A method and system for monitoring the status of a fan transmission chain based on acoustic emission and vibration

Abstract

The present invention discloses a method for monitoring the status of a wind turbine transmission chain based on acoustic emission and vibration, comprising: selecting to collect only acoustic emission data or simultaneously collecting acoustic emission data and vibration data as monitoring data based on the operating status of the wind turbine; calculating characteristic indicators of the monitoring data in frequency bands, transmitting and saving the original waveform data and each characteristic indicator; calculating historical trend indicators of the monitoring data based on the characteristic indicators within each frequency band; integrating the characteristic indicators and historical trend indicators of the monitoring data as monitoring indicators, and determining the status of the transmission chain based on the monitoring indicator threshold; and also discloses a wind turbine transmission chain status monitoring system using this method. The present invention's condition monitoring method based on acoustic emission and vibration can overcome the disadvantages of a single vibration signal in diagnosing and warning low-speed and early-stage high-frequency bearing fault signals. It can effectively monitor high-frequency signals generated by early-stage defects and faults, improving the accuracy of condition warnings, and facilitating the timely and accurate detection of fault problems.

Description

Technical Field

[0001] The present invention relates to the technical field of wind power monitoring, and in particular to a method and system for monitoring the state of a wind turbine transmission chain based on acoustic emission and vibration. Background Art

[0002] During the twenty-year life cycle of a wind turbine, stable and reliable operation is the most direct guarantee of a wind farm's economic benefits. Passively performing unplanned repairs and replacements after a breakdown not only increases the cost of repairs and replacements, but also results in a loss of power generation. Furthermore, repeated overhauls of intact equipment can increase operating and maintenance costs, and in severe cases, can lead to excessive maintenance, shortening the equipment's service life. Industry surveys and statistics show that failures in the transmission chain, including those of wind turbine generators, gearboxes, and main bearings, result in the longest periods of downtime. Furthermore, the cost of repairing and replacing these transmission chain components is high, accounting for a significant portion of the wind turbine's power generation costs. Therefore, researching and developing condition monitoring technology for wind turbine transmission chain equipment and providing early warning of transmission chain equipment failures are key areas for cost reduction and efficiency improvement in the wind power industry. At present, the transmission chain status monitoring system mainly obtains the vibration signals of the equipment on the transmission chain of the wind turbine by installing vibration sensors at relevant positions of the wind turbine. Then, the operating status of the main equipment of the wind turbine is obtained through dedicated data acquisition and analysis, and early warning judgment is made on possible faults of the wind turbine. However, the vibration monitoring method has limited effect on low-speed equipment and non-full-cycle situations. In addition, the frequency detection range of the vibration sensor is generally in the low frequency band below 2000Hz, which easily misses the high-frequency band signals of early fault defects. Therefore, its fault warning accuracy is low.

[0003] The Chinese patent document "A Method and Apparatus for Detecting Faults in a Wind Turbine Drive Chain Without Speed ​​Measurement," published under the publication number CN114624023A and published on June 14, 2022, includes the following steps: obtaining a time-discretized vibration signal; extracting the instantaneous shaft velocity by processing the vibration signal, and performing time-domain integration of the instantaneous shaft velocity to obtain the instantaneous shaft phase; generating a tachometer based on the instantaneous shaft phase; performing synchronous phase angle discretization based on the tachometer information; and performing synchronous analysis based on the discrete shaft phase angle signals to detect component damage characteristics in the wind turbine drive chain. When the real-time shaft speed signal is missing, the instantaneous shaft velocity of the wind turbine drive chain's high-speed shaft is extracted from the vibration signal, and the vibration signal is synchronously re-collected and analyzed, improving the accuracy of fault detection in the wind turbine drive chain. However, this technology still relies solely on condition monitoring based on a single vibration signal, failing to overcome the limitations of vibration monitoring for low-speed equipment and in non-full-cycle scenarios. Furthermore, the limited monitoring frequency range results in low accuracy of condition warnings. Summary of the Invention

[0004] The present invention aims to overcome the problem in the prior art that when the status of a fan transmission chain is monitored only by vibration signals, the vibration monitoring has limited effect on low-speed equipment and non-full-cycle situations, and the monitoring frequency range is restricted, resulting in low accuracy of status warning. A fan transmission chain status monitoring method and system based on acoustic emission and vibration is provided. Through the detection of acoustic emission, the elastic waves released by deformation or fracture in the material are converted into electrical signals, and the status of the fan transmission chain is monitored together with the vibration signal, thereby overcoming the problem that a single vibration signal cannot effectively detect low-speed equipment and the detection frequency is limited. The method can effectively monitor the high-frequency band signals generated by early defect faults, improve the accuracy of status warnings, and facilitate timely and accurate detection of fault problems.

[0005] In order to achieve the above object, the present invention adopts the following technical solutions:

[0006] A method for monitoring the status of a fan transmission chain based on acoustic emission and vibration, comprising:

[0007] Depending on the operating status of the fan, choose to collect only acoustic emission data or both acoustic emission data and vibration data as monitoring data;

[0008] Calculate the characteristic indicators of the monitoring data in different frequency bands, and send and save the original waveform data and each characteristic indicator;

[0009] Calculate historical trend indicators based on characteristic indicators of monitoring data in each frequency band;

[0010] The characteristic indicators and historical trend indicators of the comprehensive monitoring data are used as monitoring indicators, and the transmission chain status is judged according to the monitoring indicator thresholds.

