Equipment and method for monitoring abrasion condition of friction plate of yaw brake of fan

Abstract

The invention relates to the technical field of fan yaw brakes, in particular to a device and method for monitoring the abrasion condition of a friction plate of a fan yaw brake, and the device comprises a sound wave receiver, a signal processor, a signal analyzer and an indicator lamp type fault display. The sound wave receiver receives sound generated when each yaw brake acts and transmits the sound to the signal processor; the signal processor carries out filtering noise reduction, positioning, sound pressure analysis and spectrum analysis processing on the received sound waves; the signal analyzer can judge the working state of the yaw brake according to the sound position, the sound pressure and the frequency spectrum condition; the indicator lamp type fault display can display the service life condition of the friction plate; by analyzing the position, the frequency spectrum and the sound pressure of the sound signals, the abrasion conditions of the friction plates of the yaw brake can be accurately judged at the same time, false alarm or missing alarm is not likely to happen when the conditions of oil leakage and the like of the yaw brake exist, the friction plates can be replaced in time conveniently, and manpower and equipment cost is saved.

Description

Technical Field

[0001] The invention relates to the technical field of wind turbine yaw brakes, and in particular to a device and a method for monitoring the wear of a friction plate of a wind turbine yaw brake. Background Art

[0002] The wind turbine uses the yaw motor to drive the brake of the yaw assembly to brake the yaw gear ring, which can make the wind turbine glide stably when yawed, and the yaw motor will also brake through its own friction plate when the wind turbine is stable. However, the friction plate of the yaw motor will wear out more with the increase of usage time. The yaw brake is usually used to ensure the smooth rotation of the cabin when the cabin is facing the wind, and to lock and maintain the position of the cabin after the wind is facing.

[0003] The current detection methods are divided into two categories: periodic indirect measurement or non-contact sensor measurement such as laser or eddy current. The periodic indirect measurement method does not require the installation of instruments and equipment, and is simple to operate. It only needs to use a caliper to measure the remaining thickness to calculate the worn thickness. However, the periodic indirect measurement method has a large error, and it is impossible to climb the wind turbine to carry out measurement in severe weather such as strong winds, and the timeliness is poor. It is often replaced in advance or replaced after a long period of failure; the non-contact sensor method has higher accuracy and the cost of using the non-contact sensor method is also high. The Chinese patent number "CN106286651A" and the patent name "Wind power brake friction plate wear state detector" discloses a detector for monitoring the wear state of the friction plate, including a detection module and a signal acquisition module. The detection module includes a sensor, and the sensor is used to feedback parameters in real time and make judgments, which brings convenience to maintenance and repair. However, the patent requires a detector to be installed near each yaw brake, which is costly and occupies a large area. When some yaw brakes have oil leakage, false alarms or missed alarms are prone to occur. Summary of the invention

[0004] The object of the present invention is to provide a device and method for monitoring the wear condition of the friction plate of the yaw brake of a wind turbine, which can accurately determine the wear condition of the friction plate of the yaw brake. One device can monitor multiple yaw brakes, saving manpower and floor space, and is not prone to false alarms when some yaw brakes have oil leakage or the like.

[0005] To achieve the above-mentioned purpose, the present invention provides the following technical solutions: a device for monitoring the wear of the friction plate of the yaw brake of a wind turbine, comprising an acoustic wave receiver, a signal processor, a signal analyzer and an indicator light-type fault display; the signal processor and the signal analyzer are both arranged in the same housing, the number of the acoustic wave receivers is set to 2-6, the acoustic wave receivers are evenly fixed to the outside of the housing by a fixed bracket and are 10cm-30cm away from the housing, the indicator light-type fault display is arranged on the outside of the housing, the indicator light-type fault display is provided with indicator lights, the number of the indicator lights is consistent with the number of yaw brakes, and the indicator lights correspond to the yaw brakes.

[0006] A method for monitoring the wear of a friction plate of a wind turbine yaw brake comprises the following steps:

[0007] Step 1: The acoustic wave receiver receives the sound signal generated when each yaw brake is actuated and transmits the sound signal to the signal processor;

[0008] Step 2: The signal processor performs filtering and noise reduction processing on the collected sound signal;

[0009] Step 3: There is a time delay between the sound signals emitted by the same yaw vibrator received by the sound wave receiver. The time delay value can be obtained through the functional relationship between the sound signals. According to the Wiener-Khinchin theorem and Fourier transform, the generalized cross-correlation function between the two sound wave signals can be obtained. A rectangular coordinate system is established according to the generalized cross-correlation function, and the position coordinates of the sound signal can be obtained;

[0010] Step 4: The signal processor extracts the spectrum of the sound signal through Fourier transform and forms a spectrum diagram, and determines the main frequency of the sound wave and the sound pressure at the main frequency according to the spectrum diagram;

[0011] Step 5: The signal analyzer obtains the sound position, main frequency and sound pressure value outputted from the signal processor. The signal analyzer can determine the deviation between the main frequency and sound pressure value of each yaw brake and the average value of the main frequency and sound pressure value of the remaining yaw brakes through analysis, and the severity of the fault can be determined by the size of the deviation value;

[0012] Step 6: Depending on the severity of the fault of each yaw brake, the corresponding indicator light displays the corresponding color.

