CN110554090A - Wind turbine generator and crack monitoring system and method of variable-pitch bearing of wind turbine generator - Google Patents

Abstract

The invention provides a crack monitoring system and method for a wind turbine generator and a variable-pitch bearing thereof, wherein the crack monitoring system comprises: a plurality of acoustic emission sensors configured to acquire acoustic emission signals of corresponding pitch bearings; the controller is installed in the hub and configured to control each acoustic emission sensor to acquire the acoustic emission signal of the corresponding variable pitch bearing, acquire the acoustic emission signal of the corresponding variable pitch bearing from each acoustic emission sensor, and determine whether each variable pitch bearing has cracks according to the acoustic emission signals of all the variable pitch bearings. According to the wind turbine generator and the crack monitoring system of the pitch bearing of the wind turbine generator, online crack monitoring of the pitch bearing can be achieved through an acoustic emission technology, the pitch bearing is not affected by a shielding object, reliability is high, crack faults can be found in the crack initiation stage, compared with monitoring technologies such as manual troubleshooting and video troubleshooting, cracks can be found earlier, and planned maintenance can be achieved.

Description

Crack monitoring system and method for wind turbine and its pitch bearing

technical field

The invention relates to the field of wind power generation, and more specifically, to a crack monitoring system and method for a wind turbine and its pitch bearing.

Background technique

The working environment of pitch bearings of wind turbines is mostly harsh environments such as wind and sand, rain, humidity and low temperature, and various failures, such as fractures, are prone to occur. And because the replacement of the pitch bearing is very inconvenient, it is very important to monitor the crack of the pitch bearing. Existing methods of monitoring pitch bearings for cracks suffer from various deficiencies. For example, Fig. 1 shows a schematic structural diagram of an existing video monitoring system. The method of monitoring cracks through video requires setting up two imaging systems 101 to scan and shoot high-risk areas where cracks are prone to occur in pitch bearings. The imaging systems 101 are fixed On the hub, the imaging system 101 completes the monitoring of the high-risk areas where the pitch bearings are prone to cracks by virtue of the wind turbine retracting action. Fig. 2 shows a block diagram of an existing video surveillance system. As shown in Figure 2, the imaging system 101 of the three pitch bearings sends the captured images to the processor 202 in the hub measurement and control cabinet 201, and the processor 202 analyzes the crack faults of each pitch bearing according to the captured images, and sends The analysis results are sent to the fan PLC 203 . However, the video monitoring area is often covered by oil pollution, and on wind turbines with stiffening rings, the pitch bearing cannot be directly monitored, so the reliability of the monitoring results is not high. In addition, the existing monitoring technologies such as video, tin foil, and conductive paint can only be found after the crack has extended to the surface of the pitch bearing, which cannot achieve planned maintenance.

Contents of the invention

The purpose of the present invention is to provide a crack monitoring system and method for wind turbines and pitch bearings thereof, so as to solve the problems of low reliability and inability to realize planned maintenance in the existing crack monitoring methods.

One aspect of the present invention provides a crack monitoring system for a pitch bearing of a wind turbine. The crack monitoring system includes: a plurality of acoustic emission sensors, wherein at least one acoustic emission sensor is provided corresponding to each pitch bearing, and the at least one The acoustic emission sensor is installed on the hub close to the position of the corresponding pitch bearing, and is configured to collect the acoustic emission signal of the corresponding pitch bearing; the controller, installed in the hub, is configured to control each acoustic emission sensor Acoustic emission signals of corresponding pitch bearings are collected, acoustic emission signals of corresponding pitch bearings are acquired from each acoustic emission sensor, and whether cracks appear in each pitch bearing is determined according to the acoustic emission signals of all pitch bearings.

Optionally, one acoustic emission sensor is provided corresponding to each pitch bearing, and the one acoustic emission sensor is installed on the hub close to the position of the 0 scale of the pitch bearing when the blade is in the retracted state.

Optionally, the controller is further configured to: for a cracked faulty pitch bearing, during the process of collecting acoustic emission signals, when the acoustic emission signal with the strongest energy appears at the first moment, the acoustic emission sensor is The position on the pitch bearing of the fault is determined as the position of crack initiation.

Optionally, the controller is further configured to: for the faulty pitch bearing with cracks, according to the initial position, rotation speed and the first moment of the faulty pitch bearing before retraction, determine that the At the first moment, the position on the faulty pitch bearing facing the acoustic emission sensor.

Optionally, the controller is configured to: perform Fourier transform on the acoustic emission signals of each pitch bearing, and filter the Fourier transformed acoustic emission signals to obtain acoustic emission signals within a predetermined frequency range, Calculate the comparison data of each pitch bearing respectively, compare the comparison data of each pitch bearing with a predetermined multiple of the larger value in the comparison data of the other two pitch bearings, and compare the comparison data greater than the larger value A predetermined multiple of pitch bearings is determined as a faulty pitch bearing with cracks, wherein the comparison data of each pitch bearing is a sum of squares of amplitudes of acoustic emission signals within the predetermined frequency range of each pitch bearing.

Optionally, the controller is further configured to: for a faulty pitch bearing, divide the comparison data of the faulty pitch bearing by the larger value of the comparison data of the other two pitch bearings to obtain a quotient, and The size of the extension region of the crack is determined from the quotient.

Optionally, three acoustic emission sensors are provided corresponding to each pitch bearing, wherein one acoustic emission sensor is installed on the hub close to the 0 scale position of the pitch bearing when the blade is in the retracted state, and the other two acoustic emission sensors respectively installed on both sides of the one acoustic emission sensor.

