A vibration detection method for mechanical failure of circuit breaker

Abstract

The present invention relates to a vibration detection method for a circuit breaker mechanical fault, comprising the following steps: 1. magnetically adsorbing or adhering a vibration sensor to the surface of a housing at an operating mechanism of a circuit breaker under test; 2. closing the circuit breaker to excite vibration, whereupon the vibration sensor collects a vibration signal, transmits the signal to a vibration signal analysis and processing instrument, and records the vibration signal; 3. the vibration signal analysis and processing instrument obtains the magnitude of the vibration amplitude, performs a fast Fourier transform on the vibration signal within a range of 0-8000 Hz, and performs a three-layer wavelet packet decomposition on the vibration signal using a packet processing unit of an industrial computer; and 4. calculating the maximum amplitude a of the vibration, and obtaining the energy value e of each frequency band. i 5. The original data of the load vibration test of the circuit breaker during commissioning are also subjected to fast Fourier transform and three-layer wavelet packet decomposition in the range of 0-8000Hz to calculate the maximum amplitude A of the vibration and the energy E of each frequency band. i ; 6. Determine whether there is a spring fatigue loosening fault; 7. Determine whether there is a transmission mechanism jamming fault; 8. Display the fault condition.

Description

Technical Field

[0001] The present invention belongs to the technical field of circuit breaker fault detection in power systems, and in particular relates to a vibration detection method for mechanical faults of circuit breaker. Background Art

[0002] Circuit breakers are important equipment in power grids and play two roles in power systems: one is control, which means putting some power equipment or lines into or out of service according to the needs of power system operation; the other is protection, which means when a power equipment or line fails, the relay protection device acts on the circuit breaker to quickly cut off the faulty part from the power system, ensuring the normal operation of the fault-free part of the power system.

[0003] A circuit breaker typically consists of an operating mechanism, a conductive rod, and an energy storage spring. When the circuit is connected, the two conductive rods are in contact, and the energy storage spring is stretched and storing energy. To disconnect the circuit, the operating mechanism activates, and the energy storage spring rapidly contracts, separating the two connecting rods. To close the circuit, an electric motor drives the operating mechanism to bring the conductive rods into contact, thus energizing the circuit. Because circuit breakers frequently operate, and the energy storage spring drives the conductive rods and operating mechanism to move violently, mechanical failures such as loose springs and stuck connecting rods are common. Currently, there are no effective testing methods to detect and diagnose these circuit breaker failures. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies of the prior art and provide a vibration detection method for a mechanical fault of a circuit breaker.

[0005] The above-mentioned purpose of the present invention is achieved through the following technical solutions:

[0006] A vibration detection method for a circuit breaker mechanical fault, characterized in that the vibration detection device involved in the vibration detection method includes a vibration sensor, a vibration signal analysis and processing instrument, an industrial computer, and a display output device, wherein the industrial computer includes a wavelet packet processing unit, a calculation unit, a storage unit, and an output unit; and the method includes the following steps:

[0007] Step 1: The vibration sensor is magnetically adsorbed or adhered to the housing surface of the operating mechanism of the circuit breaker under test;

[0008] Step 2: Close the circuit breaker to excite vibration, and the vibration sensor collects the vibration signal and transmits it to the vibration signal analysis and processing instrument to record the vibration signal;

[0009] Step 3: The vibration amplitude is obtained by the vibration signal analysis and processing instrument, and the input vibration signal is subjected to fast Fourier transform in the range of 0-8000 Hz by the vibration signal analysis and processing instrument, and the vibration signal is subjected to three-layer wavelet packet decomposition by the packet processing unit of the industrial computer;

[0010] Step 4: Calculate the maximum amplitude a of the vibration, and obtain the energy value e of each frequency band based on fast Fourier transform and three-layer wavelet packet decomposition. i , where i = 1,…,8;

[0011] Step 5: The original data of the on-load vibration test of the circuit breaker during commissioning are also subjected to fast Fourier transform and three-layer wavelet packet decomposition in the range of 0-8000 Hz to calculate the maximum amplitude A of the vibration and the energy E of each frequency band. i , where i = 1,…,8;

[0012] Step 6: Determine whether there is a spring fatigue loosening fault. Compare the results calculated in Step 4 and Step 5. If the maximum amplitude a of the current measurement value is less than the maximum amplitude A of the original data by 20% or more, and the energy value e3 of the third frequency band of the tested circuit breaker is greater than the energy value E3 of the third frequency band in the original data by 20% or more, then it can be determined that the tested circuit breaker has a spring fatigue loosening fault. Otherwise, it is determined that there is no spring fatigue loosening fault.