[0013] Preferably, in step 2, the calculation formula for filtering and noise reduction by the signal processor is:

[0014]

[0015] Where ωi is the weight coefficient, e(n) is the output signal error, Xi(n) is the i-th sound signal collected, i = 1, 2, ..., N, in order to minimize the error signal, the iterative formula of the weight coefficient ωi is:

[0016] W(n+1)=W(n)-μ*e(n)*X(n)

[0017] Where μ is the convergence factor, e(n) is the output signal error, and X(n) is the input signal.

[0018] Preferably, in step 3, the specific method of determining the position coordinates of the sound signal is as follows:

[0019] Step 3.1: Determine the functional relationship between the two sound signals:

[0020]

[0021] Among them, x1(t) and x2(t) are the sound signals received by the two sound wave receivers, S1 and S2 are the amplitude parameters of the transmitted signals, n1(t) and n2(t) are noise signals, τ1 and τ2 are the time delays of signal transmission, and τ1≤τ2;

[0022] Step 3.2: According to the Wiener-Hinchin theorem and Fourier transform, the cross-correlation function between the two sound wave signals is as follows:

[0023]

[0024] Where G12(ω) is the cross power spectrum function of x1(t) and x2(t), e jωt represents a rotation factor, j is an imaginary unit, ω is the angular frequency, and t is time; when x1(t) and x2(t) are filtered, the cross-power spectrum function of the output signal is:

[0025]

[0026] Where H1 and H2 represent filter functions, H2*(ω) represents the conjugate function of H2(ω), and G12(ω) is the cross-power spectrum function of x1(t) and x2(t);

[0027] Step 3.3: Through Fourier transform of R12(τ) and Gg12(ω), the generalized cross-correlation function of x1(t) and x2(t) is obtained as:

[0028]

[0029] Where G12(ω) is the cross-power spectrum function of x1(t) and x2(t), ejωt represents a rotation factor, j is an imaginary unit, ω is the angular frequency, t is time, and φ12(ω) is the weighting function in the frequency domain;

[0030] Step 3.4: Establishing a rectangular coordinate system according to the generalized cross-correlation function: the acoustic wave receiver divides the rectangular coordinate system plane into equal areas;

[0031] Step 3.5: The sampling period difference of the sound wave signals received by each two sound wave receivers can be obtained by cross-correlation calculation, and the specific position of the sound signal in the rectangular coordinate system can be determined by the positive and negative values ​​of the sampling period difference;

[0032] Step 3.6: Determine the position of the sound signal: The distance between the sound signal and the sound wave receiver is obtained through the sampling period difference and the sampling frequency, so the hyperbolic trajectory of the sound signal in the rectangular coordinate system can be determined, and the position coordinates of the sound signal relative to the sound wave receiver can be obtained according to the direction of the sound signal and the coordinates of the intersection between the hyperbolas.

[0033] Preferably, in step 4, the steps of calculating the sound pressure and spectrum signal are:

[0034] Step 4.1: Perform Fourier transform on the sound wave signal. The specific formula is as follows:

[0035]

[0036] Where f(t) is the time domain signal, f(ω) is the frequency domain signal, ω is the angular frequency, and j is the imaginary unit;

[0037] Step 4.2: According to the formula in step 4.1, perform spectrum extraction on the sound signal. The formula after extraction is as follows:

[0038]

[0039] Among them, ωn is the angular frequency, an and bn are Fourier coefficients, and an and bn can be obtained by integration:

[0040]

[0041]

[0042] By using the above formula, the sound pressure corresponding to different frequencies is plotted in the coordinate system to form a spectrum diagram;

[0043] Step 4.3: According to step 4.2, determine the main frequency of the sound signal, that is, the frequency corresponding to the maximum sound pressure value is the main frequency;

[0044] Step 4.4: Through the above method, the main frequency and the sound pressure magnitude at the main frequency of all yaw brakes can be obtained, which are recorded as fn and pn respectively;

[0045] Step 4.5: Perform summary analysis on all fn and pn.