Optionally, each of the acoustic emission sensors stores acoustic emission signals whose amplitudes are greater than a threshold value among the collected acoustic emission signals, and the controller obtains the stored acoustic emission signals from each acoustic emission sensor.

Optionally, the threshold value is the peak value of the amplitude of the acoustic emission signal of the pitch bearing when the wind turbine has not been put into use, or is the variable value when multiple wind turbines of the same model have not been put into use. The mean value of the peak value of the amplitude of the acoustic emission signal of the propeller bearing.

Optionally, the controller is specifically configured to control each acoustic emission sensor to collect the acoustic emission signal of the corresponding pitch bearing when the following trigger conditions are met: the controller detects the pitch retraction signal and the distance from the last crack monitoring exceeds The first predetermined time, or the second predetermined time from the last crack monitoring.

Another aspect of the present invention provides a wind turbine, which includes the crack monitoring system for a pitch bearing as described above.

Another aspect of the present invention provides a crack monitoring method for a pitch bearing of a wind turbine. The crack monitoring method includes: controlling each acoustic emission sensor to collect the acoustic emission signal of the corresponding pitch bearing, wherein, corresponding to each pitch The bearing is provided with at least one acoustic emission sensor, and the at least one acoustic emission sensor is installed on the hub at a position close to the corresponding pitch bearing; the acoustic emission signal of the corresponding pitch bearing is obtained from each acoustic emission sensor; according to all pitch bearings The acoustic emission signal of each pitch bearing to determine whether there is a crack.

Optionally, it also includes: for the faulty pitch bearing with cracks, during the process of collecting the acoustic emission signal, at the first moment when the acoustic emission signal with the strongest energy appears, the faulty pitch bearing facing the acoustic emission sensor The location on the bearing, identified as the location of crack initiation.

Optionally, it also includes: for the faulty pitch bearing with cracks, according to the initial position, rotation speed and the first moment of the faulty pitch bearing before retracting the pitch, determine that at the first moment, the acoustic The location on the failed pitch bearing where the transmitter sensor is facing.

Optionally, the step of determining whether cracks appear in each pitch bearing according to the acoustic emission signals of all pitch bearings includes: performing Fourier transform on the acoustic emission signals of each pitch bearing; performing filtering to obtain acoustic emission signals within a predetermined frequency range; calculating comparative data for each pitch bearing separately; comparing the comparative data for each pitch bearing with a predetermined multiple of a larger value among the comparative data for the other two pitch bearings making a comparison, and determining the pitch bearing whose comparison data is greater than a predetermined multiple of the larger value as a faulty pitch bearing with cracks, wherein the comparison data of each pitch bearing is within the predetermined frequency range of each pitch bearing The sum of the squares of the amplitudes of the acoustic emission signal.

Optionally, it also includes: for the faulty pitch bearing, dividing the comparison data of the faulty pitch bearing by the larger value of the comparison data of the other two pitch bearings to obtain a quotient, and calculating Determine the size of the crack growth region.

Optionally, the step of controlling each acoustic emission sensor to collect the acoustic emission signal of the corresponding pitch bearing includes: when the following trigger conditions are met, controlling each acoustic emission sensor to collect the corresponding acoustic emission signal of the pitch bearing: Paddle signal and the distance from the last crack monitoring exceeds the first predetermined time, or the distance from the last crack detection exceeds the second predetermined time.

Another aspect of the present invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to perform the crack monitoring method as described above.

According to the crack monitoring system and method of wind turbines and their pitch bearings, online monitoring of pitch bearing cracks can be realized through acoustic emission technology, which is not affected by obstructions, has high reliability, and crack faults can be found in the crack initiation stage, compared to Monitoring technologies such as manual inspection and video inspection can detect cracks earlier and realize planned maintenance.

Additional aspects and/or advantages of the present invention will be set forth in part in the following description, and some will be clear from the description, or can be learned through practice of the present invention.

Description of drawings

The above-mentioned and other objects, features and advantages of the present invention will become more clear through the following detailed description in conjunction with the accompanying drawings, wherein:

Fig. 1 is a schematic structural view showing an existing video monitoring system;

Fig. 2 is a block diagram showing an existing video surveillance system;

3 is a block diagram showing a crack monitoring system for a pitch bearing of a wind turbine according to an embodiment of the present invention;

4 is a diagram illustrating an installation position of an acoustic emission sensor according to an embodiment of the present invention;

FIG. 5 is a diagram illustrating an installation position of an acoustic emission sensor according to another embodiment of the present invention;

Figure 6A is a distribution diagram showing valid data for a faulty pitch bearing according to an embodiment of the present invention;

Fig. 6B is a distribution diagram showing valid data of a normal pitch bearing according to an embodiment of the present invention;

Fig. 7A is a schematic diagram showing an original collected acoustic emission signal of a faulty pitch bearing according to an embodiment of the present invention;

Fig. 7B is a schematic diagram showing an original collected acoustic emission signal of a normal pitch bearing according to an embodiment of the present invention;

Fig. 8A is a schematic diagram showing the Fourier-transformed acoustic emission signal of a faulty pitch bearing according to an embodiment of the present invention;

Fig. 8B is a schematic diagram showing the Fourier-transformed acoustic emission signal of a normal pitch bearing according to an embodiment of the present invention;

Fig. 9 is a flowchart illustrating a crack monitoring method of a pitch bearing of a wind turbine according to an embodiment of the present invention.

Detailed ways

Embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

Fig. 3 is a block diagram showing a crack monitoring system of a pitch bearing of a wind turbine according to an embodiment of the present invention.

As shown in FIG. 3 , the crack monitoring system for a pitch bearing of a wind turbine according to an embodiment of the present invention includes a plurality of acoustic emission sensors 301 and a controller 302 .

