Method and system for identifying a fault in a drive train of a wind turbine based on an acoustic signal

Abstract

This application provides a method and system for fault identification of wind turbine drivetrain based on acoustic signals. The method includes: setting up multiple sound acquisition devices inside the equipment and labeling each sound acquisition device; establishing a three-dimensional data model based on the equipment's shape and connecting each sound acquisition device to its corresponding position in the data model according to its label; comparing and verifying the target sound collected by adjacent sound acquisition devices to confirm the fault point; adjusting the verification of the fault sound according to different environments by monitoring the temperature of the working environment and changing the corresponding comparison data; adjusting the verification data of the fault sound by considering the metal fatigue wear during mechanical transmission and changing the verification conditions; and issuing an early warning based on the fault information. This application significantly improves the accuracy, efficiency, and reliability of wind turbine drivetrain fault diagnosis, helps ensure the stable operation of wind turbines in harsh environments, reduces maintenance costs, and improves power generation performance.

Description

Technical Field

[0001] This application belongs to the field of wind power generation technology, and relates to a method for identifying faults in the transmission chain of a wind turbine generator, and particularly to a method and system for identifying faults in the transmission chain of a wind turbine generator based on acoustic signals. Background Technology

[0002] With the rapid development of renewable energy, wind power, as a clean energy technology, has seen continuous increases in installed capacity and single-unit power. However, wind turbines are typically installed in harsh natural environments such as deserts and coastal areas, enduring complex operating conditions (such as strong winds, sandstorms, and salt spray corrosion) over long periods, leading to a high failure rate. Among these, the transmission system, as a core component of the wind turbine, directly affects the overall power generation efficiency and reliability. In particular, low-speed bearings and gears (such as main shaft bearings, planetary carrier bearings, planetary gear bearings, planetary gears, internal gear rings, and sun gears) are prone to wear and tooth breakage, severely impacting the stable operation of the wind turbine.

[0003] Currently, several technical solutions based on sound or image analysis exist for fault monitoring of mechanical equipment. For example, patent CN112484980A proposes a mechanical fault monitoring system and method based on sound and image analysis. This system collects operating sounds and images of the equipment, calculates the similarity between the fault sounds and images, and then assesses the risk of machine failure. However, this method has the following limitations:

[0004] (1) Poor environmental adaptability: The operating sound of the equipment is easily affected by temperature changes. The vibration characteristics of metal parts are different at different temperatures, which causes the sound characteristics to deviate and reduces the accuracy of fault identification.

[0005] (2) Long-term wear interference: As metal parts accumulate fatigue wear, their acoustic characteristics will gradually change, making comparative analysis based on fixed sound samples fail.

[0006] (3) Single verification dimension: It relies only on sound and image data and lacks collaborative analysis of multiple physical quantities (such as vibration, temperature, torque, etc.), making it difficult to fully verify fault characteristics, leading to misjudgment or omission.

[0007] Therefore, existing technologies are insufficient to meet the high-precision fault diagnosis requirements of wind turbine transmission systems under complex operating conditions. There is an urgent need for a fault monitoring method that can overcome environmental interference, adapt to component degradation, and integrate multi-dimensional information to improve the accuracy and reliability of fault diagnosis. Summary of the Invention

[0008] The purpose of this application is to provide a method and system for identifying wind turbine drivetrain faults based on acoustic signals, which solves the problems of inaccurate fault diagnosis caused by the reduced accuracy of existing wind turbine drivetrain fault monitoring methods based on sound recognition under different temperature environments, the influence of metal fatigue wear, and the difficulty in verifying fault results from multiple angles.

[0009] In a first aspect, this application provides a method for identifying wind turbine drivetrain faults based on acoustic signals. The method includes: acquiring sound signals at different locations on a wind turbine to be tested, and acquiring location information corresponding to the sound signals; filtering the sound signals to obtain a first sound signal; comparing and analyzing the first sound signals at adjacent locations with preset sample acoustic wave signals to confirm the fault point of the wind turbine and acquire fault point information; and performing residual cross-validation based on the fault point information to obtain valid fault point information, thereby achieving the final identification of the wind turbine drivetrain fault point.

[0010] In one implementation of the first aspect, obtaining the location information corresponding to the sound signal includes the following steps: establishing a three-dimensional device data model based on the sound signal and extracting key structural feature parameters; assigning a unique identifier to each sound collector; spatially associating the physical coordinates of the collector with the nodes of the three-dimensional device data model to obtain the location information corresponding to the sound signal; the spatial association includes: mapping the location information to the coordinate system of the three-dimensional device data model through a coordinate transformation algorithm; and establishing a correspondence table between the collector identifier and the model topology nodes.

[0011] In one implementation of the first aspect, comparing and analyzing the first sound signal at an adjacent location with a preset sample sound wave signal to identify the fault point of the wind turbine generator and obtaining fault point information includes: comparing the first sound signal with the preset sample sound wave signal to obtain a preliminary fault point; acquiring the real-time operating environment temperature, switching the corresponding sound comparison data of the preliminary fault point according to temperature changes, and dynamically adjusting the sample sound wave parameters of the preliminary fault point; and adjusting the verification data threshold of the fault sound according to the degree of metal fatigue wear of the mechanical transmission components to adapt to the verification requirements of different wear stages and form a comparison result; confirming the fault point and analyzing the fault type based on the comparison result to obtain fault point information.