[0046] Preferably, in step 5, the method for determining the deviation between the main frequency and sound pressure value of each yaw brake and the other yaw brakes is as follows:

[0047] When a yaw brake is working, if the deviation between its main frequency and the average value of the main frequencies of other yaw brakes is less than 30 Hz and the sound pressure generated is less than 10% of the average sound pressure of the other yaw brakes, that is, it is within the qualified range, it can be judged that the life of the yaw brake friction plate is in a qualified state.

[0048] When a yaw brake is working, if the deviation between its main frequency and the average value of the main frequencies of other yaw brakes is between 30Hz and 50Hz and the sound pressure generated is between 10% and 20% of the average sound pressure of other yaw brakes, it is in the critical range. It can be judged that the life of the friction plate of the yaw brake is about to end and the situation needs to be checked as soon as possible.

[0049] When a yaw brake is working, if the deviation between its main frequency and the average value of the main frequencies of other yaw brakes is greater than 50Hz and the sound pressure generated is more than 20% greater than the average sound pressure of the other yaw brakes, that is, it exceeds the critical range, it can be judged that the life of the yaw brake friction plate has ended and needs to be replaced in time.

[0050] Preferably, in step 6, the color of the indicator light is displayed as follows:

[0051] When the signal analyzer determines that the life of a yaw brake friction pad is within the qualified range, the indicator light displays a green light; when the signal analyzer determines that the life of a yaw brake friction pad is about to end, the indicator light displays a yellow light; when the signal analyzer determines that the life of a yaw brake friction pad has ended, the indicator light displays a red light.

[0052] Preferably, in step 4.1, the calculation formula for the sound pressure and the spectrum signal may also adopt the Mel-frequency cepstral coefficient method.

[0053] Compared with the prior art, the present invention has the following beneficial effects:

[0054] 1. The present invention can process and analyze the sound signals generated by multiple yaw vibrators at the same time, that is, it can monitor the wear of the friction plates of multiple yaw vibrators at the same time, saving manpower and equipment costs.

[0055] 2. The present invention can accurately judge the remaining life of the yaw brake friction plate by analyzing the spectrum and sound pressure of the sound signal, and is less likely to cause false alarms or missed alarms when some yaw brakes have oil leakage.

[0056] 3. The present invention can display the wear condition of each yaw vibrator through an indicator light, and the friction plate can be adjusted in time according to the wear condition. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] Figure 1 A flow chart of a method for monitoring the wear of a friction plate of a wind turbine yaw brake according to an embodiment of the present invention;

[0058] Figure 2 A schematic diagram of the three-dimensional structure of a device according to an embodiment of the present invention;

[0059] Figure 3 A distribution diagram of the positions of the sound wave receivers according to an embodiment of the present invention in a rectangular coordinate system;

[0060] Figure 4 A waveform diagram of a sound signal received by a sound wave receiver according to an embodiment of the present invention;

[0061] Figure 5 A waveform diagram of a sound signal received by a sound wave receiver according to an embodiment of the present invention after filtering and noise reduction;

[0062] Figure 6 FIG. 4 is a spectrum diagram of a sound signal generated by a yaw vibrator according to an embodiment of the present invention.

[0063] In the figure: 1. sound wave receiver; 2. signal processor; 3. signal analyzer; 4. indicator light type fault display; 5. housing; 6. fixing bracket; 7. indicator light; 8. first sound wave receiver; 9. second sound wave receiver; 10. third sound wave receiver; 11. fourth sound wave receiver; 12. first hyperbola; 13. second hyperbola. DETAILED DESCRIPTION

[0064] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0065] Please refer to the embodiment of the present invention. Figure 1-2A device for monitoring the wear of the friction plate of the yaw brake of a wind turbine comprises a sound wave receiver 1, a signal processor 2, a signal analyzer 3 and an indicator light type fault display 4; the sound wave receiver 1 is used to receive the sound generated when each yaw brake is in action, and transmit it to the signal processor 2; the signal processor 2 is used to filter and reduce noise, locate, analyze sound pressure, and analyze spectrum for the received sound waves; the signal analyzer 3 can determine the working state of each yaw brake through the sound position, sound pressure, and spectrum obtained by the signal processor 2, and perform comparative analysis; the indicator light type fault display 4 is used to display the current life of the yaw brake friction plate; the signal processor 2 and the signal analyzer 3 are both arranged at the same In a shell 5, in this embodiment, the shell 5 is configured as a cylinder; the number of sound wave receivers 1 is set to 2-6, in this embodiment, the number of sound wave receivers 1 is set to 4, each adjacent sound wave receiver 1 is distributed at 90°, the sound wave receiver 1 is evenly fixed to the outside of the shell 5 by a fixed bracket 6 and is 10cm-30cm away from the shell 5, the fixed bracket 6 is also used to transmit the sound signal to the signal processor 2, in this embodiment, the distance between the sound wave receiver 1 and the shell 5 is set to 20cm, the indicator light type fault display 4 is arranged on the outside of the shell 5, the indicator light type fault display 4 is provided with an indicator light 7, the number of the indicator lights 7 is consistent with the number of yaw brakes, and the indicator lights 7 correspond to the yaw brakes.