[0011] The collection of acoustic emission data in the present invention is similar to the collection of vibration data. It is a collection method that requires contact with the object to be measured. Acoustic emission detection uses an acoustic emission sensor coupled to the surface of a sample or structure. When deformation or fracture occurs in the material or structure, elastic waves are released inside. The elastic waves generated by the acoustic emission source in the material are converted into electrical signals, thereby inferring the defect state and severity inside the material. Compared with vibration detection, which fails to effectively monitor low-speed rotating equipment such as main bearings and the frequency detection range of vibration sensors is generally in the low frequency band below 2000Hz, it cannot effectively detect early faults of transmission chain equipment. Acoustic emission detection can compensate for the shortcomings of vibration detection, thereby increasing the sources of acquisition of status monitoring data and improving the accuracy of status warnings, so as to facilitate timely and accurate detection of fault problems.

[0012] Preferably, the wind turbine operating state includes a low-wind standby state and a normal power generation state, and only acoustic emission data is collected in the low-wind standby state;

[0013] In the normal power generation state, when the high-speed shaft speed is lower than the speed threshold, only the acoustic emission data is collected; when the high-speed shaft speed is greater than or equal to the speed threshold, the acoustic emission data and vibration data are collected simultaneously.

[0014] In the present invention, the operating status data of the wind turbine is obtained in real time to determine whether the wind turbine is in normal power generation state or low-wind standby state. Since vibration monitoring is limited in detecting low-speed equipment, acoustic emission monitoring is used for normal power generation state and low-wind standby state below the speed threshold, and a combination of the two is used for normal power generation state above the speed threshold.

[0015] Preferably, for the acoustic emission data, the characteristic indexes of the acoustic emission data are calculated respectively according to the three acoustic emission frequency bands of (0, f1], (f1, f3] and greater than f3; for the vibration data, the characteristic indexes of the vibration data are calculated respectively according to the three vibration frequency bands of (0, f1], (f1, f2] and (f2, f3].

[0016] In the present invention, different faults will occur in different frequency bands. The purpose of frequency band calculation is to determine in which frequency band the abnormality occurs, and then determine the specific fault type. The specific frequency band value can be given based on expert experience and adjusted later. f1 <f2<f3。

[0017] As a preference, when only acoustic emission data is collected, the effective value AE of the acoustic emission data in each acoustic emission frequency band is calculated respectively. RMS And the average ASL as the characteristic indicator:

[0018] When collecting acoustic emission data and vibration data at the same time, in addition to calculating the effective value AE of acoustic emission data in each acoustic emission frequency band, RMS In addition to the average value ASL as characteristic indicators, it is also necessary to calculate the effective value V of the vibration data in each vibration frequency band RMS and crest factor V F as a characteristic indicator.

[0019] In the present invention, when monitoring acoustic emission, the effective value and the average value are calculated in the three acoustic emission frequency bands as characteristic indicators, so six characteristic indicators can be obtained; when acoustic emission detection and vibration monitoring are used in combination, six characteristic indicators belonging to acoustic emission monitoring and six characteristic indicators belonging to vibration monitoring can be obtained.

[0020] Preferably, the process of calculating the historical trend index of the monitoring data in each frequency band according to the characteristic index is as follows: for the characteristic index in each frequency band, the rate of change of the characteristic index in the time interval Δt is used as the historical trend index in the frequency band.

[0021] In the present invention, the historical trend indicators are reflected by the rate of change of the characteristic indicators. Therefore, when acoustic emission monitoring is used, six historical trend indicators can be obtained; when acoustic emission detection and vibration monitoring are used in combination, six historical trend indicators belonging to acoustic emission monitoring and six historical trend indicators belonging to vibration monitoring can be obtained.

[0022] Preferably, for determining that a device in the fan transmission chain is in an abnormal state, any one of the following conditions needs to be met: among the monitoring indicators collected and calculated from the device, any monitoring indicator exceeds its monitoring indicator threshold, and the excess amplitude is greater than a preset amplitude, and the excess duration is greater than a first preset time; or

[0023] Among the monitoring indicators collected and calculated from the device, the proportion of the number of monitoring indicators exceeding the monitoring indicator threshold is greater than the preset proportion, and the duration of the exceeding limit is greater than the second preset time.

[0024] In the present invention, when acoustic emission monitoring is performed, the six characteristic indicators of acoustic emission and the six historical trend indicators are all used as monitoring indicators to perform status monitoring and judgment; when acoustic emission detection and vibration monitoring are used in combination, the six characteristic indicators and six historical trend indicators belonging to acoustic emission monitoring and the six characteristic indicators and six historical trend indicators belonging to vibration monitoring are all used as monitoring indicators to perform status monitoring and judgment; for the abnormal state judgment of a certain device in the fan transmission chain, the judgment is performed based on the monitoring indicators calculated based on the data collected by the acoustic emission sensor unit and the vibration sensor unit arranged on the device.

[0025] As a preference, the wind turbine drive train status assessment is divided into three levels: normal, warning, and alarm:

[0026] When any device in the fan transmission chain is in an abnormal state and the abnormality lasts longer than the preset abnormality time, it is a warning level; when two or more devices are in an abnormal state and the abnormality lasts longer than the preset abnormality time, it is an alarm level;

[0027] The fan transmission chains that have not reached the warning level or alarm level are judged to be at the normal level.