[0012] In one implementation of the first aspect, the sample acoustic wave parameters include the propagation frequency of the acoustic wave on the metal within the device at a certain temperature; the formula for calculating the propagation frequency of the acoustic wave on the metal within the device at a certain temperature is:

[0013]

[0014] Where f2 represents the sound wave frequency at the current temperature T2; f1 represents the fixed sound wave frequency at a fixed temperature T1; α represents the temperature coefficient of the elastic modulus; and β represents the coefficient of thermal expansion.

[0015] In one implementation of the first aspect, adjusting the verification data threshold for fault sounds based on the degree of metal fatigue wear of the mechanical transmission components includes: obtaining the actual usage time of the metal components in the transmission chain of the mechanical transmission components; calculating the degree of fatigue wear of the metal components based on the actual usage time; and adjusting the verification data threshold for fault sounds based on the degree of fatigue wear.

[0016] In one implementation of the first aspect, the frequency of the sound wave during transmission is determined based on the wear value of the transmission chain; the greater the wear value of the transmission chain, the louder the sound wave; the smaller the wear value of the transmission chain, the softer the sound wave.

[0017] In one implementation of the first aspect, the formula for calculating the stress value of the transmission chain during actual transmission is as follows:

[0018] σ eff =σk

[0019] Where, σ eff Indicates the actual stress value; σ represents the nominal stress value; k represents the correction factor;

[0020] The formula for calculating the material failure state after N cycles under actual stress is as follows:

[0021]

[0022] Where N represents the number of cycles until fatigue failure of the material; m and G represent constants related to the material.

[0023] The formula for calculating the fatigue state of the material at the current cycle number is:

[0024]

[0025] Where, n i This indicates the number of cycles performed under different stress levels.

[0026] In one implementation of the first aspect, the formula for calculating the sound wave frequency is:

[0027]

[0028] Where P represents the sound wave frequency when the transmission chain wears; S represents the wear intensity of the transmission chain; T represents the ambient temperature; and q represents the correction coefficient, which is a positive number.

[0029] The greater the wear intensity of the transmission chain, the higher the sound wave frequency; the higher the temperature, the lower the sound wave frequency.

[0030] In one implementation of the first aspect, obtaining effective fault point information by performing aftershock cross-validation based on the fault point information includes: extracting the fault sound wave divergence aftershock signal based on the fault point information; acquiring adjacent aftershock data of adjacent sound signals of the target sound signal; and performing feature difference analysis on the fault sound wave divergence aftershock signal and the adjacent aftershock data to obtain effective fault point information.

[0031] Secondly, this application provides a method for identifying wind turbine drivetrain faults based on acoustic signals. The system includes: an acquisition module for acquiring sound signals at different locations on the wind turbine to be tested and acquiring the location information corresponding to the sound signals; a filtering module for filtering the sound signals to obtain a first sound signal; a comparison and analysis module for comparing and analyzing the first sound signals at adjacent locations with preset sample sound wave signals to confirm the fault point of the wind turbine and acquire fault point information; and a verification module for performing residual cross-verification based on the fault point information to obtain valid fault point information, thereby achieving the final identification of the wind turbine drivetrain fault point.

[0032] As described above, the wind turbine drivetrain fault identification method and system based on acoustic signals described in this application have the following beneficial effects:

[0033] The wind turbine drivetrain fault identification method provided in this application, through the setting of multiple sound acquisition devices, can perform divergence verification of fault sound sources on the generator drivetrain. By comparing and analyzing the sound from the sound acquisition devices around the fault sound source, more accurate fault data can be obtained. This application also utilizes a temperature monitor to monitor the current operating temperature, thereby determining the metal propagation efficiency of the generator drivetrain and thus the current sound wave condition, making the comparison data more accurate. Furthermore, by converting the fatigue wear degree of the generator drivetrain data, the application can determine the changes in fault sound waves based on the usage time and wear of the generator drivetrain, thus making the comparison data more accurate when updating the sample sound waves. Attached Figure Description

[0034] Figure 1 The diagram shows a hardware application scenario of the wind turbine drivetrain fault identification method based on acoustic signals described in this application embodiment.

[0035] Figure 2 The diagram shown is a schematic of the hardware connection structure of the wind turbine drivetrain fault identification device based on acoustic signals described in this application embodiment.

[0036] Figure 3 The diagram shown is a structural block diagram of the wind turbine drivetrain fault identification device based on acoustic signals described in an embodiment of this application.

[0037] Figure 4 The diagram shown is a schematic overall flow chart of the wind turbine drivetrain fault identification method based on acoustic signals described in the embodiments of this application.

[0038] Figure 5 The diagram shows a data processing flow of the wind turbine drivetrain fault identification method based on acoustic signals as described in the embodiments of this application.

[0039] Figure 6 The diagram shown is a flowchart of step S1 of the wind turbine drivetrain fault identification method based on acoustic signals described in this application embodiment.

[0040] Figure 7 The diagram shown is a flowchart of step S3 of the wind turbine drivetrain fault identification method based on acoustic signals described in this application embodiment.