[0066] A method for monitoring the wear of a friction plate of a wind turbine yaw brake comprises the following steps:

[0067] Step 1: The sound wave receiver 1 receives the sound signal generated when each yaw brake is actuated and transmits the sound signal to the signal processor 2;

[0068] Step 2: The signal processor 2 performs filtering and noise reduction processing on the collected sound signal;

[0069]

[0070] where ω i is the weight coefficient, e(n) is the output signal error, X i (n) is the i-th sound signal collected, i = 1, 2, ..., N, and the weight coefficient ω is used to minimize the error signal. i The iteration formula is:

[0071] W(n+1)=W(n)-μ*e(n)*X(n)

[0072] Where μ is the convergence factor, e(n) is the output signal error, and X(n) is the input signal.

[0073] The specific filtering and noise reduction steps are as follows:

[0074] Step 2.1: Initialize the filter coefficient vector W(0);

[0075] Step 2.2: Calculate the output signal y(n);

[0076] Step 2.3: Calculate the error of the output signal as:

[0077] e(n)=d(n)-y(n)

[0078] Where d(n) is the desired output signal;

[0079] Step 2.4: Update the filter coefficient vector:

[0080] W(n+1)=W(n)-μ*e(n)*X(n)

[0081] Step 2.5: Repeat steps 2.2-2.4 until convergence is reached, which means the denoising is complete.

[0082] Step 3: There is a time delay between the sound signals emitted by the same yaw vibrator received by the sound wave receiver 1. The time delay value can be obtained through the functional relationship between the sound signals. According to the Wiener-Khinchin theorem and Fourier transform, the generalized cross-correlation function between the two sound wave signals can be obtained; a rectangular coordinate system is established based on the generalized cross-correlation function, and the position coordinates of the sound signal can be calculated.

[0083] Step 3.1: Determine the functional relationship between the two sound signals:

[0084]

[0085] Among them, x1(t) and x2(t) are the sound signals received by the two sound wave receivers 1, S1 and S2 are the amplitude parameters of the transmitted signals, n1(t) and n2(t) are noise signals, τ1 and τ2 are the delays of signal transmission, and τ1≤τ2; normalized with x1(t) as the standard, formula (1) can be converted to:

[0086]

[0087] By performing discrete transformation on equation (2), we can obtain:

[0088]

[0089] Among them, λ=S1S2, which represents the ratio of the amplitudes of the two signals S1 and S2, D=τ2-τ1, which represents the obtained delay value, and the correlation function of x1(t) and x2(t) is:

[0090]

[0091] Assuming that n1(t) and n2(t) are white noises independent of the signal source emitting the signal s(t) and are uncorrelated with each other, the cross-correlation function between the signal s(t) and the two noise sources n1(t) and n2(t) is:

[0092]

[0093]

[0094] The functional relationship between the two noise sources can be expressed as:

[0095]

[0096] It can be introduced:

[0097] R12(τ)=λR SS (τ-D)

[0098] By the autocorrelation function |R SS From the main characteristics and basic properties of (τ)|≤Rss(0), it can be obtained that when τ=D, Rss(τ-D) is the maximum, that is, the value of R12(τ) also reaches the maximum. Therefore, τ corresponding to the peak value of R12(τ) is the harmonic frequency and signal delay between the two sound wave receivers 1.

[0099] Step 3.2: According to the Wiener-Khinchin theorem, the power spectral density of any wide-sense stationary random process with a constant mean is the Fourier transform of its autocorrelation function. The cross-correlation function of x1(t) and x2(t) can be obtained as:

[0100]

[0101] Among them, G 12 (ω) is the cross power spectrum function of x1(t) and x2(t), e jωt represents a rotation factor, j is an imaginary unit, ω is the angular frequency, and t is time; when x1(t) and x2(t) are filtered, the cross-power spectrum function of the output signal is:

[0102]

[0103] Among them, H1 and H2 represent filter functions, H2 * (ω) represents the conjugate function of H2(ω), G 12 (ω) is the cross power spectrum function of x1(t) and x2(t);