[0028] The present invention not only monitors the abnormal status of each individual device on the fan transmission chain, but also comprehensively analyzes the abnormal status of all devices to determine whether the operating status of the entire fan transmission chain is healthy, and takes different measures according to different health levels.

[0029] Preferably, for the original waveform data, the original waveform data will be sent and saved only when the waveform sending and saving rules are met, otherwise only the corresponding characteristic indicators will be sent and saved.

[0030] In the present invention, if all the original waveform data are sent to the server, the amount of data transmitted will increase greatly. Therefore, in order to reduce the total amount of data sent to the server and improve the efficiency of calculation and analysis, corresponding waveform sending and saving rules can be set, which can not only reduce the amount of data transmitted but also avoid the omission of important original waveform data.

[0031] A wind turbine transmission chain status monitoring system based on acoustic emission and vibration, comprising:

[0032] Several groups of data acquisition devices are arranged on the fan transmission chain equipment, each group of data acquisition devices includes a pair of acoustic emission sensor units and vibration sensor units;

[0033] The fan data real-time acquisition unit acquires the fan operation data in real time and determines the fan operation status;

[0034] The data collected by the data acquisition device and the fan data real-time acquisition unit are processed by the edge processing unit and sent to the server. The server receives and saves the data and then determines the status of the fan transmission chain.

[0035] In the present invention, the vibration sensor unit is used to obtain the vibration signal of the fan transmission chain equipment, and the acoustic emission sensor unit is used to obtain the acoustic emission signal of the fan transmission chain equipment. These sensors are respectively installed on the rotating parts of the main bearing, gearbox, and generator; the fan data real-time acquisition unit 3 can obtain the fan operation status data from the fan control system in real time, and the edge processing unit is used to perform a certain amount of calculation and data processing and execute relevant acquisition strategies, and then transmit the processed data to the server. The server processes and analyzes the data and comprehensively analyzes multiple indicators to judge the operation status of the fan transmission chain.

[0036] Preferably, the acoustic emission signal acquired by the acoustic emission sensor unit is collected by the acoustic emission data acquisition unit and transmitted to the edge processing unit, and the vibration signal acquired by the vibration sensor unit is collected by the vibration data acquisition unit and transmitted to the edge processing unit; the wind turbine data real-time acquisition unit, the acoustic emission data acquisition unit, the vibration data acquisition unit and the edge processing unit are built into an integrated device.

[0037] In the present invention, the calculations in the condition monitoring method of the present invention must be performed for each sensor unit, that is, the data collected by each sensor unit will be calculated to generate six characteristic indicators and six historical trend indicators, and then the condition judgment is made based on these monitoring indicators; in addition, some units near the fan transmission chain are integrated into an integrated device, which can facilitate the installation and maintenance of the condition monitoring system.

[0038] The present invention has the following beneficial effects: through the detection of acoustic emission, the elastic waves released by deformation or fracture in the material are converted into electrical signals, and combined with the vibration signal to monitor the status of the wind turbine transmission chain, thereby overcoming the problem that a single vibration signal cannot effectively detect low-speed equipment and the detection frequency is limited, and can effectively monitor the high-frequency band signals generated by early defect faults, improve the accuracy of status warnings, so as to timely and accurately detect fault problems; can quickly and accurately detect early failure problems of transmission chain equipment, realize the transformation from passive maintenance after the fault occurs to planned maintenance before the fault occurs, effectively reduce the failure rate of equipment, and improve the operating reliability of the unit, thereby achieving the purpose of reducing wind power operation and maintenance costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a flow chart of the method for monitoring the status of the fan transmission chain in the present invention;

[0040] Figure 2 is a schematic diagram of a fan transmission chain status monitoring system in the present invention;

[0041] Figure 3 is a schematic diagram of a data acquisition device on a fan transmission chain according to an embodiment of the present invention;

[0042] In the figure: 1. Vibration sensor unit; 2. Acoustic emission sensor unit; 3. Real-time acquisition unit for wind turbine data; 4. Vibration data acquisition unit; 5. Acoustic emission data acquisition unit; 6. Edge processing unit; 7. Switch unit; 8. Server; 81. Data storage unit; 82. Data analysis and early warning unit; 11. First acoustic emission sensor; 12. First vibration sensor; 13. Second acoustic emission sensor; 14. Second vibration sensor; 15. Third acoustic emission sensor; 16. Third vibration sensor; 17. Fourth acoustic emission sensor; 18. Fourth vibration sensor; 19. Fifth acoustic emission sensor; 20. Fifth vibration sensor; 21. Sixth acoustic emission sensor; 22. Sixth vibration sensor; 31. Integrated device. DETAILED DESCRIPTION

[0043] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0044] like Figure 1 As shown, a method for monitoring the status of a wind turbine transmission chain based on acoustic emission and vibration includes:

[0045] Depending on the operating status of the fan, you can choose to collect only acoustic emission data or collect both acoustic emission data and vibration data as monitoring data.

