CN1232834C - Online detection method for vacuum circuit breaker contact on-off time based on vibration analysis - Google Patents

Abstract

The invention discloses an on-line detection method for the closing moment of a vacuum circuit breaker based on vibration analysis. length and trigger mode and store the sampling data; 3) preprocess the vibration data; 4) select the type and length of the moving window function, and use the short-term energy algorithm for analysis; 5) select the threshold value to determine the three-phase contact 6) Calculate the closing synchronism and closing time according to the closing time of the three-phase contacts. The invention improves the event resolution and anti-interference ability of vibration signal analysis by selecting the installation position of the vibration sensor and the type and length of the short-term energy analysis time window function, thereby realizing the timing of the closing time of the three-phase contacts in the online state. detection.

Description

On-line Detection Method of Contact Closing Moment of Vacuum Circuit Breaker Based on Vibration Analysis

technical field

The invention relates to the field of electrotechnical technology, in particular to an online detection method for the closing moment of contacts of a vacuum circuit breaker based on vibration analysis.

Background technique

The closing moment of the three-phase contacts of a vacuum circuit breaker is an important parameter to describe its mechanical operation state. The calculation of many mechanical characteristic parameters such as closing time, closing speed and closing synchronism all depend on the determination of the contact closing moment. It is relatively easy to detect this point off-line, and it is generally realized by adding an auxiliary electrical circuit to the three-phase fracture. detect accurately. However, when the circuit breaker is working under load, the three-phase contacts are at high potential, so it cannot be detected by traditional offline electrical methods. In the previous on-line monitoring of circuit breakers, the stroke characteristics of the moving contacts were used to roughly estimate the closing time of the contacts, and the error was large, which was very unfavorable for improving the calculation accuracy of the entire mechanical parameters.

Vibration analysis is a non-invasive detection method. Because the vibration sensor is installed on the grounding part of the circuit breaker, it is very suitable for online condition monitoring and has been widely used in recent years. During the opening and closing operation of the circuit breaker, due to the mutual friction and collision between the internal parts, a time series consisting of multiple shock attenuation waves will be generated on the surface of the mechanism. By extracting the starting points of these evanescent waves, the occurrence moments of different vibration events can be determined. Attached Figure 2 is the vibration signal collected on the surface of the mechanism during the closing process of the ZN12 vacuum circuit breaker, which contains 6 typical vibration events: Event 1 is caused by the closing electromagnet hitting the closing engine; Event 2 is It is caused by the collision of the cam and the lever mechanism; events 3, 4, and 5 are caused by the collision of the three-phase contacts during the closing process; event 6 is caused by the collision when the opening engine is in place. Existing methods to determine the onset time of vibration events include median filter method, time domain envelope method, amplitude method and autocorrelation method. Due to the limitations of event time resolution and anti-interference ability, these methods are limited to the analysis of single vibration events that can be clearly distinguished in the time domain, such as event 1, event 2 and event 6 in Figure 2. The three-phase contacts of vacuum circuit breakers usually use the same operating mechanism. During the closing process, the vibration events caused by the collision of the three-phase contacts are very close in time, and the difference between each other is generally only a few milliseconds (that is, the closing Synchrony). The vibration waves thus generated are superimposed on each other in the time domain when propagating to the mechanism, and have comparable amplitudes (event 3, event 4, and event 5 in Figure 2). For this kind of vibration signal mixed by multiple events, the traditional analysis method can only get one initial moment, but it is difficult to know which phase of contact vibration this moment represents. When the closing synchronism is in the normal state (that is, the maximum time difference of the three-phase closing time is within the allowable range of the normal operation of the circuit breaker), it is possible to use this time to approximately replace the closing time of the circuit breaker; but when the closing synchronous When the characteristics of the circuit breaker change greatly due to the change of the mechanism state, it is necessary to know the closing time of the three-phase contacts separately, so as to correctly evaluate the mechanical operation state of the circuit breaker. In the domestic and foreign references, there is no relevant report on the online measurement of the three-phase contact closing time of the vacuum circuit breaker.

Contents of the invention

The purpose of the present invention is to provide an on-line detection method of the contact closing moment of the vacuum circuit breaker based on vibration analysis, so as to realize the detection of the three-phase contact closing moment of the vacuum circuit breaker in the online state.