Those skilled in the art know that a wind turbine generally includes three blades, and each blade corresponds to a pitch bearing, that is, a wind turbine generally includes three pitch bearings. At least one acoustic emission sensor 301 is provided corresponding to each pitch bearing.

The at least one acoustic emission sensor 301 is installed on the hub at a position close to the corresponding pitch bearing, and is configured to collect an acoustic emission signal of the corresponding pitch bearing. The acoustic emission sensor 301 may be various devices capable of collecting acoustic emission signals. For example, the acoustic emission sensor 301 may be an R6A acoustic emission probe. The data line of the acoustic emission sensor 301 is arranged along the hub, enters the hub, and is connected with the measurement and control cabinet in the hub.

In a preferred embodiment, in order to comprehensively collect acoustic emission signals in high-risk areas where cracks are likely to occur in the pitch bearing and reduce monitoring costs, an acoustic emission sensor 301 is provided corresponding to each pitch bearing.

FIG. 4 is a diagram illustrating an installation position of an acoustic emission sensor according to an embodiment of the present invention. As shown in FIG. 4 , the acoustic emission sensor 301 is installed on the hub close to the 0 scale position of the pitch bearing when the blade is in the retracted state. In this way, in the process of blade retraction, the pitch bearing will rotate clockwise, and the trailing edge region where cracks are prone to occur will pass through the position where the acoustic emission sensor 301 is installed in turn, so that the acoustic emission sensor 301 can be installed to facilitate Acoustic emission signals of the trailing edge area where cracks are prone to occur in the pitch bearing can be collected.

Controller 302 is mounted within the hub. For example, the controller 302 is installed in a measurement and control cabinet inside the hub. A power supply, a wireless module and a lightning protection module can also be installed in the measurement and control cabinet.

The controller 302 is configured to, when the blade is in the state of pitch retraction, start the crack monitoring, that is, control each acoustic emission sensor 301 to collect the acoustic emission signal of the corresponding pitch bearing, and obtain the corresponding pitch pitch from each acoustic emission sensor 301. Acoustic emission signals of the bearings, and based on the acoustic emission signals of all pitch bearings, it is determined whether each pitch bearing is cracked.

In a preferred embodiment, in order to reduce unnecessary monitoring, the controller 302 is configured to start crack monitoring when a trigger condition is met.

The trigger condition is used to start each acoustic emission sensor 301 , that is, to control each acoustic emission sensor 301 to start collecting the acoustic emission signal of the corresponding pitch bearing.

The trigger condition includes that the controller 302 detects the pitch retraction signal and the first predetermined time has elapsed since the last crack detection. The controller 302 detects the pitch retraction signal, indicating that the pitch retraction operation is about to be performed. The first predetermined time is the shortest interval of crack monitoring, which can be set according to actual conditions, for example, it can be one day. When the controller 302 detects that the pitch retraction operation is about to be performed, and the last crack detection exceeds the minimum interval time, the controller 302 starts the crack detection, so that the crack detection and the pitch retraction operation are performed simultaneously.

For another example, the trigger condition may also include that the time since the last crack monitoring exceeds a second predetermined time. The second predetermined time is the longest interval of crack monitoring, which can be set according to the actual situation, for example, it can be three days. When the controller 302 detects that the last crack monitoring exceeds the longest interval time, the controller 302 starts the crack monitoring and starts the propeller retracting operation, so that the crack monitoring and the propeller retracting operation are performed simultaneously.

In addition, in order to ensure the accuracy of crack monitoring, it is necessary to stop the wind turbine before starting crack monitoring.

When the trigger condition is satisfied and the blades are retracting, the controller 302 controls each acoustic emission sensor 301 to start collecting the acoustic emission signal of the corresponding pitch bearing, and after the pitch retraction operation is completed, controls each acoustic emission sensor 301 to end Acoustic emission signals of corresponding pitch bearings are collected.

The acoustic emission sensor 301 stores the collected acoustic emission signals.

In a preferred embodiment, in order to reduce the amount of stored data of the acoustic emission sensor 301, each acoustic emission sensor 301 uses the acoustic emission signal whose amplitude is greater than the threshold value in the acoustic emission signal collected as valid data, and stores it. The acoustic emission signal whose amplitude is smaller than the threshold value is regarded as invalid data and is not stored.

The threshold value is the peak value of the amplitude of the acoustic emission signal of the pitch bearing without cracks.

Optionally, the peak value of the amplitude of the acoustic emission signal of the pitch bearing of the wind turbine may be measured when the wind turbine has not been put into use, and the peak value may be used as the threshold value.

Or, measure the peak value of the amplitude of the acoustic emission signal of the pitch bearing when multiple wind turbines of the same type have not been put into use, and use the average value of the measured multiple peak values as the threshold value.

In addition, the threshold value can also be updated in a machine learning manner according to the historical data of the acoustic emission signal of the normal pitch bearing. For example, to obtain the normal pitch bearings of multiple wind turbines of the same type that have been put into use, the peak value of the amplitude of the acoustic emission signal in each previous crack monitoring, and the average value of the obtained multiple peak values as the current threshold value.

The controller 302 acquires the corresponding acoustic emission signal of the pitch bearing from each acoustic emission sensor 301 . That is to say, the controller 302 acquires the stored acoustic emission signals from each acoustic emission sensor 301 .

The controller 302 determines whether a crack occurs in each pitch bearing according to all acquired acoustic emission signals (that is, acoustic emission signals of all pitch bearings). That is, for any pitch bearing, the controller 302 determines whether any pitch bearing has a crack according to the acoustic emission signal of any pitch bearing and the acoustic emission signals of the other two pitch bearings.