[0104] Step 3.3: R 12 (τ) and G g12 (ω) The generalized cross-correlation function of x1(t) and x2(t) is obtained by Fourier transform:

[0105]

[0106] Among them, G 12 (ω) is the cross power spectrum function of x1(t) and x2(t), e jωt represents a rotation factor, j is the imaginary unit, ω is the angular frequency, t is the time, φ 12 (ω) is the weighting function in the frequency domain;

[0107] Step 3.4: According to the generalized correlation function, establish a rectangular coordinate system, see Appendix Figure 3 The four sound wave receivers 1 are numbered as the first sound wave receiver 8, the second sound wave receiver 9, the third sound wave receiver 10 and the fourth sound wave receiver 11 respectively. The first sound wave receiver 8 is located in the fourth quadrant, the second sound wave receiver 9 is located in the first quadrant, the third sound wave receiver 10 is located in the second quadrant, and the fourth sound wave receiver 11 is located in the third quadrant. The four sound wave receivers 1 are distributed in a square shape and are connected by the coordinate axes OX, OY and the diagonals of the four sound wave receivers 1. The entire plane can be divided into eight areas, which are named in clockwise direction for ease of calculation: +1, +2, +3, +4, -4, -3, -2, and -1 areas.

[0108] Step 3.5: max12 represents the sampling period difference between the sound wave signal received by the first sound wave receiver 8 and the sound wave signal received by the second sound wave receiver 9, which is calculated by cross-correlation. max23 represents the sampling period difference between the sound wave signal received by the second sound wave receiver 9 and the sound wave signal received by the third sound wave receiver 10, which is calculated by cross-correlation. max13 represents the sampling period difference between the sound wave signal received by the first sound wave receiver 8 and the sound wave signal received by the third sound wave receiver 10, which is calculated by cross-correlation. max24 represents the sampling period difference between the sound wave signal received by the second sound wave receiver 9 and the sound wave signal received by the fourth sound wave receiver 11, which is calculated by cross-correlation. The positive or negative value of max12 can determine whether the sound signal is above or below the OX axis. If it is positive, the sound signal is above the OX axis, and if it is negative, the sound signal is below the OX axis. The sound signal can be divided into eight quadrants by max23, max13, and max24, and the specific area of ​​the sound signal in the eight quadrants of the rectangular coordinate system can be determined.

[0109] Step 3.6: After determining the direction of the sound signal, the method for determining the location of the sound signal is as follows: Figure 3, the distance of the sound signal from the first sound wave receiver 8 and the second sound wave receiver 9 can be obtained by max12 and the signal sampling frequency. The trajectory at a certain distance from the two fixed points is a hyperbola, and it can be determined that the sound signal is on the hyperbola. Therefore, the trajectory of the first hyperbola 12 of the sound signal can be determined by max12, and the second hyperbola trajectory 13 of the sound signal can be determined by the value of max23. The intersection of the first hyperbola 12 and the second hyperbola 13 can obtain four intersection points. Combined with the direction of the sound signal determined in step 3.5, one of the four intersection points can be determined as the position of the sound signal. By obtaining the coordinates of the intersection points, the position coordinates of the sound signal relative to the sound wave receiver 1 can be obtained. According to the principle of coordinate relativity, the position coordinates of the sound signal can be obtained from the position coordinates of the sound signal.

[0110] Step 4: The signal processor 2 extracts the spectrum of the sound signal through Fourier transform and forms a spectrum diagram, and determines the main frequency of the sound signal and the sound pressure at the main frequency according to the spectrum diagram.

[0111] Step 4.1: Perform Fourier transform on the sound wave signal. The specific formula is as follows:

[0112]

[0113] Among them, f(t) is the time domain signal, f(ω) is the frequency domain signal, ω is the angular frequency, and j is the imaginary unit; the Fourier transform extracts the frequency components contained in the signal by weighted averaging the time domain signal. The Fourier transform method is simple and reliable, and other frequency conversion methods can also be used, such as the Mel-frequency cepstrum coefficient method.

[0114] Step 4.2: According to the formula in step 4.1, perform spectrum extraction on the sound signal. The formula after extraction is as follows:

[0115]

[0116] Among them, ω n is the angular frequency, a n and b n is the Fourier coefficient, a n and b n It can be obtained by integration:

[0117]

[0118]

[0119] By using the above formula, the sound pressure corresponding to different frequencies is plotted in the coordinate system to form a spectrum diagram;

[0120] Step 4.3: According to step 4.2, determine the main frequency of the sound signal, that is, the frequency corresponding to the maximum sound pressure value is the main frequency;

[0121] Step 4.4: Through the above method, the main frequency and the sound pressure at the main frequency of all yaw brakes can be obtained, which are denoted as f and n and p n ;

[0122] Step 4.5: For all f n and p n Conduct summary analysis.