[0046] The wind turbine operating status includes low wind standby state and normal power generation state:

[0047] 1) When the wind speed condition meets the start-up operating conditions and the operating status is normal, the wind turbine is judged to be in normal power generation state, otherwise the wind turbine is in fault or maintenance state;

[0048] 2) When the wind speed conditions do not meet the start-up conditions and the unit is not in a fault or maintenance state, the fan is determined to be in low wind standby state.

[0049] In the low-wind standby state, only acoustic emission data is collected; in the normal power generation state, when the high-speed shaft speed is lower than the speed threshold, only acoustic emission data is collected; when the high-speed shaft speed is greater than or equal to the speed threshold, acoustic emission data and vibration data are collected at the same time.

[0050] Calculate the characteristic index of the monitoring data by frequency band, and send and save the original waveform data and each characteristic index.

[0051] For acoustic emission data, the characteristic indices of acoustic emission data are calculated according to the three acoustic emission frequency bands (0, f1], (f1, f3] and greater than f3; for vibration data, the characteristic indices of vibration data are calculated according to the three vibration frequency bands (0, f1], (f1, f2] and (f2, f3].

[0052] When only acoustic emission data is collected, the effective value AE of the acoustic emission data in each acoustic emission frequency band is calculated separately. RMS and mean ASL as characteristic indicators;

[0053] When collecting acoustic emission data and vibration data at the same time, in addition to calculating the effective value AE of acoustic emission data in each acoustic emission frequency band, RMS In addition to the average value ASL as characteristic indicators, it is also necessary to calculate the effective value V of the vibration data in each vibration frequency band RMS and crest factor V F as a characteristic indicator.

[0054] For the original waveform data, the original waveform data will be sent and saved only when the waveform sending and saving rules are met, otherwise only the corresponding characteristic indicators will be sent and saved.

[0055] The rules for sending and saving waveforms are as follows:

[0056] 1) Waveform data whose characteristic index calculated at the current moment exceeds any set alarm threshold;

[0057] 2) The characteristic index calculated at the current moment has increased by more than 20% compared to the average index of the previous 72 hours;

[0058] 3) The interval between the current time and the last time waveform data was sent is greater than 6 hours.

[0059] The historical trend indicator is calculated based on the characteristic indicator of the monitoring data in each frequency band. The process of calculating the historical trend indicator based on the characteristic indicator of the monitoring data in each frequency band is as follows: for the characteristic indicator in each frequency band, the rate of change of the characteristic indicator within the time interval Δt is used as the historical trend indicator in the frequency band.

[0060] The characteristic indicators and historical trend indicators of the integrated monitoring data are used as monitoring indicators, and the transmission chain status is judged according to the monitoring indicator thresholds. When the corresponding equipment status of the transmission chain is abnormal, a corresponding maintenance plan is formulated. The monitoring indicator thresholds and abnormality judgment conditions can be continuously optimized and improved based on the results of subsequent early warning feedback.

[0061] To determine whether a device in the fan transmission chain is in an abnormal state, any of the following conditions must be met:

[0062] Any monitoring indicator among the monitoring indicators collected and calculated from the device exceeds its monitoring indicator threshold, and the excess is greater than the preset extent, and the duration of the excess is greater than the first preset time; or

[0063] Among the monitoring indicators collected and calculated from the device, the proportion of the number of monitoring indicators exceeding the monitoring indicator threshold is greater than the preset proportion, and the duration of the exceeding limit is greater than the second preset time.

[0064] The status assessment of the fan drive chain is divided into three levels: normal, warning and alarm:

[0065] When any device in the fan transmission chain is in an abnormal state and the abnormality lasts longer than the preset abnormality time, it is a warning level; when two or more devices are in an abnormal state and the abnormality lasts longer than the preset abnormality time, it is an alarm level;

[0066] The fan transmission chains that have not reached the warning level or alarm level are judged to be at the normal level.

[0067] The collection of acoustic emission data in the present invention is similar to the collection of vibration data. It is a collection method that requires contact with the object to be measured. Acoustic emission detection uses an acoustic emission sensor coupled to the surface of a sample or structure. When deformation or fracture occurs in the material or structure, elastic waves are released inside. The elastic waves generated by the acoustic emission source in the material are converted into electrical signals, thereby inferring the defect state and severity inside the material. Compared with vibration detection, which fails to effectively monitor low-speed rotating equipment such as main bearings and the frequency detection range of vibration sensors is generally in the low frequency band below 2000Hz, thus failing to effectively detect early faults of transmission chain equipment, acoustic emission detection can compensate for the shortcomings of vibration detection, thereby improving the accuracy of status warnings, so as to facilitate timely and accurate detection of fault problems.

[0068] The present invention utilizes a transmission chain equipment status monitoring method based on acoustic emission and vibration. This method overcomes the shortcomings of a single vibration signal in diagnosing and warning low-speed and early-stage high-frequency bearing fault signals. The present invention employs different signal acquisition strategies for the different operating states of wind turbines, enabling comprehensive, real-time monitoring of the health status of wind turbine transmission chain equipment. Furthermore, based on acoustic emission and vibration signals, the present invention establishes multiple characteristics sensitive to the health status of transmission chain equipment and their historical trend monitoring indicators. This method can capture key information about early-stage equipment failures and better characterize the severity of the failures.

[0069] In the present invention, the operating status data of the wind turbine is obtained in real time to determine whether the wind turbine is in normal power generation state or low-wind standby state. Since vibration monitoring is limited in detecting low-speed equipment, acoustic emission monitoring is used for normal power generation state and low-wind standby state below the speed threshold, and a combination of the two is used for normal power generation state above the speed threshold.