Here, in order to make the monitoring result more accurate, the controller 302 may firstly perform preprocessing on all acoustic emission signals. The preprocessing may include: performing Fourier transform on the acoustic emission signal, and filtering the Fourier transformed acoustic emission signal to obtain the acoustic emission signal within a predetermined frequency range.

The predetermined frequency range refers to the frequency range where the acoustic emission signal of the material used to make the pitch bearing is located.

The predetermined frequency range may be determined empirically, for example, may be determined as 100 kHz to 150 kHz empirically.

In addition, the predetermined frequency range may also be updated in a machine learning manner according to historical data of acoustic emission signals of faulty pitch bearings. For example, to obtain the faulty pitch bearings of multiple wind turbines of the same type that have been put into use, the frequency of the acoustic emission signal with the strongest energy in each previous crack monitoring, the average value of the obtained multiple frequencies is used as the current The center frequency of the predetermined frequency range.

The bandwidth of the current predetermined frequency range can be determined empirically, for example, it can be determined empirically as 50 kHz. In addition, the bandwidth can also be updated by machine learning based on historical data of acoustic emission signals of faulty pitch bearings. For example, to calculate the faulty pitch bearings of multiple wind turbines of the same model that have been put into use, the bandwidth of the predetermined frequency range in each previous crack monitoring will be the largest bandwidth among the bandwidths of the multiple predetermined frequency ranges calculated as the bandwidth of the current predetermined frequency range. The bandwidth of the predetermined frequency range in any previous crack monitoring can be determined in the following manner: in any previous crack monitoring, the frequency of the acoustic emission signal with the strongest energy of the faulty pitch bearing is used as the frequency of the crack monitoring center frequency, the spectrum envelope centered on the center frequency covers an area of a predetermined ratio (for example 80%), and the bandwidth of the corresponding frequency range is used as the bandwidth of the predetermined frequency in any crack monitoring.

The controller 302 respectively calculates the comparison data of each pitch bearing, and the comparison data of each pitch bearing is the sum of the squares of the amplitudes of the acoustic emission signals of each bearing within a predetermined frequency range.

The controller 302 compares the comparison data of each pitch bearing with a predetermined multiple of a larger value among the comparison data of the other two pitch bearings, and determines that the pitch bearing whose comparison data is greater than a predetermined multiple of the larger value is present Cracked faulty pitch bearing.

The predetermined multiple can be determined empirically, for example, it can be 10 times. That is, the controller 302 compares the comparison data of each pitch bearing with 10 times the larger value of the comparison data of the other two pitch bearings, and determines the pitch bearing whose comparison data is greater than 10 times the larger value For the faulty pitch bearing with cracks, the pitch bearing whose comparative data is less than or equal to 10 times the larger value is determined as a normal pitch bearing without cracks.

The controller 302 can send the crack monitoring result to the main controller of the wind turbine. The main controller can prompt the user to repair or maintain the faulty pitch bearing according to the crack monitoring results. For example, the controller 302 may send the crack monitoring result by sending a predetermined data form, and different data represent different crack monitoring results. For example, 000 means no fault, 100 means that No. 1 pitch bearing has a crack, 010 means that No. 2 pitch bearing has a crack, and 001 means that No. 3 pitch bearing has a crack.

The time when the pitch bearing is first detected to have a crack is the time of crack initiation.

In a preferred embodiment, the controller 302 can also determine the crack propagation of the faulty pitch bearing according to all the acoustic emission signals. The expansion situation may include expansion area and expansion speed. The user can carry out planned maintenance on the faulty pitch bearing according to the expansion of the crack of the faulty pitch bearing.

For the expansion area, the crack of the pitch bearing generally expands in the axial direction of the pitch bearing. The controller 302 divides the comparison data of the faulty pitch bearing by the larger value of the comparison data of the other two pitch bearings to obtain a quotient, and determines the size of the crack extension area according to the quotient. Here, the corresponding relationship between the quotient value and the size of the extension region of the crack can be set in advance. This correspondence can be determined empirically. Then, according to the quotient value and the corresponding relationship between the quotient value and the size of the crack extension area, the size of the crack extension area of the faulty pitch bearing is determined.

Regarding the growth speed, the controller 302 may determine the crack growth speed according to the energy change speed of the acoustic emission signal with the strongest energy in multiple crack monitorings of the faulty pitch bearing. The faster the energy change rate, the faster the crack growth rate.

In a preferred embodiment, the controller 302 can also determine the location of crack initiation according to the acoustic emission signal of the faulty pitch bearing. The controller 302 also aims at the faulty pitch bearing with cracks, and in the process of collecting the acoustic emission signal of the faulty pitch bearing, at the first moment when the acoustic emission signal with the strongest energy appears, the faulty pitch bearing facing the acoustic emission sensor will The location on the bearing was determined as the location of crack initiation in the faulty pitch bearing. That is to say, the position on the faulty pitch bearing facing the acoustic emission sensor at the first moment is determined as the crack initiation position in the faulty pitch bearing. In this way, the user can accurately maintain or repair the faulty pitch bearing according to the location of the crack initiation.

Specifically, for the faulty pitch bearing with cracks, the controller 302 determines that the acoustic emission sensor is normal at the first moment according to the initial position of the faulty pitch bearing before pitch retraction, the rotation speed of the faulty pitch bearing, and the first moment. on the location of the faulty pitch bearing. The initial position can be expressed as the position of the faulty pitch bearing on which the acoustic emission sensor is facing before the pitch retraction, and the initial position can be determined according to the pitch angle of the faulty pitch pitch bearing before pitch retraction. Here, the time from the initial moment to the first moment indicates the rotation time it takes for the faulty pitch bearing to rotate from the initial position to the crack initiation position, the rotation distance can be obtained by multiplying the rotation time by the rotation speed, and the rotation distance The distance plus the initial position can be obtained. At the first moment, the position on the faulty pitch bearing facing the acoustic emission sensor.

