the formula, pi(k) For the kth frequency component f of the ith framekCorresponding probability density, N is FFT length;
202) spectral entropy of frame i HiComprises the following steps:
203) the energy of the ith frame is:
204) the energy-entropy ratio of the ith frame is:
4. the method for evaluating the state of a high-voltage circuit breaker based on the short-time energy-entropy ratio and DTW of a vibration signal as claimed in claim 3, wherein: in step 3), the processing of DTW comprises the steps of:
301) constructing a cost matrix d with the size of m × n, wherein the matrix elements d (i, j) are Euclidean distances of x (i) and y (j), namely:
302) under the preset boundary condition and constraint rule, a path W which can minimize the accumulative cost matrix D among the elements of the matrix D can be found according to the dynamic programming idea*I.e. the optimal regular path;
W＝(w1,w2,...,wk),wk＝(i,j)k(10)
wherein: x and y are time series, m and n are lengths of the time series, and i and j are indexes of the sequence x and y respectively; the boundary conditions are as follows: the regular path starts from the index (1,1) element of the cost matrix and ends at the index (m, n) element, i.e., w1＝(1,1)1,wk＝(m,n)k(ii) a The constraint rules include local continuous constraints: the locally continuous constraint mode is realized by defining a matching point before each position in the regular path.
5. The method for evaluating the state of a high-voltage circuit breaker based on the vibration signal short-time energy-entropy ratio and DTW according to claim 4, wherein: the expression of the local constraint mode is as follows:
6. the method for evaluating the state of a high-voltage circuit breaker based on the short-time energy-entropy ratio and DTW of a vibration signal as claimed in claim 2, wherein: the window function ω (m) employs a hamming window.
7. The method for evaluating the state of a high-voltage circuit breaker based on the short-time energy-entropy ratio and DTW of a vibration signal as claimed in claim 1, wherein: in the step 1), the frame length is less than 8ms and more than or equal to 3 ms.
8. The method for evaluating the state of a high-voltage circuit breaker based on the short-time energy-entropy ratio and DTW of a vibration signal as claimed in claim 7, wherein: the frame length is 6ms, the sampling frequency is set to be 10kHz, and when the frame length is converted into points, the frame length is 60; the frame shift inc is 20.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911008524.4A CN110824344A (en) | 2019-10-22 | 2019-10-22 | High-voltage circuit breaker state evaluation method based on vibration signal short-time energy-entropy ratio and DTW |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911008524.4A CN110824344A (en) | 2019-10-22 | 2019-10-22 | High-voltage circuit breaker state evaluation method based on vibration signal short-time energy-entropy ratio and DTW |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110824344A true CN110824344A (en) | 2020-02-21 |
Family
ID=69549990
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911008524.4A Pending CN110824344A (en) | 2019-10-22 | 2019-10-22 | High-voltage circuit breaker state evaluation method based on vibration signal short-time energy-entropy ratio and DTW |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110824344A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114544205A (en) * | 2022-01-05 | 2022-05-27 | 北京交通大学 | Automatic fault diagnosis system for locomotive wheel drive running-in test |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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) |
| CN107817098A (en) * | 2017-10-12 | 2018-03-20 | 东北电力大学 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method |
| CN109164381A (en) * | 2018-09-07 | 2019-01-08 | 广西电网有限责任公司电力科学研究院 | A kind of mechanical state of high-voltage circuit breaker on-line monitoring and fault identification method and device |
| CN109270441A (en) * | 2018-07-17 | 2019-01-25 | 国网浙江省电力有限公司检修分公司 | High-voltage circuitbreaker separating brake characterisitic parameter online test method based on vibration signal |
-
2019
- 2019-10-22 CN CN201911008524.4A patent/CN110824344A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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) |
| CN107817098A (en) * | 2017-10-12 | 2018-03-20 | 东北电力大学 | A kind of Mechanical Failure of HV Circuit Breaker diagnostic method |
| CN109270441A (en) * | 2018-07-17 | 2019-01-25 | 国网浙江省电力有限公司检修分公司 | High-voltage circuitbreaker separating brake characterisitic parameter online test method based on vibration signal |
| CN109164381A (en) * | 2018-09-07 | 2019-01-08 | 广西电网有限责任公司电力科学研究院 | A kind of mechanical state of high-voltage circuit breaker on-line monitoring and fault identification method and device |
Non-Patent Citations (1)
| Title |
|---|
| 王振浩 等: "动态时间规整算法诊断高压断路器故障", 《高电压技术》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114544205A (en) * | 2022-01-05 | 2022-05-27 | 北京交通大学 | Automatic fault diagnosis system for locomotive wheel drive running-in test |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Liu et al. | Classifying transformer winding deformation fault types and degrees using FRA based on support vector machine | |
| CN109270441B (en) | High-voltage circuit breaker opening characteristic parameter online detection method based on vibration signal | |
| CN106199412B (en) | A kind of permanent magnet mechanism high-pressure vacuum breaker method of fault pattern recognition | |
| Goumas et al. | Classification of washing machines vibration signals using discrete wavelet analysis for feature extraction | |
| CN109460618A (en) | A kind of rolling bearing remaining life on-line prediction method and system | |
| Wang et al. | Classification of power quality events using optimal time-frequency representations-part 1: Theory | |
| Yang et al. | Fault identification for circuit breakers based on vibration measurements | |
| Wang et al. | Few-shot transfer learning with attention mechanism for high-voltage circuit breaker fault diagnosis | |
| CN108489717B (en) | On-load tap changers of transformers machine performance monitoring method and system | |
| CN112557966A (en) | Transformer winding looseness identification method based on local mean decomposition and support vector machine | |
| Zhu et al. | Adaptive fault diagnosis of HVCBs based on P-SVDD and P-KFCM | |
| CN108919104B (en) | Breaker fault diagnosis method based on Fisher discriminant classification method | |
| CN113125095A (en) | Universal circuit breaker contact system residual mechanical life prediction method based on deep learning | |
| CN111639586A (en) | Non-invasive load identification model construction method, load identification method and system | |
| CN113537328A (en) | Rotary machine fault diagnosis method and device based on deep learning | |
| CN112599134A (en) | Transformer sound event detection method based on voiceprint recognition | |
| CN110263944A (en) | A kind of multivariable failure prediction method and device | |
| Hucker et al. | Requirements of automated PD diagnosis systems for fault identification in noisy conditions | |
| CN115098962A (en) | Method for predicting residual life of mechanical equipment in degradation state based on hidden half Markov model | |
| CN105954695A (en) | Synchronization-based homogeneous-sensor mutation parameter recognizing method and device | |
| CN117076955A (en) | Fault detection method and system for high-voltage frequency converter | |
| CN110824344A (en) | High-voltage circuit breaker state evaluation method based on vibration signal short-time energy-entropy ratio and DTW | |
| CN112636328B (en) | Medium-voltage distribution network equipment load state identification method | |
| CN113010985A (en) | Non-invasive load identification method based on parallel AANN | |
| US20210116504A1 (en) | Method for monitoring circuit breaker and apparatus and internet of things using the same |
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: 20200221 |
|
| RJ01 | Rejection of invention patent application after publication |