[0070] In the present invention, different faults will occur in different frequency bands. The purpose of frequency band calculation is to determine in which frequency band the abnormality occurs, and then determine the specific fault type. The specific frequency band value can be given based on expert experience and adjusted later. f1 <f2<f3。

[0071] In the present invention, when monitoring acoustic emission, the effective value and the average value are calculated in the three acoustic emission frequency bands as characteristic indicators, so six characteristic indicators can be obtained; when acoustic emission detection and vibration monitoring are used in combination, six characteristic indicators belonging to acoustic emission monitoring and six characteristic indicators belonging to vibration monitoring can be obtained.

[0072] In the present invention, the historical trend indicators are reflected by the rate of change of the characteristic indicators. Therefore, when acoustic emission monitoring is used, six historical trend indicators can be obtained; when acoustic emission detection and vibration monitoring are used in combination, six historical trend indicators belonging to acoustic emission monitoring and six historical trend indicators belonging to vibration monitoring can be obtained.

[0073] In the present invention, when acoustic emission monitoring is performed, the six characteristic indicators of acoustic emission and the six historical trend indicators are all used as monitoring indicators to perform status monitoring and judgment; when acoustic emission detection and vibration monitoring are used in combination, the six characteristic indicators and six historical trend indicators belonging to acoustic emission monitoring and the six characteristic indicators and six historical trend indicators belonging to vibration monitoring are all used as monitoring indicators to perform status monitoring and judgment; for the abnormal state judgment of a certain device in the fan transmission chain, the judgment is performed based on the monitoring indicators calculated based on the data collected by the acoustic emission sensor unit and the vibration sensor unit arranged on the device.

[0074] The present invention not only monitors the abnormal status of each individual device on the fan transmission chain, but also comprehensively analyzes the abnormal status of all devices to determine whether the operating status of the entire fan transmission chain is healthy, and takes different measures according to different health levels.

[0075] In the present invention, if all the original waveform data are sent to the server, the amount of data transmitted will increase greatly. Therefore, in order to reduce the total amount of data sent to the server and improve the efficiency of calculation and analysis, corresponding waveform sending and saving rules can be set, which can not only reduce the amount of data transmitted but also avoid the omission of important original waveform data.

[0076] like Figure 2 As shown, a wind turbine transmission chain status monitoring system based on acoustic emission and vibration includes:

[0077] Several groups of data acquisition devices are arranged on the fan transmission chain equipment, each group of data acquisition devices includes a pair of acoustic emission sensor units 2 and vibration sensor units 1;

[0078] The fan data real-time acquisition unit 3 acquires the fan operation data in real time and determines the fan operation status;

[0079] The data collected by the data acquisition device and the fan data real-time acquisition unit 3 are processed by the edge processing unit 6 and sent to the server 8 through the switch unit 7. The server 8 receives and saves the data and then determines the status of the fan transmission chain.

[0080] The acoustic emission signal acquired by the acoustic emission sensor unit 2 is collected by the acoustic emission data acquisition unit 5 and transmitted to the edge processing unit 6, and the vibration signal acquired by the vibration sensor unit 1 is collected by the vibration data acquisition unit 4 and transmitted to the edge processing unit 6; the wind turbine data real-time acquisition unit 3, the acoustic emission data acquisition unit 5, the vibration data acquisition unit 4 and the edge processing unit 6 are built into an integrated device 31.

[0081] In the present invention, the vibration sensor unit is used to obtain the vibration signal of the fan transmission chain equipment, and the acoustic emission sensor unit is used to obtain the acoustic emission signal of the fan transmission chain equipment. These sensors are respectively installed on the rotating parts of the main bearing, gearbox, and generator; the fan data real-time acquisition unit 3 can obtain the fan operation status data from the fan control system in real time, and the edge processing unit is used to perform a certain amount of calculation and data processing and execute relevant acquisition strategies, and then transmit the processed data to the server. The server processes and analyzes the data and comprehensively analyzes multiple indicators to judge the operation status of the fan transmission chain.

[0082] In the present invention, the calculations in the condition monitoring method of the present invention must be performed for each sensor unit, that is, the data collected by each sensor unit will be calculated to generate six characteristic indicators and six historical trend indicators, and then the condition judgment is made based on these monitoring indicators; in addition, some units near the fan transmission chain are integrated into an integrated device, which can facilitate the installation and maintenance of the condition monitoring system.