[0123] Step 5: The signal analyzer 3 obtains the sound position, main frequency and sound pressure value output by the signal processor 2. The signal analyzer 3 can determine the deviation between the main frequency and sound pressure value of each yaw brake and the average value of the main frequency and sound pressure value of the remaining yaw brakes through analysis. The severity of the fault can be determined by the size of the deviation value. The specific determination method is as follows:

[0124] When a yaw brake is working, if the deviation between the main frequency and the average value of the main frequencies of other yaw brakes is less than 30Hz and the sound pressure generated is less than 10% of the average sound pressure of other yaw brakes, it is within the qualified range, and it can be judged that the life of the yaw brake friction plate is in a qualified state;

[0125] When a yaw brake is working, if the deviation between its main frequency and the average value of the main frequencies of other yaw brakes is between 30Hz and 50Hz and the sound pressure generated is between 10% and 20% of the average sound pressure of other yaw brakes, that is, it is in the critical range, then it can be judged that the life of the friction pad of the yaw brake is about to end and the condition of the friction pad needs to be checked as soon as possible.

[0126] When a yaw brake is working, if the deviation between its main frequency and the average value of the main frequencies of other yaw brakes is greater than 50Hz and the sound pressure generated is more than 20% greater than the average sound pressure of the remaining yaw brakes, that is, it exceeds the critical range, it can be judged that the life of the yaw brake friction plate has ended and the friction plate needs to be replaced in time.

[0127] Step 6: According to the severity of the fault of each yaw brake, the corresponding indicator light 7 displays a corresponding color, and the displayed colors are as follows:

[0128] When the signal analyzer 3 determines that the service life of a yaw brake friction pad is within the qualified range, the corresponding indicator light 7 displays a green light; when the signal analyzer 3 determines that the service life of a yaw brake friction pad is about to end, the corresponding indicator light 7 displays a yellow light; when the signal analyzer 3 determines that the service life of a yaw brake friction pad has ended, the corresponding indicator light 7 displays a red light.

[0129] In order to better illustrate the effectiveness and superiority of the technical solution of the present invention, the following comparative analysis is made on the embodiments of the present invention after applying the above technical solution:

[0130] The present invention is placed in the center of the wind turbine yaw platform. In this embodiment, the number of yaw vibrators and indicator lights 7 is 8. First, the sound signal generated when the yaw brake is in action is collected. Please refer to the attached Figure 4 , which is a waveform diagram of a sound signal received by one of the sound wave receivers 1. The sound signal is filtered and denoised by the signal processor 2. Please refer to the attached Figure 5 , is the sound signal after filtering and noise reduction; Figure 4 and Figure 5 In contrast, the sound signal is significantly filtered and denoised by the signal processor 2, and the signal-to-noise ratio is increased from 40 dB before filtering and denoising to more than 110 dB.

[0131] According to the technical solution of the present invention, the specific positions and spectrograms of the eight sound signals can be obtained. Figure 6 , which is a spectrum diagram of the sound signals of eight yaw vibrators in an embodiment of the present invention. By performing sound pressure and spectrum analysis on the eight sound signals, the wear conditions of the eight yaw brake friction plates are obtained.

[0132] The main frequencies and sound pressure values ​​of the 8 yaw brake sound signals are shown in the following table:

[0133] Yaw brake number Main frequency (Hz) Sound pressure value (Pa) 1 6.0 8.4 2 6.1 8.8 3 6.4 9.5 4 6.1 8.6 5 38.4 30.7 6 6.4 9.4 7 6.4 9.5 8 6.5 9.6

[0134] After calculating the main frequency and sound pressure deviation value of each yaw brake, the main frequency of yaw brake No. 5 is 38.4Hz, and the sound pressure value is 30.71Pa. The average main frequency of the other 7 yaw brakes is 6.2Hz, and the average sound pressure value is 9.1Pa. The deviation between the main frequency of yaw brake No. 5 and the average main frequency of other yaw brakes is greater than 30Hz, and the sound pressure generated is more than 10% higher than the average sound pressure of the other yaw brakes. It can be judged that the life of the friction plate of yaw brake No. 5 is about to end, and the condition of the friction plate needs to be checked as soon as possible. Except for yaw brake No. 5, the main frequency and sound pressure values ​​of the other yaw brakes are within the acceptable range of deviation values, and there is no need to replace the friction plate.