[0041] Figure 8 The diagram shown is a flowchart of step S4 of the wind turbine drivetrain fault identification method based on acoustic signals described in this application embodiment.

[0042] Figure 9 The diagram shown is a schematic diagram of the acoustic wave verification divergence structure in one embodiment of the wind turbine drivetrain fault identification method based on acoustic signals described in this application.

[0043] Figure 10 The diagram shown is a schematic diagram of the principle structure of the wind turbine drivetrain fault identification system based on acoustic signals described in this application.

[0044] Figure 11 The diagram shown is a structural schematic of the electronic device described in an embodiment of this application.

[0045] Component designation explanation

[0046] 1. Wind turbine drivetrain fault identification device based on acoustic signals

[0047] 11 Terminal Controller

[0048] 12 Sound Collectors

[0049] 13 Temperature monitor

[0050] 111 Sound Acquisition Module

[0051] 112 Noise Removal Module

[0052] 113 Sound Contrast Module

[0053] 114 Fault Analysis Module

[0054] 115 Fault Verification Module

[0055] 116 Data Conversion Module

[0056] 117 Temperature Monitoring Module

[0057] 1121 Bandpass Filter

[0058] 1122 Transducer

[0059] 1131 Key Marking Unit

[0060] 1161 Data storage unit

[0061] 101 Acquisition Module

[0062] 102 Filtering Module

[0063] 103 Comparative Analysis Module

[0064] 104 Verification Module

[0065] 111 Memory

[0066] 112 processor

[0067] 113 Monitor

[0068] Steps S1 to S4 Detailed Implementation

[0069] The following specific examples illustrate the implementation of this application. Those skilled in the art can easily understand other advantages and effects of this application from the content disclosed in this specification. This application can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of this application. It should be noted that, unless otherwise specified, the following embodiments and features in the embodiments can be combined with each other.

[0070] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of this application. Therefore, the drawings only show the components related to this application and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0071] The wind turbine drivetrain fault identification method based on acoustic signals provided in the following embodiments of this application includes, but is not limited to, hardware application scenarios such as sound collectors and temperature monitors. The following description will take the hardware application scenario of the wind turbine drivetrain fault identification device based on acoustic signals as an example.

[0072] like Figure 1 As shown, this embodiment provides a hardware application scenario for a wind turbine drivetrain fault identification device based on acoustic signals, specifically including: a terminal controller 11, on which a temperature monitor 13 and multiple sound collectors 12 are connected. The terminal controller 11 is equipped with a sound acquisition module 111, a noise filtering module 112, a sound comparison module 113, a fault analysis module 114, and a fault verification module 115.

[0073] Please see Figure 2 and Figure 3 The figures shown are a schematic diagram of the hardware connection structure of the wind turbine drivetrain fault identification device based on acoustic signals according to the embodiments of this application, and a structural block diagram of the wind turbine drivetrain fault identification device based on acoustic signals according to the embodiments of this application.

[0074] The sound acquisition module 111 connects to multiple sound acquisition devices, which are installed at different positions on the wind turbine transmission mechanism. Each acquisition device is assigned a different number to correspond to different positions, so as to represent the position in the model.

[0075] The noise filtering module 112 filters noise through a bandpass filter 1121.

[0076] The sound comparison module 113 is connected to a data conversion module 116 and a temperature monitoring module 117. The sound comparison module 113 compares the incoming sound with the sample sound wave and performs fault analysis through the fault analysis module 114. During this process, the temperature monitoring module 117 monitors the working environment and adjusts the sample sound wave through the data conversion module 116. The data conversion module 116 is equipped with a data storage unit 1161 for storing temperature data and metal wear data. The sound comparison module 113 is equipped with a key marking unit 1131 for focused monitoring and comparison of frequently occurring fault points, and the number of comparisons is adjusted according to the frequency.

[0077] The fault analysis module 114 is used to analyze and confirm the fault sound waves, and to save and send the fault information.

[0078] The fault verification module 115 uses the residual waves generated during sound wave divergence to perform sound wave comparison analysis on nearby sound collectors other than the target sound collector, thereby making a sound wave verification judgment.

[0079] In this embodiment, the noise filtering module 112 receives sound waves of different frequencies through a bandpass filter 1121. When it is working, it converts the sound waves into electrical signals through a transducer 1122 (such as a microphone), and inputs the collected electrical signals into the bandpass filter 1121. The bandpass filter 1121 has a specific frequency range, allowing only frequency components within this range to pass through, while blocking other frequency components.

[0080] The formula for calculating the center frequency within a band filter is as follows:

[0081]

[0082] Where f0 represents the center frequency value within the bandpass filter, which is adjusted by changing the value of the inductor L or the capacitor C; L represents the inductor; C represents the capacitor.

[0083] Initially, the sound comparison module 113 records the sound of the device operating normally, and the recorded data consists of multiple sets. The average sound wave value of the multiple sets of sound is used as the sample sound wave to form a comparison model.