[0013] Step 7: Determine whether there is a transmission mechanism jam fault. Compare the results calculated in Step 4 and Step 5. If the currently measured maximum amplitude a is less than the maximum amplitude A in the original data by 20% or more, and the sum of the energies of the 6th, 7th, and 8th frequency bands e0 of the tested circuit breaker is less than the sum of the energies of the 6th, 7th, and 8th frequency bands E0 in the original data by 20% or more, then it can be determined that the tested circuit breaker has a jam fault. Otherwise, it is determined that there is no jam fault.

[0014] Step 8: Display the fault of the combination electrical circuit breaker through the display output device.

[0015] Furthermore: the vibration sensor is an acceleration vibration sensor with a sensitivity of ≥500mV / g and a sampling frequency of ≥20kHz.

[0016] Furthermore: the sampling rate of the vibration signal analysis and processing instrument is ≥16kHz.

[0017] Further: In step 4, the energy value e of each frequency band i for:

[0018]

[0019] Where, X 3i(k)(i=1,…,8;k=1,2,…,N) is the amplitude of each discrete point of the wavelet packet reconstructed signal; i is the frequency band number of the third layer of wavelet decomposition; N is the number of sampling points; the characteristic quantity is constructed as the percentage of the energy of each frequency band to the total energy.

[0020] T=(E1,E2,E3…,E8)

[0021] The present invention has the following advantages and positive effects:

[0022] 1. The vibration detection method for circuit breaker mechanical faults proposed in the present invention can accurately determine whether the circuit breaker has mechanical faults, including spring fatigue loosening faults and / or jamming faults.

[0023] 2. The method proposed by the present invention is simple, fast, highly accurate, has no electrical connection with the equipment, and has no impact on the operating status of the equipment.

[0024] 3. The present invention introduces vibration testing technology into electrical equipment fault diagnosis, which can detect potential faults earlier than electrical testing methods. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 1 is a simplified structural diagram of a vibration detection device involved in the vibration detection method of the present invention;

[0026] Figure 2 is a flow chart of the vibration detection method of the present invention;

[0027] Figure 3 It is a time domain signal diagram of the circuit breaker of the present invention when it is in operation;

[0028] Figure 4 yes Figure 3 The spectrum obtained after fast Fourier transform;

[0029] Figure 5 This is the frequency band decomposition diagram of the three-layer wavelet packet of the present invention. DETAILED DESCRIPTION

[0030] The structure of the present invention will be further described below with reference to the accompanying drawings and through examples. It should be noted that the present examples are descriptive rather than restrictive.

[0031] A vibration detection method for mechanical faults in circuit breakers, see Figure 1-5The invention utilizes circuit breaker operation to excite vibration, uses a vibration sensor to collect the vibration signal, and outputs it to a vibration signal analysis and processing instrument. The instrument then determines the vibration amplitude and performs a fast Fourier transform on the signal to produce a vibration signal spectrum. The specific method involves selecting and arranging the vibration sensor, stimulating the vibration signal, analyzing the vibration signal, and selecting and determining the characteristic signal.

[0032] First, select an accelerometer with a sensitivity of 500mV / g or higher and a sampling frequency of 20kHz or higher. The vibration signal analysis and processing instrument (such as the Bentley 408 Vibration Data Acquisition and Analyzer or the Bentley SCOUT140) must have Fast Fourier Transform (FT) capabilities (such as the Bentley Vibration Data Acquisition and Analyzer software ADRE Sxp) and a sampling rate of 16kHz or higher. The industrial computer should include a wavelet packet processing unit, a calculation unit, a storage unit, and an output unit.

[0033] Circuit breakers have various shapes and structures. When conducting vibration tests, the sensor should be installed on the housing surface of the action mechanism. The installation method can be based on the actual situation. The installation should be firm. Figure 1 The basic principle of the test is to use the huge energy generated by the circuit breaker to stimulate vibration, use a vibration sensor to collect the vibration signal, use an analyzer and industrial computer to perform fast Fourier transform and wavelet packet decomposition on the signal, compare the current measurement data with the original data during the commissioning period, and compare the amplitude and energy ratio of different frequency bands to determine whether the abnormal circuit breaker has mechanical faults such as spring fatigue loosening or transmission mechanism jamming. The specific steps are as follows:

[0034] Step 1: The vibration sensor is magnetically adsorbed or adhered to the housing surface of the operating mechanism of the circuit breaker under test;

[0035] Step 2: Close the circuit breaker to excite vibration, and the vibration sensor collects the vibration signal and transmits it to the vibration signal analysis and processing instrument to record the vibration signal;