[0135] In order to verify the ability of the technical solution described in the present invention to accurately determine the wear condition of the friction plate, this embodiment uses a manual measurement method to compare with the results of the embodiment of the present invention. The results are shown in the following table, which uses a caliper to measure the remaining thickness of 8 yaw brake friction plates.

[0136] Yaw brake number Caliper measurement of the remaining thickness of the friction plate (mm) 1 4.5 2 4.3 3 4.2 4 4.4 5 2.2 6 4.2 7 4.2 8 4.1

[0137] The results show that the remaining thickness of the friction pad of the No. 5 yaw brake is significantly lower than that of the other yaw brakes, and the friction pad of the No. 5 yaw brake is close to the safe value of the friction pad thickness specified by the manufacturer, which is 2mm. It is recommended to replace it as soon as possible. According to the comparison results, the present invention can accurately and quickly determine the life of the yaw brake friction pad and provide early warning in time.

[0138] The present invention can accurately judge the remaining service life of the yaw brake friction plate by analyzing the position, spectrum and sound pressure of the sound signal, and is not prone to false alarms or missed alarms when some yaw brakes have oil leakage, etc., so it is convenient to replace the friction plate in time. At the same time, the wear of multiple yaw vibrator friction plates can be monitored, saving manpower and equipment costs.

[0139] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0140] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A device for monitoring the wear of the friction plate of a wind turbine yaw brake, characterized in that: include: An acoustic wave receiver (1), a signal processor (2), a signal analyzer (3) and an indicator light type fault display (4); The signal processor (2) and the signal analyzer (3) are both arranged in the same housing (5); the number of the sound wave receivers (1) is set to be 2-6; the sound wave receivers (1) are evenly fixed to the outside of the housing (5) by a fixing bracket (6) and are 10 cm-30 cm away from the housing (5); the indicator light type fault display (4) is arranged outside the housing (5); the indicator light type fault display (4) is provided with an indicator light (7); the number of the indicator lights (7) is consistent with the number of yaw brakes; the indicator lights (7) correspond to the yaw brakes.

2. A method for monitoring the wear of a friction plate of a wind turbine yaw brake, characterized in that: The following steps are involved: Step 1: The acoustic wave receiver (1) receives the acoustic signal generated when each yaw brake is actuated and transmits the acoustic signal to the signal processor (2); Step 2: The signal processor (2) performs filtering and noise reduction processing on the collected sound signal; Step 3: There is a time delay between the sound signals emitted by the same yaw vibrator and received by the sound wave receiver (1). The time delay value can be obtained through the functional relationship between the sound signals. According to the Wiener-Khinchin theorem and Fourier transform, the generalized cross-correlation function between the two sound wave signals can be obtained. A rectangular coordinate system is established according to the generalized cross-correlation function, and the position coordinates of the sound signal can be obtained. Step 4: The signal processor (2) extracts the spectrum of the sound signal through Fourier transform and forms a spectrum diagram, and determines the main frequency of the sound wave and the sound pressure at the main frequency according to the spectrum diagram; Step 5: The signal analyzer (3) obtains the sound position, main frequency and sound pressure value outputted from the signal processor. The signal analyzer (3) can determine the deviation between the main frequency and sound pressure value of each yaw brake and the average value of the main frequency and sound pressure value of the remaining yaw brakes through analysis, and the severity of the fault can be determined by the size of the deviation value; Step 6: According to the severity of the fault of each yaw brake, the corresponding indicator light (7) displays a corresponding color.

3. The method for monitoring the wear of the friction plate of the wind turbine yaw brake according to claim 2, characterized in that: In step 2, the calculation formula for filtering and noise reduction performed by the signal processor (2) is: where ω i is the weight coefficient, e(n) is the output signal error, X i (n) is the i-th sound signal collected, i = 1, 2, ..., N, and the weight coefficient ω is used to minimize the error signal. i The iteration formula is: W(n+1)=W(n)-μ*e(n)*X(n) Where μ is the convergence factor, e(n) is the output signal error, and X(n) is the input signal.