[0084] Specifically, the sound comparison module 113 initially records the normal operating sound of the device, and the recorded data consists of multiple sets. The average sound wave value of the multiple sets of sound is used as the sample sound wave to form a comparison model. This comparison model can be adjusted and changed according to changes in time and temperature. For example, in winter, due to the low temperature, the sound waves emitted by the metal transmission are significantly different from those at room temperature. This is because at room temperature, metals undergo thermal expansion due to their higher temperature, resulting in a relatively large distance between molecules. During metal transmission, molecular vibration is relatively easy to generate sound waves through collisions and friction between components. Furthermore, due to the relatively loose structure of metal, the vibration propagation speed is also slightly faster, making it easier to generate higher frequency sound waves. However, at low temperatures, the metal contracts, the distance between molecules decreases, the structure becomes more compact, vibration transmission becomes relatively difficult, and the generated sound wave frequency is relatively low.

[0085] In summary, this application, through the setting of a temperature monitoring module and a temperature monitor, can monitor the current operating temperature, thereby determining the metal propagation efficiency of the generator drive chain, and thus determining the current sound wave condition, making the comparison data more accurate; by converting the fatigue wear degree of the generator drive chain into data, the changes in the fault sound waves of the generator drive chain can be determined based on the usage time and wear of the generator drive chain, thus making the comparison data more accurate when updating the sample sound waves; and through the setting of the fault verification module and the function of multiple sound collectors, the radiation verification of the fault sound source on the generator drive chain can be performed, and the sound comparison analysis of the sound collectors around the fault sound source can be performed to obtain more accurate fault data.

[0086] The technical solutions in the embodiments of this application will be described in detail below with reference to the accompanying drawings.

[0087] This application first installs multiple sound acquisition devices inside the equipment and assigns digital labels to each device. Then, a three-dimensional data model is created based on the equipment's form, and each sound acquisition device is connected to its corresponding position in the data model according to its label. Next, the target sound collected by adjacent sound acquisition devices is compared and verified to confirm the fault point. Then, by monitoring the temperature of the working environment, the corresponding comparison data is changed, thereby adjusting the verification of the fault sound according to different environments. Furthermore, the verification data for the fault sound is adjusted based on metal fatigue wear during mechanical transmission, thus changing the verification conditions. Finally, the fault information is transmitted to the terminal, and an early warning is issued. This application, through sound recognition technology combined with other possible monitoring methods (such as image recognition), can more accurately identify and diagnose faults in the wind turbine drivetrain, including bearing and gear wear, broken teeth, and other problems. This improves the accuracy of fault diagnosis and equipment maintenance efficiency, while also reducing the failure rate of the wind turbine and extending the equipment's service life.

[0088] Please see Figure 4 and Figure 5 The figures shown are a general flowchart of the wind turbine drivetrain fault identification method based on acoustic signals according to the embodiments of this application and a data processing flowchart of the wind turbine drivetrain fault identification method based on acoustic signals according to the embodiments of this application.

[0089] like Figure 4 and Figure 5 As shown, this embodiment provides a method for identifying wind turbine drivetrain faults based on acoustic signals. The method includes the following steps:

[0090] S1, acquire sound signals from different locations on the wind turbine to be tested, and acquire the location information corresponding to the sound signals.

[0091] Please see Figure 6 The diagram shows a flowchart of step S1 of the wind turbine drivetrain fault identification method based on acoustic signals described in this application embodiment. Figure 6 As shown, S1 includes the following steps:

[0092] S11, Based on the sound signal, establish a three-dimensional device data model and extract key structural feature parameters;

[0093] S12, assign a unique identifier to each sound collector;

[0094] S13, Spatially associate the physical coordinates of the data collector with the nodes of the 3D device data model to obtain the location information corresponding to the sound signal. The spatial association includes: mapping the location information to the coordinate system of the 3D device data model through a coordinate transformation algorithm; and establishing a correspondence table between the data collector identifier and the model topology nodes.

[0095] In this embodiment, several sound collectors are installed on the wind turbine to be tested. These sound collectors are installed at different positions on the wind turbine's transmission mechanism. Each collector is assigned a different number corresponding to its position, which is used to represent the location within the model.

[0096] Specifically, multiple sound acquisition devices are set up inside the device, and each sound acquisition device is digitally labeled; a three-dimensional data model is established based on the device's shape, and each sound acquisition device is connected to the corresponding position in the data model according to its label.

[0097] Geometric data of the device surface can be obtained using 3D scanning technology or photogrammetry; the physical location coordinates and corresponding number information of each sound acquisition device (such as a microphone) can be recorded. The technical parameters of the microphones can be collected (or a multi-microphone array can be used, and the geometric configuration information of the array can be recorded).

[0098] A three-dimensional device data model is generated based on several collected sound signal data, and the physical coordinates of each collector are correlated with the sound signals, that is, their respective position coordinate information is matched one by one, and a space-device relationship database is established.

[0099] S2, the sound signal is filtered to obtain the first sound signal.

[0100] In this embodiment, the collected sound signal is filtered for noise to obtain a filtered sound signal, which is used as the first sound signal.

[0101] Specifically, firstly, a transducer (such as a microphone) is used to receive the sound waves generated by the wind turbine during operation, converting the sound waves in the target frequency band into electrical signals, and ensuring that the sensitivity matches the target frequency band. Then, the acquired electrical signals are input to a bandpass filter for frequency filtering. This involves determining the frequency range and stopband attenuation requirements of the bandpass filter, allowing only sound frequencies within the preset frequency range to pass through the filter, while blocking and filtering out sound frequencies outside the preset frequency range. The signal passing through the bandpass filter is the sound signal after filtering out noise; these retained signals are used for subsequent sound analysis and fault diagnosis. Finally, the signal processed by the bandpass filter is output to the comparative analysis module for further analysis.