In a preferred embodiment, in order to more accurately locate the location of crack initiation, three acoustic emission sensors 301 are arranged corresponding to each pitch bearing, that is, when the above-mentioned installed on the hub is close to the blade in the pitch retracting state In addition to the one acoustic emission sensor 301 at the position of the 0 scale of the bearing, two other acoustic emission sensors 301 are provided.

FIG. 5 is a diagram illustrating an installation position of an acoustic emission sensor according to another embodiment of the present invention. As shown in FIG. 5 , the other two acoustic emission sensors 301 are respectively arranged on both sides of the one acoustic emission sensor 301 . The other two acoustic emission sensors 301 are used to further determine the location of crack initiation. For example, at the first moment, the energy intensity of the acoustic emission signals collected by the other two acoustic emission sensors 301 can be compared, and the side where the acoustic emission sensor 301 of the collected acoustic emission signal with stronger energy intensity is located is determined as a crack The location of the germination.

The above mainly describes the crack monitoring method for the trailing edge region where cracks are likely to occur in the pitch bearing. It can be understood that other regions of the pitch bearing can also be monitored for cracks by installing acoustic emission sensors at other positions. For example, an acoustic emission sensor is installed on the hub close to the position of the pitch bearing 180 scale when the blade is in the retracted state to monitor cracks in the leading edge area of the pitch bearing. Except for the inconsistent installation position of the acoustic emission sensor, other The monitoring method is the same as the above-mentioned crack monitoring method in the trailing edge region, and will not be repeated here.

FIGS. 6A to 8B respectively show relevant information of acoustic emission signals of a faulty pitch bearing and a normal pitch bearing according to an embodiment of the present invention.

6A and 6B respectively show distribution diagrams of valid data of a faulty pitch bearing and a normal pitch bearing according to an embodiment of the present invention. As shown in FIG. 6A , the acoustic emission signal of the faulty pitch bearing includes a plurality of valid data 601 . As shown in FIG. 6B , there is no valid data in the AE signal of a normal pitch bearing.

FIG. 7A and FIG. 7B respectively show schematic diagrams of original collected acoustic emission signals of a faulty pitch bearing and a normal pitch bearing according to an embodiment of the present invention. As shown in FIG. 7A , the amplitude of some acoustic emission signals of the faulty pitch bearing is relatively high. As shown in Fig. 7B, the amplitudes of all acoustic emission signals of a normal pitch bearing are within the normal range.

FIG. 8A and FIG. 8B respectively show schematic diagrams of acoustic emission signals after Fourier transform of a faulty pitch bearing and a normal pitch bearing according to an embodiment of the present invention. As shown in FIG. 8A , the amplitude of the acoustic emission signal with a frequency between 100 kHz and 150 kHz after Fourier transform of the faulty pitch bearing is relatively high. As shown in FIG. 8B , the amplitude of the acoustic emission signal with a frequency between 100 kHz and 150 kHz after Fourier transform of a normal pitch bearing is within a normal range.

Fig. 9 is a flowchart illustrating a crack monitoring method of a pitch bearing of a wind turbine according to an embodiment of the present invention.

Those skilled in the art know that a wind turbine generally includes three blades, and each blade corresponds to a pitch bearing, that is, a wind turbine generally includes three pitch bearings. In the crack monitoring method of the pitch bearing of the wind turbine set according to the embodiment of the present invention, at least one acoustic emission sensor is provided corresponding to each pitch bearing.

The at least one acoustic emission sensor is installed on the hub at a position close to the corresponding pitch bearing, and is configured to collect an acoustic emission signal of the corresponding pitch bearing. Acoustic emission sensors can be various devices that can collect acoustic emission signals. For example, the acoustic emission sensor may be an R6A acoustic emission probe. The data line of the acoustic emission sensor is arranged along the hub, enters the hub, and is connected with the measurement and control cabinet inside the hub.

In a preferred embodiment, in order to comprehensively collect the acoustic emission signals of the high-risk areas where the pitch bearings are prone to cracks and reduce the monitoring cost, an acoustic emission sensor is set corresponding to each pitch bearing, and the specific installation position can be Refer to Figure 4.

Referring to FIG. 9 , in step S901 , when the blade is in the state of retracting the pitch, crack monitoring is started, that is, each acoustic emission sensor is controlled to collect the acoustic emission signal of the corresponding pitch bearing.

In step S902, acoustic emission signals of corresponding pitch bearings are acquired from each acoustic emission sensor.

In step S903, it is determined whether each pitch bearing has a crack according to the acoustic emission signals of all the pitch bearings.

In a preferred embodiment, in step S901, crack monitoring is started when the trigger condition is met.

The trigger condition is used to start each acoustic emission sensor, that is, to control each acoustic emission sensor to start collecting the acoustic emission signal of the corresponding pitch bearing.

The trigger condition includes detecting the pitch retraction signal and the first predetermined time has passed since the last crack monitoring. The detection of the retraction signal indicates that the retraction operation is imminent. The first predetermined time is the shortest interval of crack monitoring, which can be set according to actual conditions, for example, it can be one day. When it is detected that the propeller retraction operation is about to be performed, and the last crack monitoring exceeds the minimum interval time, the crack monitoring is started, so that the crack monitoring and the propeller retraction operation are performed simultaneously.