The technical idea of the present invention is to select a suitable installation position of the vibration sensor, highlight the vibration signal of the phase contact, and reduce the influence of the vibration of the other two phase contacts. Limiting filter preprocessing is performed on the collected vibration signal, and then the closing moment of each phase contact is extracted by using the algorithm of short-term energy.

The status detection of the circuit breaker is an important content of electrical intelligence. In order to improve the service life of vacuum circuit breakers and reduce their daily maintenance costs, it is necessary to continuously improve the content of online detection of mechanical characteristics. The correctness of state detection depends on the measurement accuracy of various mechanical parameters, among which the closing time of the three-phase contacts is very important for the calculation of closing parameters. From the above prior art of circuit breaker vibration detection, finding a vibration detection method with high event time resolution and strong anti-interference ability is the key to realize the online detection of the three-phase contact closing time.

The purpose of the present invention is achieved by the following technical solutions: The online detection method of the contact closing moment of the vacuum circuit breaker based on vibration analysis includes the following steps:

1) Select the installation location and installation direction of the vibration sensor. The installation location of the sensor is point P at the end where the A, B, and C three-phase insulating rods are connected to the grounding part of the circuit breaker. Three vibration acceleration sensors of the same type are used to simultaneously acquire three The vibration signal of the phase contact, the installation direction is parallel to the insulating rod;

2) Set the sampling sequence, sampling frequency, sampling length and trigger mode and store the sampling data;

3) For the preprocessing of the vibration data, a limiting filter algorithm is used to smooth the mutation points caused by noise in the vibration signal;

4) Select the type and length of the moving window function, and use the short-term energy algorithm for analysis;

5) Select the threshold value to determine the closing moment of the three-phase contacts;

6) Calculate the closing synchronism and closing time according to the closing time of the three-phase contacts.

The sampling sequence of the above settings is A phase-B phase-C phase; the sampling frequency is ≥25kHz; the sampling starts with the rising edge of the closing coil current, and the sampling length is twice the closing time of the circuit breaker.

The limiting filtering algorithm is as follows: first obtain the envelope of the absolute value of the vibration signal, by comparing the envelope amplitudes y(n) and y(n-1) of two sampling points at adjacent moments, if they If the difference exceeds the set limit value, it is considered that mutation interference has occurred, and the sampling value x(n) is regarded as an illegal value, and it is eliminated, and x(n-1) is used instead of x(n ); the limiting value is determined by the product of the maximum change rate V _max of the envelope at the beginning of the vibration signal and the sampling period T.

Select the type and length of the moving window function, and use the short-term energy algorithm to analyze as follows:

Suppose the vibration sequence is x(i), i=0,...,N-1, then the short-term energy function S(n) is defined as:

$\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; \&infin;</span>}{\overset{<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&infin;</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>$

In the formula, w(n), n=0, . . . , M-1 is a moving window function, and M is the length of the window function. S(n) represents the local energy of the signal at time n. When analyzing the closing vibration signal of the vacuum circuit breaker, the Hamming window is selected. When the sampling frequency is determined, the window length M is 50-100. The Hamming window function is defined as:

$w\left(\mathrm{no}\right)=0.54-0.46\cos \left(\frac{2\mathrm{\&pi;n}}{m-1}\right)$ 0≤n≤M-1

The selection range of the threshold value is: 0 to half of the first peak value (E _s ) in the short-term energy analysis waveform of the vibration signal. Thus, the starting time of the three-phase contact vibration event is obtained, which corresponds to the closing time of each phase contact of the circuit breaker; if the maximum value of the short-term energy analysis of the external noise before the vibration event is N _max , then the threshold value T _h Choose as:

${\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}& \mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}\\ \mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; <figure-callout\; id="10"\; label="s"\; filenames="C20041002626200161.png"\; state="\{\{state\}\}">s</figure-callout></span>}}& \mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}\end{array}\right.$

The calculation of closing synchronism and closing time is as follows: judge the first closing phase and the last closing phase according to the order of closing time, and the time difference between them is the closing synchronism; the starting point of the coil current is used as the closing point The moment the command arrives, the time difference between it and the closing moment of the last closing phase is the closing time of the circuit breaker.

The present invention applies the idea of short-term energy to the analysis of the closing vibration signal of the vacuum circuit breaker, and improves the event time resolution ability and Anti-interference ability, so as to realize the detection of the closing moment of the three-phase contacts in the online state.