[0102] This step can effectively extract useful signals from sounds containing multiple frequency components, providing an accurate data basis for sound identification of wind turbine drivetrain faults.

[0103] S3, compare the first sound signal at the adjacent position with the preset sample sound wave signal and perform fault analysis to confirm the fault point of the wind turbine and obtain the fault point information.

[0104] Please see Figure 7 The diagram shows a flowchart of step S3 of the wind turbine drivetrain fault identification method based on acoustic signals described in this application embodiment. Figure 7 As shown, step S3 includes the following steps:

[0105] S31, compare the first sound signal with the preset sample sound wave signal to obtain the preliminary fault point;

[0106] S32, acquire the real-time working environment temperature, switch the corresponding sound comparison data of the initial fault point according to the temperature change, and dynamically adjust the sample sound wave parameters of the initial fault point.

[0107] S33, adjust the verification data threshold of fault sound according to the degree of metal fatigue wear of mechanical transmission components, so as to adapt to the verification requirements of different wear stages and form a comparison result.

[0108] S34. Based on the comparison results, confirm the fault point and analyze the fault type to obtain fault point information.

[0109] In this embodiment, the target sound is collected by adjacent sound acquisition devices and compared and verified to confirm the fault point; the working environment temperature is monitored in real time, and the corresponding sound comparison data is switched according to the temperature change to dynamically adjust the verification conditions of the fault sound; the verification data threshold of the fault sound is adjusted according to the degree of metal fatigue wear of the mechanical transmission components to adapt to the verification requirements of different wear stages.

[0110] The sample acoustic parameters include the propagation frequency of acoustic waves on metal within the device at a certain temperature.

[0111] Specifically, the temperature monitor monitors the temperature of the equipment and calculates the frequency of sound waves propagating on the metal inside the equipment based on the real-time temperature.

[0112] The formula for calculating the frequency of sound wave propagation on metal within a device at a certain temperature is:

[0113]

[0114] Where f2 represents the sound wave frequency at the current temperature T2; f1 represents the fixed sound wave frequency at a fixed temperature T1; α represents the temperature coefficient of the elastic modulus; and β represents the coefficient of thermal expansion.

[0115] It should be noted that this formula is used to calculate the sound frequency f2 at the current temperature T2. When the temperature is higher, the elastic modulus of the metal material usually decreases, and the coefficient of thermal expansion will increase with the increase of temperature, thus increasing within a certain range. This causes the transmission frequency of sound in the metal to decrease, which in turn causes the fault sound to change at different temperatures.

[0116] In this embodiment, step S33 includes: obtaining the actual usage time of the metal components in the transmission chain of the mechanical transmission component; calculating the degree of fatigue wear of the metal components based on the actual usage time; and adjusting the verification data threshold of the fault sound based on the degree of fatigue wear. Specifically, the sound wave frequency of the transmission chain during transmission is determined based on the wear value of the transmission chain; the greater the wear value of the transmission chain, the louder the sound wave; and the smaller the wear value of the transmission chain, the quieter the sound wave.

[0117] Specifically, when converting data, the fatigue wear of the metal is calculated by measuring the usage time of the metal inside the device, thereby adjusting the verification data for the sound waves.

[0118] The formula for calculating the stress value of the transmission chain during actual transmission is as follows:

[0119] σ eff =σk

[0120] Where, σ eff σ represents the actual stress value; k represents the nominal stress value; k represents the correction factor.

[0121] The formula for calculating the stress value after N cycles under actual stress until the material fails is:

[0122]

[0123] Where N represents the number of cycles until fatigue failure; m and G represent constants related to the material. This formula is used to express the actual stress value. After N cycles, the material will reach a failure state.

[0124] Therefore, the formula for calculating the fatigue state of the material at the current number of cycles is:

[0125]

[0126] Where, n i Indicated under different stresses The number of cycles completed at the horizontal level indicates the fatigue state of the material at the current cycle number.

[0127] Substitute Ni into the above formula:

[0128]

[0129] The current fatigue wear value of the conveyor chain is obtained. Based on the wear value of the conveyor chain, the sound wave frequency of the conveyor chain during transmission is determined. When the wear fatigue value is greater, the sound wave becomes louder.

[0130] In this embodiment, the key marking unit is used to focus on monitoring and comparing frequently faulty points, and the number of comparisons is adjusted according to the frequency. For example, when multiple faults occur at point T, two comparison operations are performed through the sound comparison module after a fault occurs at point T again.

[0131] By combining temperature monitoring module with conveyor chain wear data, the acoustic frequency of the conveyor chain during wear can be calculated.

[0132] The formula for calculating the frequency of the sound wave is:

[0133]

[0134] Where P represents the sound wave frequency when the transmission chain wears; S represents the wear intensity of the transmission chain; T represents the ambient temperature; and q represents the correction coefficient, which is a positive number.