For another example, the trigger condition may also include that the time since the last crack monitoring exceeds a second predetermined time. The second predetermined time is the longest interval of crack monitoring, which can be set according to the actual situation, for example, it can be three days. When the last crack monitoring on the detection distance exceeds the longest interval time, the crack monitoring is started, and the propeller retracting operation is started, so that the crack monitoring and the propeller retracting operation are performed simultaneously.

In addition, in order to ensure the accuracy of crack monitoring, it is necessary to stop the wind turbine before starting crack monitoring.

When the trigger condition is met and the blades are retracting the pitch, each acoustic emission sensor is controlled to start collecting the acoustic emission signal of the corresponding pitch bearing, and after the pitch retraction operation is completed, each acoustic emission sensor is controlled to stop collecting the corresponding pitch Acoustic emission signal of a bearing.

The acoustic emission sensor stores the collected acoustic emission signals.

In a preferred embodiment, in order to reduce the amount of stored data of the acoustic emission sensor, each acoustic emission sensor takes the acoustic emission signal whose amplitude is greater than the threshold value in the collected acoustic emission signal as valid data, and stores it, and stores the acoustic emission signal whose amplitude is less than The acoustic emission signal of the threshold value is regarded as invalid data and is not stored.

The threshold value is the peak value of the amplitude of the acoustic emission signal of the pitch bearing without cracks.

Optionally, the peak value of the amplitude of the acoustic emission signal of the pitch bearing of the wind turbine may be measured when the wind turbine has not been put into use, and the peak value may be used as the threshold value.

Or, measure the peak value of the amplitude of the acoustic emission signal of the pitch bearing when multiple wind turbines of the same type have not been put into use, and use the average value of the measured multiple peak values as the threshold value.

In addition, the threshold value can also be updated in a machine learning manner according to the historical data of the acoustic emission signal of the normal pitch bearing. For example, to obtain the normal pitch bearings of multiple wind turbines of the same type that have been put into use, the peak value of the amplitude of the acoustic emission signal in each previous crack monitoring, and the average value of the obtained multiple peak values as the current threshold value.

In step S902, acoustic emission signals of corresponding pitch bearings are acquired from each acoustic emission sensor. That is, the stored acoustic emission signals are acquired from the respective acoustic emission sensors.

In step S903, it is determined whether each pitch bearing has a crack according to all acquired acoustic emission signals (ie, acoustic emission signals of all pitch bearings). That is, for any one of the pitch bearings, it is determined whether any one of the pitch bearings has a crack according to the acoustic emission signal of any one of the pitch bearings and the acoustic emission signals of the other two pitch bearings.

Here, in order to make the monitoring results more accurate, all acoustic emission signals can be preprocessed first. The preprocessing may include: performing Fourier transform on the acoustic emission signal, and filtering the Fourier transformed acoustic emission signal to obtain the acoustic emission signal within a predetermined frequency range.

The predetermined frequency range refers to the frequency range where the acoustic emission signal of the material used to make the pitch bearing is located.

The predetermined frequency range may be determined empirically, for example, may be determined as 100 kHz to 150 kHz empirically.

In addition, the predetermined frequency range may also be updated in a machine learning manner according to historical data of acoustic emission signals of faulty pitch bearings. For example, to obtain the faulty pitch bearings of multiple wind turbines of the same type that have been put into use, the frequency of the acoustic emission signal with the strongest energy in each previous crack monitoring, the average value of the obtained multiple frequencies is used as the current The center frequency of the predetermined frequency range.

The bandwidth of the current predetermined frequency range can be determined empirically, for example, it can be determined empirically as 50 kHz. In addition, the bandwidth can also be updated by machine learning based on historical data of acoustic emission signals of faulty pitch bearings. For example, to calculate the faulty pitch bearings of multiple wind turbines of the same model that have been put into use, the bandwidth of the predetermined frequency range in each previous crack monitoring will be the largest bandwidth among the bandwidths of the multiple predetermined frequency ranges calculated as the bandwidth of the current predetermined frequency range. The bandwidth of the predetermined frequency range in any previous crack monitoring can be determined in the following manner: in any previous crack monitoring, the frequency of the acoustic emission signal with the strongest energy of the faulty pitch bearing is used as the frequency of the crack monitoring center frequency, the spectrum envelope centered on the center frequency covers an area of a predetermined ratio (for example 80%), and the bandwidth of the corresponding frequency range is used as the bandwidth of the predetermined frequency in any crack monitoring.

In step S903, the comparison data of each pitch bearing is calculated respectively, and the comparison data of each pitch bearing is the sum of squares of amplitudes of acoustic emission signals within a predetermined frequency range of each bearing.

Comparing the comparison data of each pitch bearing with a predetermined multiple of the larger value among the comparison data of the other two pitch bearings, and determining the pitch bearing whose comparison data is greater than the predetermined multiple of the larger value as a cracked failure Pitch bearings.

The predetermined multiple can be determined empirically, for example, it can be 10 times. That is, the comparison data of each pitch bearing is compared with 10 times the larger value among the comparison data of the other two pitch bearings, and the pitch bearing whose comparison data is greater than 10 times the larger value is determined to be cracked If the pitch bearing is faulty, the pitch bearing whose comparative data is less than or equal to 10 times the larger value is determined to be a normal pitch bearing without cracks.

The crack monitoring results can be sent to the main controller of the wind turbine. The main controller can prompt the user to repair or maintain the faulty pitch bearing according to the crack monitoring results. For example, the crack monitoring result can be sent by sending a predetermined data form, and different data represent different crack monitoring results. For example, 000 means no fault, 100 means that No. 1 pitch bearing has a crack, 010 means that No. 2 pitch bearing has a crack, and 001 means that No. 3 pitch bearing has a crack.

The time when the pitch bearing is first detected to have a crack is the time of crack initiation.