Description of drawings

Fig. 1 is a software processing flowchart of the present invention;

Figure 2 is a vibration signal diagram when the vacuum circuit breaker is closed;

Figure 3 is a front view of the vacuum circuit breaker;

Fig. 4 is the schematic diagram of the installation position P point of vibration sensor;

Fig. 5 is the impact diagram of the window function length M on the short-term energy analysis result;

Figure 6 is a diagram of the closing transposition signal and closing vibration signal of the three-phase contact of the vacuum circuit breaker;

FIG. 7 is a result diagram of short-term energy analysis on the closing vibration signal of FIG. 6 .

Fig. 8 is the comparison graph of the synchronism of switching on and real value that the vibration signal analysis of the present invention obtains;

Detailed ways

The present invention will be described in more detail below in conjunction with the accompanying drawings.

The flow chart of the vibration signal processing software is shown in Figure 1:

1) Set the sampling sequence, sampling frequency, sampling length and trigger mode;

2) Collect vibration signals and store them;

3) Obtaining the envelope curve for the absolute value of the vibration signal;

4) Find the position of the noise mutation point through the limiting filter algorithm and smooth it in the original signal;

5) Select the type and length of the window function;

6) Use short-term energy algorithm for analysis;

7) Select an appropriate threshold value to determine the closing time of the vacuum circuit breaker;

8) Calculate mechanical state parameters such as closing synchronism, closing time and closing speed.

In order to obtain the closing moments of the three-phase contacts separately, the installation position of the vibration sensor must be reasonably selected first. Refer to the front view of the vacuum circuit breaker in Figure 3 and the schematic diagram of the installation position P of the vibration sensor in Figure 4. For the four-bar linkage mechanism commonly used in vacuum circuit breakers at present, three vibration acceleration sensors of the same type (integrated charge amplifiers inside) are used simultaneously. Detect vibration signals of three-phase contacts. The present invention selects the end connected with the insulating pull rod and the grounding part of the circuit breaker as the measurement position (shown at point P in Figure 4), and the installation direction is parallel to the insulating pull rod. In order not to damage the insulating performance of the insulating rod itself, the installation method adopts the method of epoxy resin paste, so as to obtain the vibration signal of the contact of each phase. Three factors are considered when selecting P point: first, this point is in the grounding part of the circuit breaker, which meets the requirements of on-line detection; The peripheral free space at this point is large, which is convenient for the installation of the vibration sensor. Third, this point is closest to the vibration source of the measuring phase contact, but far from the vibration sources of the other two phase contacts, so the vibration signal of the measuring phase contact can be highlighted to the greatest extent. A large number of comparative tests show that after the buffering and attenuation of the operating mechanism, the frequency and amplitude of the contact vibration signals of the other two phases are greatly reduced at this point, so it is very beneficial to extract the contact vibration events of the measurement phase.

The signal obtained from the vibration sensor is transformed by the sensor signal conditioner and collected by the data acquisition card into the computer for processing. The acquisition of vibration data is triggered by an external interrupt, and the interrupt signal is generated by the closing coil current. From the rising edge of the closing current, the contact vibration signals of phase A, phase B and phase C are sequentially collected. Spectrum analysis shows that the main frequency components of the vacuum circuit breaker closing vibration signal are concentrated within 10kHz. According to the sampling theorem and considering a certain margin, the sampling frequency can be set to 25kHz. The sampling length is twice the closing time of the circuit breaker.

In the process of measurement, transmission and recording of test data, sometimes there will be sudden severe external interference, which will introduce some abnormal mutation points into the recorded signal. When these mutation points are concentrated, it is equivalent to increasing the overall noise level, thereby interfering with the analysis of the data, so it is necessary to preprocess the collected raw data. The algorithm of limiting filter is used in the software to find out the position of the mutation point, and then smooth it in the original data. The algorithm of limiting filtering is as follows: By comparing the envelope amplitudes y(n) and y(n-1) of two adjacent sampling points, if their difference exceeds the set limiting value, it is considered Mutation interference occurs, and the sampling value x(n) is regarded as an illegal value, and is eliminated, and x(n-1) is used instead of x(n). The limiting value is determined by the product of the maximum rate of change V _max of the envelope at the start of the vibration signal and the sampling period T.

The preprocessed signal is analyzed using the algorithm of short-term energy, assuming that the vibration sequence is x(i), i=0,...,N-1, then the short-time energy function S(n) is defined as:

$\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; \&infin;</span>}{\overset{<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&infin;</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>$

In the formula, w(n), n=0, . . . , M-1 is a moving window function, and M is the length of the window function. S(n) represents the local energy of the signal at time n. It can be seen that the short-time energy analysis performs exponential transformation on the signal first, and then processes it by segment or frame superposition. Framing can be continuous or overlapped, and it can be achieved by weighting with movable windows of limited length.