[0135] The greater the wear intensity S of the transmission chain, the greater the sound wave frequency P; and the higher the temperature T, the smaller the sound wave frequency P.

[0136] S4. Based on the fault point information, perform residual cross-verification to obtain valid fault point information, so as to achieve the final identification of the wind turbine drivetrain fault point.

[0137] Please see Figure 8 The diagram shows a flowchart of step S4 of the wind turbine drivetrain fault identification method based on acoustic signals described in this embodiment of the application. Figure 8 As shown, step S4 includes the following steps:

[0138] S41, extract the fault acoustic wave divergence residual signal based on the fault point information;

[0139] S42, acquire the adjacent echo data of the adjacent sound signals of the target sound signal;

[0140] S43, perform feature difference analysis on the fault acoustic wave divergence residual signal and the adjacent residual data to obtain effective fault point information.

[0141] In this embodiment, the residual waves generated during sound wave divergence are used to perform sound wave comparison analysis on nearby sound collectors other than the target sound collector, thereby enabling sound wave verification and judgment.

[0142] Specifically, the target sound signal is compared with the signal characteristics of the centralized control acquisition unit, and their consistency is verified. That is, if there is a significant difference between the target sound signal and adjacent sound signals, interference is suspected; if the difference is not significant, the sound wave propagation path is verified to be as expected. The pulse-echo method can be used to analyze reflection time and locate the source of the abnormal signal. Ultrasonic interference or white noise tests are used to confirm whether the acquisition unit interferes with a specific frequency band, while ruling out signal anomalies caused by non-sound wave factors. A sound wave propagation path map and comparative analysis records are generated to obtain the final fault point, and the fault point information is saved.

[0143] The following example illustrates the process of identifying wind turbine drivetrain faults using sound recognition.

[0144] Please see Figure 9 The image shown is a schematic diagram of the acoustic wave verification divergence structure in one embodiment of the wind turbine drivetrain fault identification method based on acoustic signals described in this application.

[0145] Multiple sound acquisition devices are installed inside the device, and each sound acquisition device is digitally labeled; a three-dimensional data model is established based on the device's shape, and each sound acquisition device is connected to the corresponding position in the data model according to its label.

[0146] For example, Figure 9 In this example, with T as the target position, a sound acquisition device is set at position T, and sound acquisition devices corresponding to T1, T2, T3, T4, T5 and T6 are set at positions adjacent to T respectively.

[0147] During the recognition process, when the sound collector T (e.g.) Figure 9When a sound wave is detected at location T, it is initially filtered by a bandpass filter on the noise filtering module. Then, the filtered sound wave is compared using a sample comparison model. If an anomaly occurs, it is verified by a fault verification module. At this point, the sound wave at location T spreads to locations T1, T2, T3, T4, T5, and T6. Fault verification is then performed by monitoring the residual sound waves at locations T1 to T6, and the verification data is transmitted, stored, and used for early warning. Simultaneously, the comparison model is adjusted based on the equipment's usage time. Specifically, if the usage time is too long, the generator drive chain experiences significant fatigue wear, resulting in different sound waves. Therefore, the comparison model is modified to ensure more accurate comparison data.

[0148] In summary, the acoustic signal-based wind turbine drivetrain fault identification method described in this application can more accurately identify and diagnose wind turbine drivetrain faults, including bearing and gear wear, broken teeth, and other problems. By comprehensively analyzing multiple data sources, including sound and images, to verify fault results from multiple angles, the fault diagnosis becomes more comprehensive and reliable. Furthermore, it enables convenient and efficient fault risk prediction and diagnosis, helping to improve wind turbine maintenance efficiency, reduce downtime, and increase power generation efficiency. By timely detecting and diagnosing faults, this application helps reduce the failure rate of wind turbines and extend equipment lifespan. Moreover, by improving the accuracy of drivetrain fault diagnosis, it helps maintain the good performance of the drivetrain, thereby improving overall power generation efficiency. This application can operate normally in harsh natural environments such as deserts, rivers, and seas, and maintains the accuracy of sound recognition even under different temperature conditions, thus adapting to the complex environments in which wind turbines operate and facilitating fault identification in complex environments.

[0149] The scope of protection for the wind turbine drivetrain fault identification method based on acoustic signals described in this application is not limited to the execution order of the steps listed in this embodiment. Any solution implemented by adding, subtracting, or replacing steps in the prior art based on the principles of this application is included within the scope of protection of this application.

[0150] This application also provides a wind turbine drivetrain fault identification system based on acoustic signals. The wind turbine drivetrain fault identification system based on acoustic signals can implement the wind turbine drivetrain fault identification method based on acoustic signals described in this application. However, the implementation device of the wind turbine drivetrain fault identification method based on acoustic signals described in this application includes, but is not limited to, the structure of the wind turbine drivetrain fault identification system based on acoustic signals listed in this embodiment. All structural modifications and substitutions of the prior art made according to the principles of this application are included within the protection scope of this application.

[0151] like Figure 10As shown, this embodiment provides a wind turbine drivetrain fault identification system based on acoustic signals, including: an acquisition module 101, a filtering module 102, a comparison and analysis module 103, and a verification module 104.

[0152] The acquisition module 101 is used to acquire sound signals at different locations on the wind turbine to be tested, and to acquire the location information corresponding to the sound signals.