In a preferred embodiment, the crack monitoring method of the pitch bearing of the wind turbine according to the embodiment of the present invention may also include the following steps (not shown in the figure): determine the crack of the faulty pitch bearing according to all acoustic emission signals of the expansion. The expansion situation may include expansion area and expansion speed. The user can carry out planned maintenance on the faulty pitch bearing according to the expansion of the crack of the faulty pitch bearing.

For the expansion area, the crack of the pitch bearing generally expands in the axial direction of the pitch bearing. The comparative data of the faulty pitch bearing can be divided by the larger value of the comparative data of the other two pitch bearings to obtain a quotient, and the size of the crack propagation area can be determined according to the quotient. Here, the corresponding relationship between the quotient value and the size of the extension region of the crack can be set in advance. This correspondence can be determined empirically. Then, according to the quotient value and the corresponding relationship between the quotient value and the size of the crack extension area, the size of the crack extension area of the faulty pitch bearing is determined.

For the growth speed, the crack growth speed can be determined according to the energy change speed of the acoustic emission signal with the strongest energy in the multiple crack monitoring of the faulty pitch bearing. The faster the energy change rate, the faster the crack growth rate.

In a preferred embodiment, the crack monitoring method of the pitch bearing of the wind turbine according to the embodiment of the present invention may also include the following steps (not shown in the figure): determining crack initiation according to the acoustic emission signal of the faulty pitch bearing s position. For the faulty pitch bearing with cracks, in the process of collecting the acoustic emission signal of the faulty pitch bearing, at the first moment when the acoustic emission signal with the strongest energy appears, the acoustic emission sensor on the faulty pitch bearing facing the The location was determined as the location of crack initiation in the faulty pitch bearing. That is to say, the position on the faulty pitch bearing facing the acoustic emission sensor at the first moment is determined as the crack initiation position in the faulty pitch bearing. In this way, the user can accurately maintain or repair the faulty pitch bearing according to the location of the crack initiation.

Specifically, for the faulty pitch bearing with cracks, the initial position of the faulty pitch bearing before retraction, the rotation speed of the faulty pitch bearing, and the first moment can be used to determine the Location on faulty pitch bearing. The initial position can be expressed as the position of the faulty pitch bearing on which the acoustic emission sensor is facing before the pitch retraction, and the initial position can be determined according to the pitch angle of the faulty pitch pitch bearing before pitch retraction. Here, the time from the initial moment to the first moment indicates the rotation time it takes for the faulty pitch bearing to rotate from the initial position to the position where the crack initiates. The rotation distance can be obtained by multiplying the rotation time by the rotation speed. The rotation distance Adding the initial position can be obtained, at the first moment, the position on the faulty pitch bearing facing the acoustic emission sensor.

In a preferred embodiment, in order to locate the crack initiation position more accurately, three acoustic emission sensors are arranged corresponding to each pitch bearing, that is, when the above-mentioned pitch bearing is installed on the hub close to the blade in the retracted state In addition to the one acoustic emission sensor at the position of the 0 scale, there are two other acoustic emission sensors. Refer to Figure 5 for the specific installation positions.

The above mainly describes the crack monitoring method for the trailing edge region where cracks are likely to occur in the pitch bearing. It can be understood that other regions of the pitch bearing can also be monitored for cracks by installing acoustic emission sensors at other positions. For example, an acoustic emission sensor is installed on the hub close to the position of the pitch bearing 180 scale when the blade is in the retracted state to monitor cracks in the leading edge area of the pitch bearing. Except for the inconsistent installation position of the acoustic emission sensor, other The monitoring method is the same as the above-mentioned crack monitoring method in the trailing edge region, and will not be repeated here.

The present invention also provides a wind turbine, which includes the above crack monitoring system for pitch bearings.

The present invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, causes the processor to execute the crack monitoring method as described above.

According to the embodiment of the present invention, the crack monitoring system and method of the wind turbine and its pitch bearing can realize the online monitoring of the crack of the pitch bearing through the acoustic emission technology, which is not affected by obstructions, has high reliability, and can detect cracks in the crack initiation stage. Crack faults can be detected earlier than monitoring techniques such as manual inspection and video inspection, which can realize planned maintenance. In addition, it can avoid the labor cost, time cost and possible missed inspection caused by manual inspection, and avoid the long interval of manual inspection, which will lead to failure to find out in time and cause unit failure.

While the invention has been particularly shown and described with reference to exemplary embodiments thereof, it will be understood by those skilled in the art that changes may be made in form and detail without departing from the spirit and scope of the invention as defined by the claims. various changes.

Claims (18)

1. The utility model provides a crack monitoring system of wind turbine generator system's change oar bearing which characterized in that includes:

A plurality of acoustic emission sensors, wherein at least one acoustic emission sensor is provided corresponding to each pitch bearing, the at least one acoustic emission sensor being mounted on the hub in proximity to the corresponding pitch bearing and configured to acquire acoustic emission signals of the corresponding pitch bearing;

The controller is installed in the hub and configured to control each acoustic emission sensor to acquire the acoustic emission signal of the corresponding variable pitch bearing, acquire the acoustic emission signal of the corresponding variable pitch bearing from each acoustic emission sensor, and determine whether each variable pitch bearing has cracks according to the acoustic emission signals of all the variable pitch bearings.

2. The crack monitoring system of claim 1, wherein one acoustic emission sensor is provided for each pitch bearing, the one acoustic emission sensor being mounted on the hub close to the pitch bearing 0 scale when the blade is in the feathered state.

3. The crack monitoring system of claim 2, wherein the controller is further configured to: and aiming at the fault pitch bearing with the crack, determining the position on the fault pitch bearing, which is just opposite to the acoustic emission sensor, as the crack initiation position when the acoustic emission signal with the strongest energy occurs at the first moment in the process of acquiring the acoustic emission signal.