During on-site detection, since the circuit breaker works in a complex electromagnetic environment, there will inevitably be a large amount of random noise in the signal obtained from the vibration sensor. These include low-frequency interference signals generated by movement and friction of mechanism parts, as well as high-frequency noise signals caused by partial discharge and arc discharge. These interference signals have great randomness, which is very unfavorable for the extraction of the initial moment of the contact vibration event. After the original sampling signal is analyzed by short-term energy, the square processing of the signal highlights the useful signal with larger energy, and the noise signal with smaller value can be ignored. In addition, due to the local continuity of the shock signal, the values at several consecutive points at the beginning are large, while the noise is random, and the probability of obtaining a large value at several consecutive points is very low, so the energy accumulation through the window function can be Further weaken the influence of noise, thereby effectively improving the signal-to-noise ratio.

Different window functions (shape, length) will determine the results of short-term energy analysis. Therefore, the type and length of the appropriate window function must be selected according to the characteristics of the vibration signal. There are two commonly used window types to choose from in the program: rectangular window and Hamming window. Compared with the rectangular window, the Hamming window has larger bandwidth and faster out-of-band attenuation speed, and has less distortion to the input signal. Therefore, the Hamming window is selected in the process of processing the closing vibration signal of the vacuum circuit breaker. Extract the signal within 2 ms near event 1 in Figure 2 for short-term energy analysis, see the influence diagram of the window function length M on the short-term energy analysis results in Figure 5, the abscissa in the figure is the time axis, and the ordinate is the short-term energy , the four curves correspond to the short-term energy analysis results when the length of the Hamming window function is 10, 50, 100, and 150. It can be seen that when the sampling frequency is constant, the shorter the window length, the shorter the time lag T _di of the energy analysis (i=1, 2, 3, 4,) is smaller, and the event time obtained is closer to the real value; but at the same time, it is not conducive to the high signal-to-noise ratio of short-term energy analysis, and the application must balance the two to choose . Under the sampling rate of 25kHz, it is more appropriate to take the window length of 50-100 after repeated comparisons. The specific window length depends on the shape of the vibration signal of the circuit breaker obtained under different site environments. When the contained noise is weak, select a shorter window length to reduce the time lag of short-term energy analysis; when the contained noise is strong, select a longer window length to improve the signal-to-noise ratio.

According to the amplitude of the short-term energy analysis results of the vibration signal and the strength of the noise, select an appropriate threshold value (0~half of the first peak value (E _s ) in the short-term energy analysis waveform of the vibration signal). Thus, the starting time of the vibration event of the three-phase contacts is obtained, which respectively correspond to the closing moments of the contacts of each phase of the circuit breaker. Assuming that the maximum value of the short-term energy analysis of the external noise before the vibration event is N _max , then the threshold T _h is:

${\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; <figure-callout\; id="10"\; label="N\; max"\; filenames="C20041002626200161.png"\; state="\{\{state\}\}">N</figure-callout></span><figure-callout\; id="10"\; label="N\; max"\; filenames="C20041002626200161.png"\; state="\{\{state\}\}">\; </figure-callout>}}_{\mathrm{<span\; class="notranslate">\; </span>}}& \mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}\\ \mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; <figure-callout\; id="10"\; label="s"\; filenames="C20041002626200161.png"\; state="\{\{state\}\}">s</figure-callout></span>}}& \mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}\end{array}\right.$

According to the sequence of closing moments, the first closing phase and the last closing phase are judged, and the time difference between them is the closing synchronism. Combined with the coil current curve, the initial point of the coil current is taken as the moment when the closing command arrives, and the time difference between it and the closing moment of the last closing phase is the closing time of the circuit breaker.

Example

In the laboratory, three acceleration sensors of the type M353B18 are used to detect the closing vibration signal of the ZN12 vacuum circuit breaker. Before this, the transposition signal of the three-phase contact is measured offline to verify the effect of vibration detection. See Figure ₆ for _the closing transposition _signal and closing vibration signal waveform of the three-phase contact of the ZN12 vacuum circuit breaker. The vibration signal of C three-phase contact closing, C _a , C _b , and C _c are the transposition signals of A phase, B phase, and C phase contacts during the closing process, respectively. It can be seen that phase C is the first conjunct phase, followed by phase A and phase B in turn. The time difference dt between the two phases of B and C represents the closing synchronism of the vacuum circuit breaker.