[0083] In the embodiment of the present invention, Figure 3 The figure shows the locations of acoustic emission and vibration sensors for a doubly-fed wind turbine generator. Due to the high surface temperature of the transmission chain equipment, this embodiment uses a high-temperature broadband acoustic emission sensor. For the vibration sensors, the first vibration sensor 12 and the second vibration sensor 14 are highly sensitive piezoelectric accelerometers. These sensors are mounted on the rotating components of the main bearing, gearbox, and generator, respectively, to collect acoustic emission and vibration signals from key components of the transmission chain. One data acquisition device (including a first vibration sensor 12 and a first acoustic emission sensor 11) is installed on the main bearing; three data acquisition devices (including a second vibration sensor 14 and a second acoustic emission sensor 13, a third vibration sensor 16 and a third acoustic emission sensor 15, and a fourth vibration sensor 18 and a fourth acoustic emission sensor 17) are installed on the gearbox; and two data acquisition devices (including a fifth vibration sensor 20 and a fifth acoustic emission sensor 19, and a sixth vibration sensor 22 and a sixth acoustic emission sensor 21) are installed on the generator. The sensor locations and number shown in this embodiment represent a preferred solution; sensors can also be placed in the axial, radial, horizontal, and radial vertical positions of the rotating components. The wind turbine real-time data acquisition unit 3, vibration data acquisition unit 4, acoustic emission data acquisition unit 5, and edge processing unit 6 are built into the integrated device 31. The wind turbine real-time data acquisition unit 3 is used to obtain real-time unit data such as wind turbine operating status, power generation, wind speed, high-speed shaft speed, and rotor speed from the SCADA wind turbine control system. The vibration data acquisition unit 4 is responsible for collecting vibration data and performing preliminary processing. The acoustic emission data acquisition unit 5 is responsible for collecting acoustic emission data and performing preliminary processing. The edge processing unit 6 is responsible for performing a certain amount of calculation and data processing, executing relevant acquisition strategies, and then transmitting the processed data to the server. The switch unit 7 is responsible for transmitting data to the server via the wind farm fiber optic ring network. The data storage unit 81 of the server 8 is responsible for receiving and storing data. The data analysis and early warning unit 82 of the server 8 is responsible for processing and analyzing the data, analyzing multiple indicators to determine the operating status of the wind turbine drive train, and outputting early warning results.

[0084] Example 1: Obtaining the wind turbine's operating status, power generation, wind speed, high-speed shaft speed, and rotor speed and other operating parameters in real time from the wind turbine SCADA system; using the SCADA data to determine the wind turbine's operating status:

[0085] When the wind speed condition does not meet the start-up conditions and the unit is not in a fault or maintenance state, the fan is determined to be in low wind standby state, and only acoustic emission data is collected at this time;

[0086] Or when the wind speed conditions meet the start-up conditions and the operating state is normal, the wind turbine is judged to be in a normal power generation state. At the same time, when the measured speed of the high-speed shaft is lower than the speed threshold of 1000RPM, only acoustic emission data is collected.

[0087] Calculate the effective value AE of the acoustic emission data collected by each acoustic emission sensor RMS And the average ASL as the characteristic indicator:

[0088]

[0089]

[0090] Where V is the acoustic emission signal, t0 is the initial sampling time, T is the sampling duration, and N is the number of data points, i.e., the signal length.

[0091] The characteristic index of acoustic emission data is calculated according to the acoustic emission frequency band (0, 10 Hz], (10 Hz, 5 kHz] and the frequency band greater than 5 kHz to obtain the effective value AE of the three frequency bands. RMS10 AE RMS10-5k and AE RMS>5k and the average ASL of the three frequency bands 10 、ASL 10-5k and ASL >5k There are six characteristic indicators in total.

[0092] Then, in each frequency band, the corresponding historical trend indicators are reflected according to the changes in the slope of the curve of the characteristic indicators of acoustic emission:

[0093]

[0094]

[0095] Where T0 represents the current time, and Δt represents the time interval. In this embodiment, a 14-day monitoring period is used. This yields six historical trend indicators, corresponding to the effective value and the average value, for the three frequency bands of acoustic emission. The six characteristic indicators and six historical trend indicators calculated from the acoustic emission data are used as monitoring indicators to determine the operating status.

[0096] For the first acoustic emission sensor 11 provided on the main bearing of the transmission chain, when the monitoring index of its signal meets the following conditions, it is determined that the state of the main bearing of the transmission chain is abnormal:

[0097] 1) A monitoring indicator exceeds its monitoring indicator threshold, and the excess is greater than 50%, and the duration of the excess is greater than 48 hours; 2) The proportion of the number of monitoring indicators exceeding the limit is greater than 50%, and the duration is greater than 72 hours.

[0098] For the second acoustic emission sensor 13, the third acoustic emission sensor 15, and the fourth acoustic emission sensor 17 provided on the transmission chain gearbox, when the monitoring index of any sensor signal meets the following conditions, it is determined that the transmission chain gearbox part is abnormal:

[0099] 1) A monitoring indicator exceeds its monitoring indicator threshold, and the excess is greater than 30%, and the duration of the excess is greater than 24 hours; 2) The proportion of monitoring indicators exceeding the limit is greater than 20%, and the duration is greater than 48 hours.

[0100] For the fifth acoustic emission sensor 19 and the sixth acoustic emission sensor 21 provided on the transmission chain generator, when the monitoring index of any sensor signal meets the following conditions, it is determined that the state of the transmission chain generator is abnormal:

[0101] 1) A monitoring indicator exceeds the monitoring indicator threshold, and the excess margin is greater than 40%, and the duration of the excess is greater than 24 hours;

[0102] 2) The number of monitoring indicators exceeding the limit accounts for more than 30% and the duration is greater than 48 hours.