[0153] In this embodiment, several sound collectors are installed on the wind turbine to be tested. These sound collectors are installed at different positions on the wind turbine's transmission mechanism. Each collector is assigned a different number corresponding to its position, which is used to represent the location within the model.

[0154] The filtering module 102 is used to filter the sound signal to obtain a first sound signal.

[0155] In this embodiment, the collected sound signal is filtered for noise to obtain a filtered sound signal, which is used as the first sound signal.

[0156] The comparison and analysis module 103 is used to compare and analyze the first sound signal at adjacent locations with the preset sample sound wave signal to identify the fault point of the wind turbine and obtain the fault point information.

[0157] In this embodiment, the target sound is collected by adjacent sound acquisition devices and compared and verified to confirm the fault point; the working environment temperature is monitored in real time, and the corresponding sound comparison data is switched according to the temperature change to dynamically adjust the verification conditions of the fault sound; the verification data threshold of the fault sound is adjusted according to the degree of metal fatigue wear of the mechanical transmission components to adapt to the verification requirements of different wear stages.

[0158] The verification module 104 is used to perform residual cross-verification based on the fault point information to obtain valid fault point information, so as to achieve the final identification of the wind turbine drive chain fault point.

[0159] In this embodiment, the residual waves generated during sound wave divergence are used to perform sound wave comparison analysis on nearby sound collectors other than the target sound collector, thereby enabling sound wave verification and judgment.

[0160] The wind turbine drivetrain fault identification system based on acoustic signals can perform radiation verification of fault sound sources on the generator drivetrain. By comparing and analyzing the sound through sound collectors around the fault sound source, more accurate fault data can be obtained. It can also monitor the current operating temperature to determine the metal propagation efficiency of the generator drivetrain, thereby determining the current sound wave condition and making the comparison data more accurate.

[0161] It should be understood that the disclosed systems, apparatuses, or methods can be implemented in other ways, given the several embodiments provided in this application. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules / units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or units may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules or units may be electrical, mechanical, or other forms.

[0162] The modules / units described as separate components may or may not be physically separate. The components shown as modules / units may or may not be physical modules; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules / units can be selected to achieve the objectives of the embodiments of this application, depending on actual needs. For example, the functional modules / units in the various embodiments of this application may be integrated into one processing module, or each module / unit may exist physically separately, or two or more modules / units may be integrated into one module / unit.

[0163] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0164] Please see Figure 11 The diagram shows a structural schematic of the electronic device described in an embodiment of this application. Figure 11 As shown, this embodiment provides an electronic device, the electronic device 110 including a memory 111 and a processor 112.

[0165] The memory 111 is used to store computer programs; preferably, the memory 111 includes various media that can store program code, such as ROM, RAM, magnetic disk, USB flash drive, memory card or optical disk.

[0166] Specifically, memory 111 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) and / or cache memory. Electronic device 110 may further include other removable / non-removable, volatile / non-volatile computer system storage media. Memory 111 may include at least one program product having a set (e.g., at least one) of program modules configured to perform the functions of the embodiments of this application. It is understood that memory 111 may be volatile memory or non-volatile memory, or both. Non-volatile memory may be read-only memory (ROM) or programmable read-only memory (PROM), which serves as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM) and synchronous static random access memory (SSRAM). The memories described in the embodiments of the present invention are intended to include, but are not limited to, these and any other suitable categories of memory.

[0167] The processor 112 is connected to the memory 111 and is used to execute the computer program stored in the memory 111 so that the electronic device 110 executes the wind turbine drive train fault identification method based on acoustic signals as described in any embodiment of this application.

[0168] Optionally, the processor 112 may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it may also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components.

[0169] Optionally, the electronic device 110 in this embodiment may further include a display 113. The display 113 is communicatively connected to the memory 111 and the processor 112, and is used to display the relevant graphical user interface (GUI) of the wind turbine drivetrain fault identification method based on acoustic signals described in the embodiments of this application and / or the wind turbine drivetrain fault identification method based on acoustic signals described in other embodiments of this application.

[0170] This application also provides a computer-readable storage medium storing a computer program thereon. When executed by a processor, the program implements the wind turbine drivetrain fault identification method based on acoustic signals described in any embodiment of this application and / or the wind turbine drivetrain fault identification method based on acoustic signals described in other embodiments of this application.

[0171] As used in this specification, the terms "component," "module," "system," etc., are used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As illustrated, applications running on computing devices and computing devices can both be components. One or more components may reside in a process and / or an execution thread, and components may be located on a single computer and / or distributed among two or more computers. Furthermore, these components can be executed from various computer-readable media on which various data structures are stored. Components can communicate, for example, via local and / or remote processes based on signals having one or more data packets (e.g., data from two components interacting with another component between a local system, a distributed system, and / or a network, such as the Internet interacting with other systems via signals).