For the short-term energy analysis of the vibration signal in Figure 6, refer to the waveform diagram of the short-term energy analysis of the closing vibration signal of the ZN12 vacuum circuit breaker shown in Figure 7. The abscissa in the figure is the time axis, S _a , S _b , S _c are the short-term energy analysis results of the vibration signals of A, B, and C three-phase contacts, and are aligned on the time axis based on the starting point of the closing coil current i. Compared with the original vibration signal, the short-time energy analysis greatly improves the signal-to-noise ratio, making the starting point of the three-phase contact vibration event clearly stand out. By selecting an appropriate threshold value, the starting times t ₁ , t ₂ and t ₃ of the vibration event of the three-phase contacts are obtained, which correspond to the closing moments of the contacts of each phase respectively. t ₂ -t ₃ represents the closing synchronism of the circuit breaker. The starting point of the closing coil current is similar to the moment when the closing command arrives, and phase B is the last closing phase, so _t2 represents the closing time of the circuit breaker. After the closing time of the circuit breaker is obtained, combined with the travel curve of the moving contact, the closing speed can also be obtained. In order to verify the stability of the method, five repeated experiments were carried out under the same mechanism state, and the analysis results are shown in Table 1.

The data in brackets in Table 1 represents the absolute error between the analysis result and the contact transposition signal measured off-line, and it can be seen that the maximum error is 0.2ms. For the detection of the mechanical state of the vacuum circuit breaker, this error is enough to distinguish the change of the contact closing time due to the change of the mechanism state. When it is used in the calculation of other mechanical parameters such as closing time, closing speed and closing synchronism, it can also achieve high detection accuracy.

See Figure 8, for the ZN12 vacuum circuit breaker, the opening distance and overtravel of the three-phase contacts can be adjusted by adjusting the length of the insulating rod, and the three-phase synchronicity will also change accordingly, so it can be used to simulate the closing of the circuit breaker contemporaneous changes. During the test, adjust the length of the C-phase insulated rod, and measure the vibration signal of the three-phase contact for every 2mm increase on the basis of the original length. Use the vibration analysis method described in the present invention to calculate the closing synchronism in each state, and compare the closing synchronism obtained with the offline contact transposition signal as the real value to obtain the vibration analysis shown in Figure 8 The comparison curve of the obtained closing synchronism and the real value, the abscissa in the figure is the length of the insulating rod of the ZN12 vacuum circuit breaker, and the ordinate represents the closing synchronism.

It can be seen from Figure 8 that with the increase of the C-phase insulating rod, the synchronism of three-phase closing increases accordingly. Moreover, the closing synchronism calculated by the vibration analysis method maintains a good consistency with the real value, and the maximum relative error is 4.3%.

Table 1 Short-term energy analysis results of ZN12 vacuum circuit breaker closing test Number of trials t ₁ /ms t ₂ /ms t ₃ /ms (t ₂ -t ₃ )/ms (t ₂ -t ₁ )/ms 1 66.04(-0.01) 67.08 (-0.09) 65.34 (-0.09) 1.74(0.00) 1.04(-0.08) 2 66.08(-0.01) 67.24 (-0.09) 65.70(0.11) 1.54(-0.20) 1.16(-0.08) 3 66.02 (-0.09) 67.18 (-0.09) 65.60(0.11) 1.58(-0.20) 1.16(0.00) 4 66.08(0.03) 67.16 (-0.05) 65.42 (-0.01) 1.74(-0.04) 1.08(-0.08) 5 66.08 (0.07) 67.12 (-0.05) 65.34 (-0.05) 1.78(0.00) 1.04(-0.12) average value 66.06 67.16 65.48 1.68 1.10

Claims (4)