4. The method for monitoring the wear of the friction plate of the wind turbine yaw brake according to claim 2, characterized in that: In step 3, the specific method of determining the position coordinates of the sound signal is: Step 3.1: Determine the functional relationship between the two sound signals: Among them, x1(t) and x2(t) are the sound signals received by the two sound wave receivers, S1 and S2 are the amplitude parameters of the transmitted signals, n1(t) and n2(t) are noise signals, τ1 and τ2 are the time delays of signal transmission, and τ1≤τ2; Step 3.2: According to the Wiener-Hinchin theorem and Fourier transform, the cross-correlation function between the two sound wave signals is as follows: Among them, G 12 (ω) is the cross power spectrum function of x1(t) and x2(t), e jωt represents a rotation factor, j is an imaginary unit, ω is the angular frequency, and t is time; when x1(t) and x2(t) are filtered, the cross-power spectrum function of the output signal is: Among them, H1 and H2 represent filter functions, H2 * (ω) represents the conjugate function of H2(ω), G 12 (ω) is the cross power spectrum function of x1(t) and x2(t); Step 3.3: R 12 (τ) and G g12 (ω) The generalized cross-correlation function of x1(t) and x2(t) is obtained by Fourier transform: Among them, G 12 (ω) is the cross power spectrum function of x1(t) and x2(t), e jωt represents a rotation factor, j is the imaginary unit, ω is the angular frequency, t is the time, φ 12 (ω) is the weighting function in the frequency domain; Step 3.4: Establishing a rectangular coordinate system according to the generalized cross-correlation function: the acoustic wave receiver (1) divides the rectangular coordinate system plane into equal areas; Step 3.5: The sampling period difference of the sound wave signals received by each two sound wave receivers (1) can be obtained by cross-correlation calculation, and the specific position of the sound signal in the rectangular coordinate system can be determined by the positive and negative values ​​of the sampling period difference; Step 3.6: Determine the position of the sound signal: The distance between the sound signal and the sound wave receiver (1) is obtained through the sampling period difference and the sampling frequency, so that the hyperbolic trajectory of the sound signal in the rectangular coordinate system can be determined. According to the direction of the sound signal and the coordinates of the intersection between the hyperbolas, the position coordinates of the sound signal relative to the sound wave receiver (1) can be obtained.

5. The method for monitoring the wear of the friction plate of the wind turbine yaw brake according to claim 2, characterized in that: In step 4, the calculation steps of the sound pressure and spectrum signal are: Step 4.1: Perform Fourier transform on the sound wave signal. The specific formula is as follows: Where f(t) is the time domain signal, f(ω) is the frequency domain signal, ω is the angular frequency, and j is the imaginary unit; Step 4.2: According to the formula in step 4.1, perform spectrum extraction on the sound signal. The formula after extraction is as follows: Among them, ω n is the angular frequency, a n and b n is the Fourier coefficient, a n and b n It can be obtained by integration: By using the above formula, the sound pressure corresponding to different frequencies is plotted in the coordinate system to form a spectrum diagram; Step 4.3: According to step 4.2, determine the main frequency of the sound signal, that is, the frequency corresponding to the maximum sound pressure value is the main frequency; Step 4.4: Through the above method, the main frequency and the sound pressure at the main frequency of all yaw brakes can be obtained, which are denoted as f and n and p n ; Step 4.5: For all f n and p n Conduct summary analysis.

6. The method for monitoring the wear of the friction plate of the wind turbine yaw brake according to claim 2, characterized in that: In step 5, the method for determining the deviation between the main frequency and sound pressure value of each yaw brake and the other yaw brakes is as follows: When a yaw brake is working, when the deviation between the main frequency and the average value of the main frequencies of other yaw brakes is less than 30Hz and the sound pressure generated is less than 10% of the average sound pressure of other yaw brakes, it can be judged that the life of the friction plate of the yaw brake is in a qualified state; When a yaw brake is working, when the main frequency of the yaw brake deviates from the average value of the main frequencies of other yaw brakes by 30Hz-50Hz and the sound pressure generated is 10%-20% higher than the average sound pressure of other yaw brakes, it can be judged that the life of the yaw brake friction plate is about to end; When a yaw brake is working, if the deviation between its main frequency and the average value of the main frequencies of other yaw brakes is greater than 50 Hz and the sound pressure generated is greater than the average sound pressure of the remaining yaw brakes by more than 20%, it can be determined that the life of the yaw brake friction plate has ended.

7. The method for monitoring the wear of the friction plate of the wind turbine yaw brake according to claim 2, characterized in that: In step 6, the color of the indicator light (7) is displayed as follows: When the signal analyzer (3) determines that the service life of a yaw brake friction plate is within the qualified range, the indicator light (7) displays a green light; when the signal analyzer (3) determines that the service life of a yaw brake friction plate is about to end, the indicator light (7) displays a yellow light; when the signal analyzer (3) determines that the service life of a yaw brake friction plate has ended, the indicator light (7) displays a red light.

8. The method for monitoring the wear of the friction plate of the wind turbine yaw brake according to claim 5, characterized in that: In step 4.1, the calculation formula of the sound pressure and the spectrum signal may also adopt the Mel-frequency cepstral coefficient method.

Images