[0172] In summary, the wind turbine drivetrain fault identification method and system based on acoustic signals provided in this application have the following beneficial effects:

[0173] The wind turbine drivetrain fault identification method provided in this application, through the setting of multiple sound acquisition devices, can perform divergence verification of fault sound sources on the generator drivetrain. By comparing and analyzing the sound from the sound acquisition devices around the fault sound source, more accurate fault data can be obtained. This application also utilizes a temperature monitor to monitor the current operating temperature, thereby determining the metal propagation efficiency of the generator drivetrain and thus the current sound wave condition, making the comparison data more accurate. Furthermore, by converting the fatigue wear degree of the generator drivetrain data, the application can determine the changes in fault sound waves based on the usage time and wear of the generator drivetrain, thus making the comparison data more accurate when updating the sample sound waves.

[0174] The descriptions of the processes or structures corresponding to the above figures each have their own emphasis. For parts of a process or structure that are not described in detail, please refer to the relevant descriptions of other processes or structures.

[0175] The above embodiments are merely illustrative of the principles and effects of this application and are not intended to limit this application. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of this application. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in this application should still be covered by the claims of this application.

Claims

1. A method for identifying wind turbine drivetrain faults based on acoustic signals, characterized in that, include: Acquiring sound signals from different locations on a wind turbine under test and obtaining the corresponding location information for the sound signals; including: establishing a three-dimensional device data model based on the sound signals and extracting key structural feature parameters; assigning a unique identifier to each sound collector; spatially associating the physical coordinates of the collectors with the nodes of the three-dimensional device data model to obtain the location information corresponding to the sound signals; the spatial association includes: mapping the location information to the coordinate system of the three-dimensional device data model through a coordinate transformation algorithm; and establishing a correspondence table between collector identifiers and model topology nodes; The sound signal is filtered to obtain a first sound signal; The process involves comparing and analyzing the first sound signal from adjacent locations with preset sample sound wave signals to identify the fault point of the wind turbine and obtain fault point information. This includes: comparing the first sound signal with preset sample sound wave signals to obtain a preliminary fault point; acquiring the real-time operating environment temperature, switching the corresponding sound comparison data for the preliminary fault point based on temperature changes, and dynamically adjusting the sample sound wave parameters of the preliminary fault point; acquiring the actual usage time of the metal components in the transmission chain of the mechanical transmission components, calculating the fatigue wear degree of the metal components based on the actual usage time, and adjusting the verification data threshold of the fault sound according to the fatigue wear degree to adapt to the verification requirements of different wear stages and form a comparison result; confirming the fault point and analyzing the fault type based on the comparison result to obtain fault point information. Based on the fault point information, residual cross-validation is performed to obtain valid fault point information, thereby achieving the final identification of the wind turbine drivetrain fault point; including: extracting the fault acoustic wave divergence residual signal based on the fault point information; obtaining adjacent residual data of adjacent acoustic signals of the target acoustic signal; and performing feature difference analysis on the fault acoustic wave divergence residual signal and the adjacent residual data to obtain valid fault point information.

2. The method for wind turbine drivetrain fault identification based on acoustic signals according to claim 1, characterized in that, The sample acoustic parameters include the propagation frequency of acoustic waves on metal within the device at a certain temperature. The formula for calculating the frequency of sound wave propagation on metal within a device at a certain temperature is: ， in, Indicates the current temperature The frequency of the sound wave at that time; Indicates a fixed temperature The fixed sound wave frequency; Temperature coefficient representing the modulus of elasticity; This represents the coefficient of thermal expansion.

3. The method for wind turbine drivetrain fault identification based on acoustic signals according to claim 1, characterized in that, The frequency of the sound waves transmitted by the conveyor chain is determined based on the wear value of the conveyor chain. The greater the wear value of the transmission chain, the louder the sound wave; the smaller the wear value of the transmission chain, the softer the sound wave.

4. The method for identifying wind turbine drivetrain faults based on acoustic signals according to claim 1, characterized in that, The formula for calculating the stress value of the transmission chain during actual transmission is as follows: ， in, Indicates the actual stress value; Indicates the nominal stress value; Indicates the correction factor; The formula for calculating the material failure state after N cycles under actual stress is as follows: ， in, This represents the number of cycles required for the material to reach fatigue failure. , These represent constants related to the material; The formula for calculating the fatigue state of the material at the current cycle number is: ， ， ， in, This indicates the number of cycles performed under different stress levels.

5. The method for wind turbine drivetrain fault identification based on acoustic signals according to claim 3, characterized in that, The formula for calculating the frequency of the sound wave is: ， in, This indicates the frequency of the sound waves when the transmission chain wears down. Indicates the wear intensity of the transmission chain; Indicates ambient temperature; This represents the correction factor, and it must be a positive number. The greater the wear intensity of the transmission chain, the higher the sound wave frequency. The higher the temperature, the lower the frequency of the sound waves.

6. A system for performing the wind turbine drivetrain fault identification method based on acoustic signals according to any one of claims 1 to 5, characterized in that, The system includes: The acquisition module is used to acquire sound signals from different locations on the wind turbine to be tested, and to acquire the location information corresponding to the sound signals; A filtering module is used to filter the sound signal to obtain a first sound signal; The comparison and analysis module is used to compare the first sound signal at adjacent locations with the preset sample sound wave signal and perform fault analysis to identify the fault point of the wind turbine and obtain fault point information. The verification module is used to perform residual cross-verification based on the fault point information to obtain valid fault point information, so as to achieve the final identification of the wind turbine drivetrain fault point.

Images