1. An on-line detection method for the closing moment of the vacuum circuit breaker contacts based on vibration analysis, which is characterized in that it comprises the following steps: 1) Select the installation location and installation direction of the vibration sensor. The installation location of the sensor is point P at the end where the A, B, and C three-phase insulating rods are connected to the grounding part of the circuit breaker. Three vibration sensors of the same type are used to obtain three-phase The vibration signal of the contact, the installation direction is parallel to the insulating rod; 2) Set the sampling sequence, sampling frequency, sampling length and trigger mode and store the sampling data; 3) For the preprocessing of the vibration data, a limiting filter algorithm is used to smooth the mutation points caused by noise in the vibration signal; The above limiting filtering algorithm is as follows: First obtain the envelope of the absolute value of the vibration signal, by comparing the envelope amplitudes y(n) and y(n-1) of two sampling points at adjacent moments, if their difference exceeds the set limit value, it is considered that mutation interference has occurred, and the sampling value x(n) is regarded as an illegal value, and it is eliminated, and x(n-1) is used instead of x(n); the limit value is determined by the envelope The product of the maximum rate of change _Vmax at the beginning of the vibration signal and the sampling period T is determined; 4) Select the type and length of the moving window function, and use the short-term energy algorithm for analysis; The analysis of the above short-term energy algorithm is as follows: Suppose the vibration sequence is x(i), i=0,...,N-1, then the short-term energy function S(n) is defined as: $\mathrm{<span\; class="notranslate">\; S</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span><span\; class="notranslate">\; -</span><span\; class="notranslate">\; \&infin;</span>}{\overset{<span\; class="notranslate">\; +</span><span\; class="notranslate">\; \&infin;</span>}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\underset{\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; 1</span>}}{\overset{\mathrm{<span\; class="notranslate">\; no</span>}}{\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}}}{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; i</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; x</span>}}^{\mathrm{<span\; class="notranslate">\; 2</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; *</span>\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span>$ In the formula, w(n), n=0,..., M-1 is the moving window function, M is the length of the window function; S(n) represents the local energy of the signal at time n; in the analysis of the vacuum circuit breaker closing When the vibration signal is selected, the Hamming window is selected. When the sampling frequency is determined, the window length M is taken as 50-100; the Hamming window function is defined as: $\mathrm{<span\; class="notranslate">\; w</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; 0.54</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 0.46</span>}\mathrm{<span\; class="notranslate">\; cos</span>}<span\; class="notranslate">\; (</span>\frac{\mathrm{<span\; class="notranslate">\; 2</span>}\mathrm{<span\; class="notranslate">\; \&pi;n</span>}}{\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; 0</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; no</span>}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; m</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; 1</span>}<span\; class="notranslate">\; ;</span>$ 5) Select the threshold value to determine the closing moment of the three-phase contacts; 6) Calculate the closing synchronism and closing time according to the closing time of the three-phase contacts. 2. The online detection method for the closing time of vacuum circuit breaker contacts based on vibration analysis according to claim 1, characterized in that the set sampling sequence is phase A-phase B-phase C; the sampling frequency is ≥ 25kHz; Sampling starts with the rising edge of the closing coil current, and the sampling length is twice the closing time of the circuit breaker. 3. The online detection method for the closing time of vacuum circuit breaker contacts based on vibration analysis according to claim 1, characterized in that the selection range of the threshold value is: 0 to the first in the short-term energy analysis waveform of the vibration signal. half of a peak value E _s ; thus the starting time of the three-phase contact vibration event is obtained, which corresponds to the closing time of each phase contact of the circuit breaker; the maximum value of the short-term energy analysis of the external noise before the vibration event is N _max , then the selection of the threshold T _h is: ${\mathrm{<span\; class="notranslate">\; T</span>}}_{\mathrm{<span\; class="notranslate">\; h</span>}}<span\; class="notranslate">\; =</span>\left\{\begin{array}{cc}\mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}& \mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; \&le;</span>\mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}\\ \mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}& \mathrm{<span\; class="notranslate">\; 10</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; N</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; ></span>\mathrm{<span\; class="notranslate">\; 0.5</span>}<span\; class="notranslate">\; \&times;</span>{\mathrm{<span\; class="notranslate">\; E.</span>}}_{\mathrm{<span\; class="notranslate">\; the\; s</span>}}\end{array}\right.$ In the formula, E _s is the first peak value in the short-term energy analysis waveform of the vibration signal. 4. The online detection method for the closing time of vacuum circuit breaker contacts based on vibration analysis according to claim 1, characterized in that the calculation of the closing synchronism and closing time is as follows: according to the sequence of closing time Sequentially judge the first closing phase and the last closing phase, the time difference between them is the closing synchronism; take the initial point of the coil current as the moment when the closing command arrives, and the time difference between it and the closing moment of the last closing phase The time difference is the closing time of the circuit breaker.

Images