[0036] Step 3: The vibration amplitude is obtained by the vibration signal analysis and processing instrument, and the input vibration signal is subjected to fast Fourier transform in the range of 0-8000 Hz by the vibration signal analysis and processing instrument, and the vibration signal is subjected to three-layer wavelet packet decomposition by the packet processing unit of the industrial computer;

[0037] Step 4: Calculate the maximum amplitude a of the vibration, and obtain the energy value e of each frequency band based on fast Fourier transform and three-layer wavelet packet decomposition. i , where i = 1,…,8;

[0038] Step 5: The original data of the on-load vibration test of the circuit breaker during commissioning are also subjected to fast Fourier transform and three-layer wavelet packet decomposition in the range of 0-8000 Hz to calculate the maximum amplitude A of the vibration and the energy E of each frequency band. i , where i = 1,…,8;

[0039] Fast Fourier transform is a method of converting a complex time domain signal into a series of periodic sinusoidal signals. A time domain signal is a signal function with time as the horizontal coordinate and amplitude as the vertical coordinate. It is often a superposition of signals from multiple sources. After fast Fourier transform, the signal can be converted into a frequency domain signal with frequency as the horizontal coordinate and amplitude as the vertical coordinate, separating signals from different sources. Fast Fourier transform can decompose complex signals into the sum of signals of various frequencies, making it easier to analyze and study data signals. Figure 3 It is the time domain signal when the circuit breaker is operating. The horizontal axis is time. It can be seen that the circuit breaker has a huge amplitude in a very short time, showing the change characteristics of vibration in the time dimension. Figure 4 It will Figure 3 The spectrum obtained after fast Fourier transform shows the distribution characteristics of vibration in the frequency dimension.

[0040] The above-mentioned wavelet packet transform projects the signal onto a set of basis functions formed by wavelet expansion and contraction. The basic equation of Fourier transform is sine and cosine function. The basic equation of wavelet packet transform is composed of wavelet function. Wavelet packet decomposition is a method improved on the basis of wavelet transform. If the detection signal has a sufficient number of sampling points, wavelet packet decomposition can repeatedly decompose the input signal layer by layer with high accuracy on different frequency bands according to binary scale, and arrange these energies into feature vectors in scale order for identification. The decomposed signal can reflect the fault information in the entire frequency range, which is a more refined and accurate data analysis method.

[0041] The wavelet function Ψ(t) used to generate a set of different frequencies and time shifts is called the basic wavelet. The set of wavelet functions generated by it is a wavelet family of the basic wavelet, which can be expressed as: Where a is the scale parameter, which controls the scale (i.e., frequency) of the wavelet through scaling, and b is the translation parameter, which controls the position of the wavelet in the time domain through shifting. The order in which these two parameters are applied is translation first, then scaling. Normalizing this family of wavelet functions yields a set of wavelet function bases.

[0042] The vibration signal of GIS is a typical non-stationary transient signal. Wavelet packets can divide the GIS vibration signal into any fine frequency band at different scales. It has a finer resolution for the high-frequency part and is more suitable for time-frequency analysis and energy spectrum analysis of vibration signals. If the number of wavelet packet decomposition layers is small, the analysis speed is fast, the eigenvalues ​​of the frequency bands do not change significantly, and the frequency resolution is low; if the number of decomposition layers is large, the eigenvalues ​​of more frequency bands will change significantly, the analysis speed is slow, but the frequency resolution is high. Therefore, considering the time-varying and frequency-varying characteristics of the analyzed signal, the db3 wavelet function is used to perform a three-layer wavelet packet decomposition. Since the sampling frequency range of the experiment is 8000Hz, the original signal of 8000Hz is used for wavelet packet decomposition. The frequency band distribution of each frequency band is shown in the attached figure. Figure 2 As shown in the figure, the wavelet packet first decomposes the original signal W(0,0) into two parts, W(1,1) and W(1,2), according to the frequency band. W(1,1) indicates that the layer number and frequency band number of the wavelet packet decomposition are 1 and 1 respectively, and W(1,2) indicates that the layer number and frequency band number of the wavelet packet decomposition are 1 and 2 respectively. The definition of the subsequent components is similar. By performing n-layer wavelet packet decomposition, the original signal can be decomposed into 2 n Sub-bands are beneficial to provide frequency or time resolution. At the same time, wavelet packet decomposition has the characteristics of no omission and no redundancy, which is conducive to localized analysis. The energy calculation method of each frequency band is as follows:

[0043] Wavelet transforms can be used to decompose GIS vibration signals into independent frequency bands. The decomposed signals in different frequency bands all have a certain amount of energy, which is of great reference value for status monitoring and fault diagnosis during GIS system operation. Wavelet packet energy spectrum analysis is an example of this method that expresses the wavelet packet decomposition results in terms of energy in different frequency bands.