[0103] The severity of the transmission chain health status is mainly divided into three levels: normal, warning, and alarm. The rules for determining the severity of each health status are as follows:

[0104] 1) Warning: If any of the transmission chain main bearing, gearbox, or generator parts is abnormal and the abnormality lasts for more than 24 hours, the operation and maintenance personnel are required to perform maintenance on the equipment within one week;

[0105] 2) Alarm: If two or more parts in the transmission chain are abnormal and the abnormality lasts for more than 24 hours, the unit needs to be shut down for maintenance immediately;

[0106] 3) Normal: A transmission chain that has not reached the warning or alarm level is judged to be in normal state and can continue to be observed and operated.

[0107] Example 2: Obtaining the wind turbine's operating status, power generation, wind speed, high-speed shaft speed, and rotor speed and other operating parameters in real time from the wind turbine SCADA system; using the SCADA data to determine the wind turbine's operating status:

[0108] When the wind speed conditions meet the start-up operating conditions and the operating status is normal, the wind turbine is determined to be in a normal power generation state. At the same time, when the measured speed of the high-speed shaft is greater than or equal to the speed threshold of 1000RPM, acoustic emission data and vibration data are collected at the same time.

[0109] Calculate the effective value AE of the acoustic emission data collected by each acoustic emission sensor RMS And the average ASL as the characteristic indicator:

[0110]

[0111]

[0112] Where V is the acoustic emission signal, t0 is the initial sampling time, T is the sampling duration, and N is the number of data points, i.e., the signal length.

[0113] At the same time, calculate the effective value V of the vibration data collected by each vibration sensor RMS and crest factor V F As a characteristic indicator:

[0114]

[0115]

[0116] The vibration signal is X(x1,x2,…,x M ), M is the number of data points, i.e., the signal length.

[0117] The characteristic index of acoustic emission data is calculated according to the acoustic emission frequency band (0, 10 Hz], (10 Hz, 5 kHz] and the frequency band greater than 5 kHz to obtain the effective value AE of the three frequency bands. RMS10 AE RMS10-5k and AE RMS>5k and the average ASL of the three frequency bands 10 、ASL 10-5k and ASL >5k There are six characteristic indicators of acoustic emission. At the same time, the characteristic indicators of vibration data are calculated according to the vibration frequency band (0, 10Hz], (10Hz, 2kHz] and greater than (2kHz, 5kHz], and the effective value V of the three frequency bands is obtained. RMS10 、V RMS10-2k and V RMS2k-5k And the crest factor V of the three frequency bands F10 、V F10-2k and V F2k-5k There are six vibration characteristic indicators, a total of twelve characteristic indicators.

[0118] Then, in each frequency band, the corresponding historical trend indicators are reflected according to the changes in the slope of the curve of the characteristic indicators of acoustic emission:

[0119]

[0120]

[0121] The corresponding historical trend indicators are reflected according to the changes in the slope of the curve of the vibration characteristic indicators:

[0122]

[0123]

[0124] Where T0 represents the current time, and Δt represents the time interval. In this embodiment, a 14-day monitoring period is used. This yields historical trend indicators for the effective value and average value within the three frequency bands of acoustic emission, as well as historical trend indicators for the effective value and crest factor within the three frequency bands of vibration, for a total of twelve historical trend indicators. The six characteristic indicators and six historical trend indicators calculated from the acoustic emission data, as well as the six characteristic indicators and six historical trend indicators calculated from the vibration data, are used as monitoring indicators to determine the operating status.

[0125] For the first acoustic emission sensor 11 and the first vibration sensor 12 provided on the main bearing of the transmission chain, when the monitoring index of any sensor signal meets the following conditions, it is determined that the state of the main bearing of the transmission chain is abnormal:

[0126] 1) A monitoring indicator exceeds its monitoring indicator threshold, and the excess is greater than 50%, and the duration of the excess is greater than 48 hours; 2) The proportion of the number of monitoring indicators exceeding the limit is greater than 50%, and the duration is greater than 72 hours.

[0127] For the second acoustic emission sensor 13, the second vibration sensor 14, the third acoustic emission sensor 15, the third vibration sensor 16, the fourth acoustic emission sensor 17, and the fourth vibration sensor 18 provided on the transmission chain gearbox, when the monitoring indicators of any sensor signal meet the following conditions, it is determined that the state of the transmission chain gearbox is abnormal:

[0128] 1) A monitoring indicator exceeds its monitoring indicator threshold, and the excess is greater than 30%, and the duration of the excess is greater than 24 hours; 2) The proportion of monitoring indicators exceeding the limit is greater than 20%, and the duration is greater than 48 hours.

[0129] For the fifth acoustic emission sensor 19, the fifth vibration sensor 20, the sixth acoustic emission sensor 21, and the sixth vibration sensor 22 provided on the transmission chain generator, when the monitoring indicators of any sensor signal meet the following conditions, it is determined that the state of the transmission chain generator part is abnormal:

[0130] 1) A monitoring indicator exceeds the monitoring indicator threshold, and the excess margin is greater than 40%, and the duration of the excess is greater than 24 hours;

[0131] 2) The number of monitoring indicators exceeding the limit accounts for more than 30% and the duration is greater than 48 hours.

[0132] The severity of the transmission chain health status is mainly divided into three levels: normal, warning, and alarm. The rules for determining the severity of each health status are as follows:

[0133] 1) Warning: If any of the transmission chain main bearing, gearbox, or generator parts is abnormal and the abnormality lasts for more than 24 hours, the operation and maintenance personnel are required to perform maintenance on the equipment within one week;

[0134] 2) Alarm: If two or more parts in the transmission chain are abnormal and the abnormality lasts for more than 24 hours, the unit needs to be shut down for maintenance immediately;

[0135] 3) Normal: A transmission chain that has not reached the warning or alarm level is judged to be in normal state and can continue to be observed and operated.