4. The crack monitoring system of claim 3, wherein the controller is further configured to: and aiming at the fault variable pitch bearing with the crack, determining the position on the fault variable pitch bearing, which is just opposite to the acoustic emission sensor, at the first moment according to the initial position, the rotating speed and the first moment of the fault variable pitch bearing before pitch withdrawing.

5. The crack monitoring system of claim 1, wherein the controller is configured to: fourier transform is carried out on the acoustic emission signals of each variable pitch bearing, the acoustic emission signals after Fourier transform are filtered to obtain acoustic emission signals in a preset frequency range, comparison data of each variable pitch bearing are respectively calculated, the comparison data of each variable pitch bearing are compared with preset multiples of a larger value in the comparison data of other two variable pitch bearings, the variable pitch bearing of which the comparison data is larger than the preset multiples of the larger value is determined as a fault variable pitch bearing with cracks,

Wherein the comparison data for each pitch bearing is the sum of the squares of the amplitudes of the acoustic emission signals within the predetermined frequency range for each pitch bearing.

6. The crack monitoring system of claim 5, wherein the controller is further configured to: and for the fault variable-pitch bearing, dividing the comparison data of the fault variable-pitch bearing by the larger value in the comparison data of the other two variable-pitch bearings to obtain a quotient, and determining the size of the expansion area of the crack according to the quotient.

7. The crack monitoring system of claim 1, wherein three acoustic emission sensors are provided for each pitch bearing, wherein one acoustic emission sensor is mounted on the hub close to the pitch bearing 0 scale when the blade is in the feathered state, and two acoustic emission sensors are mounted on either side of the acoustic emission sensor.

8. The crack monitoring system of claim 1, wherein each acoustic emission sensor stores an acoustic emission signal having an amplitude greater than a threshold value in the collected acoustic emission signals, and the controller obtains the stored acoustic emission signal from each acoustic emission sensor.

9. The crack monitoring system of claim 8, wherein the threshold value is a peak value of the amplitude of the acoustic emission signal of the pitch bearing when the wind turbine generator has not been put into use, or an average value of the peak values of the amplitudes of the acoustic emission signals of the pitch bearing when a plurality of wind turbine generators of the same type have not been put into use.

10. The crack monitoring system of claim 1, wherein the controller is configured to control each acoustic emission sensor to acquire an acoustic emission signal of the corresponding pitch bearing when the following triggering conditions are met: the controller detects a paddle retracting signal and the distance between the paddle retracting signal and the last crack monitoring exceeds a first preset time, or the distance between the paddle retracting signal and the last crack monitoring exceeds a second preset time.

11. Wind turbine comprising a crack monitoring system for a pitch bearing according to any of claims 1-10.

12. The crack monitoring method for the variable-pitch bearing of the wind turbine generator is characterized by comprising the following steps of:

Controlling each acoustic emission sensor to collect an acoustic emission signal of the corresponding pitch bearing, wherein at least one acoustic emission sensor is arranged corresponding to each pitch bearing, and the at least one acoustic emission sensor is arranged on the hub close to the corresponding pitch bearing;

Acquiring acoustic emission signals of corresponding variable pitch bearings from each acoustic emission sensor;

And determining whether each pitch bearing has cracks according to the acoustic emission signals of all the pitch bearings.

13. The crack monitoring method of claim 12, further comprising:

And aiming at the fault pitch bearing with the crack, determining the position on the fault pitch bearing, which is just opposite to the acoustic emission sensor, as the crack initiation position at the first moment when the acoustic emission signal with the strongest energy appears in the process of acquiring the acoustic emission signal.

14. the crack monitoring method of claim 13, further comprising:

And aiming at the fault variable pitch bearing with the crack, determining the position on the fault variable pitch bearing, which is just opposite to the acoustic emission sensor, at the first moment according to the initial position, the rotating speed and the first moment of the fault variable pitch bearing before pitch withdrawing.

15. The crack monitoring method of claim 12, wherein the step of determining whether a crack has occurred in each pitch bearing based on the acoustic emission signals of all pitch bearings comprises:

Fourier transformation is carried out on the acoustic emission signals of each variable pitch bearing;

Filtering the acoustic emission signals subjected to Fourier transform to obtain acoustic emission signals in a preset frequency range;

Respectively calculating comparison data of each variable pitch bearing;

Comparing the comparison data of each pitch bearing with a predetermined multiple of the greater value of the comparison data of the other two pitch bearings, determining the pitch bearing with the comparison data greater than the predetermined multiple of the greater value as a failed pitch bearing with cracks,

Wherein the comparison data for each pitch bearing is the sum of the squares of the amplitudes of the acoustic emission signals within the predetermined frequency range for each pitch bearing.

16. The crack monitoring method of claim 15, further comprising:

And for the fault variable-pitch bearing, dividing the comparison data of the fault variable-pitch bearing by the larger value in the comparison data of the other two variable-pitch bearings to obtain a quotient, and determining the size of the expansion area of the crack according to the quotient.

17. The crack monitoring method of claim 12, wherein the step of controlling each acoustic emission sensor to acquire an acoustic emission signal of a corresponding pitch bearing comprises: when the following triggering conditions are met, controlling each acoustic emission sensor to acquire the acoustic emission signal of the corresponding pitch bearing: and detecting a propeller retracting signal and exceeding a first preset time from the last crack monitoring, or exceeding a second preset time from the last crack monitoring.

18. a computer readable storage medium storing a computer program which, when executed by a processor, causes the processor to perform the crack monitoring method as claimed in any one of claims 12 to 17.