[0044] Find the energy of GIS vibration signal distributed in each frequency band. 3i The energy corresponding to (i=1,…,8) is E 3i (i=1,…,8), calculate the energy of each frequency band.

[0045]

[0046] Where X 3i (k) (i = 1, ..., 8; k = 1, 2, ..., N) is the amplitude of each discrete point in the wavelet packet reconstructed signal; i is the frequency band number of the third layer of the wavelet decomposition; N is the number of sampling points. The characteristic value is constructed as the percentage of the energy of each frequency band to the total energy.

[0047] T=(E1,E2,E3…,E8)

[0048] According to the corresponding relationship between each node of the wavelet packet decomposition tree and the subspace frequency band of the GIS vibration signal, the frequency bands of the GIS vibration signal 3-layer wavelet packet decomposition and their corresponding characteristic quantities are shown in Table 1:

[0049] Table 1: Frequency bands and their corresponding feature scales of 3-layer wavelet packet decomposition

[0050]

[0051] When different types of faults occur in the GIS, the frequency components of the GIS vibration signal will also be different. Different types of faults may cause some frequency components of the vibration signal to attenuate, while others may be enhanced. Therefore, when different GIS faults occur, the energy of the signals in different frequency bands will be quite different. The signal energy in some frequency bands will decrease, while the signal energy in other frequency bands will increase. Therefore, constructing a characteristic quantity with energy as the element is an intuitive and accurate diagnostic method.

[0052] Step 6: Determine whether a spring fatigue loosening fault exists. Compare the results calculated in steps 4 and 5. If the maximum amplitude a of the current measurement value is less than the maximum amplitude A of the original data by 20% or more, and the energy value e3 of the third frequency band of the circuit breaker under test is greater than the energy value E3 of the third frequency band in the original data by 20% or more, then it can be determined that the circuit breaker under test has a spring fatigue loosening fault. Otherwise, it is determined that there is no spring fatigue loosening fault.

[0053] Step 7: Determine whether there is a transmission mechanism jam. Compare the results calculated in Steps 4 and 5. If the currently measured maximum amplitude a is less than the maximum amplitude A in the original data by 20% or more, and the sum of the energies in the 6th, 7th, and 8th frequency bands e0 of the tested circuit breaker is less than the sum of the energies in the 6th, 7th, and 8th frequency bands E0 in the original data by 20% or more, then the circuit breaker under test is considered to have a jam. Otherwise, it is determined that there is no jam.

[0054] Step 8: Display the fault of the combination electrical circuit breaker through the display output device.

[0055] Taking the 221A interval circuit breaker of a certain substation combination electrical appliance as an example, it was found that the closing time of the circuit breaker was prolonged during operation. It was suspected that the spring was fatigued and loose or the transmission rod was stuck, so a vibration test was performed on it to determine whether there was a mechanical failure.

[0056] According to the method of the present invention:

[0057] Step 1: Adsorb or stick the vibration sensor to the surface of the housing of the circuit breaker under test. Figure 1 As shown;

[0058] Step 2: Close the circuit breaker to stimulate vibration and record the vibration signal using a sensor and analyzer.

[0059] Step 3: Perform fast Fourier transform and three-layer wavelet packet decomposition on the signal in the range of 0-8000Hz;

[0060] Step 4: Calculate the maximum vibration amplitude a = 7m / s 2 And the energy value of each frequency band e1=1m / s 2 、e2=2m / s 2 、e3=8m / s 2 、e4=0.8m / s 2 、e5=0.5m / s 2 、e6=0.4m / s 2 、e7=0.3m / s 2 、e8=0.4m / s 2 ;

[0061] Step 5: The original data of the load vibration test of the circuit breaker during commissioning is also subjected to fast Fourier transform and 3-layer wavelet packet decomposition in the range of 0-8000Hz to calculate the maximum amplitude of vibration A = 10m / s 2 And the energy of each frequency band E1=0.9m / s 2 、E2=2m / s 2 、E3=5m / s 2 、E4=0.9m / s 2 、E5=0.5m / s 2 、E6=0.5m / s 2 、E7=0.3m / s 2 、E8=0.2m / s 2 ;