[0136] The above embodiments are further elaborations and illustrations of the present invention for ease of understanding, and are not intended to limit the present invention in any way. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for monitoring the status of a fan transmission chain based on acoustic emission and vibration, characterized in that: include: Depending on the operating status of the fan, choose to collect only acoustic emission data or both acoustic emission data and vibration data as monitoring data; In the low wind standby state, only acoustic emission data is collected; In normal power generation, when the high-speed shaft speed is lower than the speed threshold, only acoustic emission data is collected; when the high-speed shaft speed is greater than or equal to the speed threshold, both acoustic emission data and vibration data are collected; Calculate the characteristic indicators of the monitoring data in different frequency bands, and send and save the original waveform data and each characteristic indicator; In each frequency band, the historical trend index is calculated based on the characteristic index of the monitoring data; for the characteristic index in each frequency band, the historical trend index is calculated based on the characteristic index in the time interval. The rate of change within the frequency band is used as an indicator of the historical trend within the frequency band; The characteristic indicators and historical trend indicators of the comprehensive monitoring data are used as monitoring indicators, and the transmission chain status is judged according to the monitoring indicator thresholds.

2. A method for monitoring the status of a wind turbine transmission chain based on acoustic emission and vibration according to claim 1, characterized in that: The wind turbine operating state includes a low wind standby state and a normal power generation state; When the wind speed condition meets the start-up operating conditions and the operating state is normal, the wind turbine is determined to be in a normal power generation state; When the wind speed conditions do not meet the start-up conditions and the unit is not in a fault or maintenance state, the fan is determined to be in low wind standby state.

3. A method for monitoring the status of a wind turbine transmission chain based on acoustic emission and vibration according to claim 1 or 2, characterized in that: For acoustic emission data, according to 、 and greater than The characteristic indexes of acoustic emission data are calculated for each of the three acoustic emission frequency bands; for vibration data, the characteristic indexes are calculated according to 、 and The characteristic indices of vibration data are calculated for each of the three vibration frequency bands.

4. A method for monitoring the status of a wind turbine transmission chain based on acoustic emission and vibration according to claim 3, characterized in that: When only acoustic emission data is collected, the effective value of the acoustic emission data in each acoustic emission frequency band is calculated separately. and average As a characteristic indicator; When collecting acoustic emission data and vibration data at the same time, in addition to calculating the effective value of acoustic emission data in each acoustic emission frequency band, and average In addition to the characteristic index, the effective value of the vibration data in each vibration frequency band needs to be calculated and crest factor as a characteristic indicator.

5. A method for monitoring the status of a wind turbine transmission chain based on acoustic emission and vibration according to claim 1, 2 or 4, characterized in that: To determine whether a device in the fan transmission chain is in an abnormal state, any of the following conditions must be met: Any monitoring indicator collected and calculated from the device exceeds its monitoring indicator threshold, and the magnitude of the excess exceeds the preset magnitude, and the duration of the excess exceeds the first preset time; or Among the monitoring indicators collected and calculated from the device, the proportion of the number of monitoring indicators exceeding the monitoring indicator threshold is greater than the preset proportion, and the duration of the exceeding limit is greater than the second preset time.

6. The method for monitoring the status of a wind turbine transmission chain based on acoustic emission and vibration according to claim 5, characterized in that: The status assessment of the fan drive chain is divided into three levels: normal, warning and alarm: When any device in the fan transmission chain is in an abnormal state and the abnormality lasts longer than the preset abnormality time, it is a warning level; When two or more devices are in abnormal state and the abnormality lasts longer than the preset abnormality time, it is an alarm level; The fan transmission chains that have not reached the warning level or alarm level are judged to be at the normal level.

7. A method for monitoring the status of a wind turbine transmission chain based on acoustic emission and vibration according to claim 1 or 4, characterized in that: For the original waveform data, the original waveform data will be sent and saved only when the waveform sending and saving rules are met, otherwise only the corresponding characteristic indicators will be sent and saved.

8. A wind turbine transmission chain status monitoring system based on acoustic emission and vibration, applicable to the method according to any one of claims 1 to 7, characterized in that: include: Several groups of data acquisition devices are arranged on the fan transmission chain equipment, each group of data acquisition devices includes a pair of acoustic emission sensor units and vibration sensor units; The fan data real-time acquisition unit acquires the fan operation data in real time and determines the fan operation status; The data collected by the data acquisition device and the fan data real-time acquisition unit are processed by the edge processing unit and sent to the server. The server receives and saves the data and then determines the status of the fan transmission chain.

9. The wind turbine transmission chain status monitoring system based on acoustic emission and vibration according to claim 8, characterized in that: The acoustic emission signal acquired by the acoustic emission sensor unit is collected by the acoustic emission data acquisition unit and transmitted to the edge processing unit, and the vibration signal acquired by the vibration sensor unit is collected by the vibration data acquisition unit and transmitted to the edge processing unit; The wind turbine data real-time acquisition unit, acoustic emission data acquisition unit, vibration data acquisition unit and edge processing unit are built into an integrated device.

Images