[0062] Step 6: Determine whether there is a spring fatigue loosening fault: Compare the results calculated in steps 4 and 5, a = 7m / s 2 Less than the maximum amplitude of the original data A = 10m / s 2 Reach 30% and exceed 20%, and the energy value of the third frequency band of the tested circuit breaker is e3=8m / s 2 Greater than the energy value of the third frequency band in the original data E3 = 5m / s 2 Reaching 60% and exceeding 20% ​​meets the characteristics of determining that a spring fatigue loosening fault exists, so it is determined that the circuit breaker has a spring fatigue loosening fault;

[0063] Step 7: Determine whether there is a transmission mechanism jam fault: Compare the results calculated in steps 4 and 5, a = 7m / s 2 Less than the maximum amplitude of the original data A = 10m / s 2Reaching 30% and exceeding 20%, the sum of the energy of the 6th, 7th, and 8th frequency bands e0=1.1m / s 2 Greater than the sum of the energy of the 6th, 7th and 8th frequency bands in the original data E0 = 1m / s 2 , does not meet the characteristics of determining that there is a transmission mechanism jam fault, so it is determined that there is no transmission mechanism jam fault in the circuit breaker.

[0064] Through the above steps, it is determined that the circuit breaker has a loose spring fault, and there is no transmission mechanism jamming fault.

[0065] Although the embodiments and drawings of the present invention are disclosed for illustrative purposes, those skilled in the art will understand that various replacements, changes and modifications are possible without departing from the spirit of the present invention and the appended claims. Therefore, the scope of the present invention is not limited to the contents disclosed in the embodiments and drawings.

Claims

1. A vibration detection method for a circuit breaker mechanical fault, characterized in that: The vibration detection method involves a vibration detection device including a vibration sensor, a vibration signal analysis and processing instrument, an industrial computer, and a display output device. The industrial computer includes a wavelet packet processing unit, a calculation unit, a storage unit, and an output unit. The method includes the following steps: Step 1: The vibration sensor is magnetically adsorbed or adhered to the housing surface of the circuit breaker operating mechanism under test; Step 2: Close the circuit breaker to stimulate vibration, and the vibration sensor collects the vibration signal and transmits it to the vibration signal analysis and processing instrument to record the vibration signal; the vibration sensor is an acceleration vibration sensor with a sensitivity of ≥500mV / g and a sampling frequency of ≥20kHz; Step 3: The vibration amplitude is obtained by the vibration signal analysis and processing instrument, and the input vibration signal is subjected to fast Fourier transform in the range of 0-8000 Hz by the vibration signal analysis and processing instrument, and the vibration signal is subjected to three-layer wavelet packet decomposition by the packet processing unit of the industrial computer; Step 4: Calculate the maximum amplitude a of the vibration, and obtain the energy value e of each frequency band based on fast Fourier transform and three-layer wavelet packet decomposition. i , where i = 1,…,8; the energy value of each frequency band is e i for: Where, X 3i (k) (i = 1, ..., 8; k = 1, 2, ..., N) is the amplitude of each discrete point of the wavelet packet reconstructed signal; i is the frequency band number of the third layer of wavelet decomposition; N is the number of sampling points; the characteristic quantity is constructed as the percentage of each frequency band energy to the total energy as follows: T=(E1,E2,E3…,E8); Step 5: The original data of the on-load vibration test of the circuit breaker during commissioning are also subjected to fast Fourier transform and three-layer wavelet packet decomposition in the range of 0-8000 Hz to calculate the maximum amplitude A of the vibration and the energy E of each frequency band. i， Where i = 1,…,8; Step 6. Determine whether a spring fatigue loosening fault exists: Compare the results calculated in Step 4 and Step 5. If the maximum amplitude a of the current measurement value is less than the maximum amplitude A of the original data by 20% or more, and the energy value e3 of the third frequency band of the circuit breaker under test is greater than the energy value E3 of the third frequency band in the original data by 20% or more, then it can be determined that the circuit breaker under test has a spring fatigue loosening fault. Otherwise, it is determined that no spring fatigue loosening fault exists. Step 7: Determine whether there is a transmission mechanism jam fault: Compare the results calculated in Step 4 and Step 5. If the currently measured maximum amplitude a is less than the maximum amplitude A in the original data by 20% or more, and the sum of the energies in the 6th, 7th, and 8th frequency bands e0 of the tested circuit breaker is less than the sum of the energies in the 6th, 7th, and 8th frequency bands E0 in the original data by 20% or more, then it can be determined that the tested circuit breaker has a jam fault. Otherwise, it is determined that there is no jam fault. Step 8: Display the fault of the combination electrical circuit breaker through the display output device.

2. The vibration detection method for circuit breaker mechanical failure according to claim 1, characterized in that: The sampling rate of the vibration signal analysis and processing instrument is ≥16kHz.

Images