Wavelet multi-scale fan blade defect detection method based on rotation speed synchronization time difference correction

Abstract

The invention relates to the field of fan blade defect detection, and discloses a wavelet multi-scale fan blade defect detection method based on rotation speed synchronization time difference correction, and the method comprises the steps: installing a sound pickup at the root of a fan blade; mounting a rotating speed sensor on a fan shaft; the method comprises the following steps: acquiring a sound signal of fan blade rotation, and digitizing; sound signal time differences of different blades are eliminated through a rotating speed sensor, and correction signals are obtained; performing feature extraction on each correction signal to obtain a feature signal; obtaining a correlation coefficient between the characteristic signals; fan blade defect detection is carried out according to the correlation coefficient, and an alarm is given; according to the method, the limitation of a traditional monitoring method is overcome, a high-precision, real-time and economical solution is further provided, and the method is suitable for various complex environments and different types of fans and has wide application prospects.

Description

Technical Field

[0001] The present invention relates to the field of fan blade defect detection, and in particular to a wavelet multi-scale fan blade defect detection method based on rotation speed synchronization time difference correction. Background Art

[0002] In recent years, with the growing global demand for renewable energy, wind power has rapidly developed as a clean and sustainable energy source. However, wind turbines are often installed in remote locations with harsh environmental conditions, such as mountains, wilderness, beaches, and islands. These environments not only increase the difficulty of equipment maintenance but also place higher demands on the reliability and durability of wind turbines.

[0003] Blades are one of the most critical components of a wind turbine, and their performance directly impacts power generation efficiency and system stability. Blades are typically designed and manufactured using composite materials. While lightweight and strong, these materials are susceptible to various environmental factors such as wind, sand, salt spray, ultraviolet light, and lightning over long-term operation, leading to surface damage and internal structural fatigue. These issues not only reduce blade efficiency but can also pose serious safety risks, such as blade breakage. Statistics show that blade failures account for up to 34% of all wind turbine downtime, severely impacting power generation efficiency and significantly increasing operational and maintenance costs. Therefore, real-time and accurate health monitoring of wind turbine blades is crucial.

[0004] Traditional blade monitoring methods primarily rely on technologies such as acoustic emission, vibration analysis, fiber Bragg grating (FBG), and infrared thermal imaging. While these methods each have their own advantages, they also have limitations, such as complex data acquisition, difficult analysis, and high costs. In practical applications, the diverse nature of wind farm locations, turbine types, fault types, and background environments makes it difficult for traditional rule- and feature-based monitoring methods to meet high-precision, real-time requirements. Furthermore, while deep learning-based methods offer superior recognition capabilities, their application in practical engineering projects remains challenging due to limitations in data acquisition scale and insufficient model generalization. Summary of the Invention

[0005] The purpose of the present invention is to propose a wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction to solve the technical problems that existing wind turbine blade defect monitoring and analysis are difficult to meet high precision and real-time requirements.

[0006] Specifically, the present invention provides a wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction, comprising the following steps: S1. Install a pickup at the root of the fan blade; install a speed sensor on the fan shaft; S2, collecting and digitizing the acoustic signals of the fan blades; S3, using a speed sensor to eliminate the time difference of acoustic signals of different blades to obtain a correction signal; S4, extracting features from each correction signal to obtain a feature signal; S5. Obtaining correlation coefficients between characteristic signals; S6. Detect fan blade defects based on the correlation coefficient and generate an alarm.

[0007] A storage device stores instructions and data for implementing a wavelet multi-scale fan blade defect detection method based on rotation speed synchronization time difference correction.

[0008] A wavelet multi-scale wind turbine blade defect detection device based on speed synchronization time difference correction includes: a processor and a storage device; the processor loads and executes instructions and data in the storage device to implement a wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction.

[0009] The present invention significantly improves wind turbine blade health monitoring, reduces maintenance costs, and ensures safe and stable wind turbine operation. This technology not only overcomes the limitations of traditional monitoring methods but also provides a highly accurate, real-time, and cost-effective solution. It is applicable to various complex environments and different types of wind turbines, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 1. It is a flow chart of the wavelet multi-scale fan blade defect detection method based on speed synchronization time difference correction of the present invention; Figure 2 This is a working diagram of the hardware equipment for this application. DETAILED DESCRIPTION

[0011] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0012] Before formally explaining the present invention, the scheme of the present invention is first generally explained for easy understanding.

[0013] Please refer to Figure 1 The present invention provides a wavelet multi-scale fan blade defect detection method based on speed synchronization time difference correction, comprising the following steps: S1. Install a pickup at the root of the fan blade; install a speed sensor on the fan shaft; It should be noted that the present invention installs a microphone near the root of each blade of the wind turbine to capture the noise signal generated during blade rotation. A speed sensor is installed on the wind turbine shaft to monitor the wind turbine's rotation speed in real time.

[0014] S2, collecting and digitizing the acoustic signals of the fan blades; Specifically, the signal collected by the pickup is The instantaneous value at . is the leaf number. These signals are sampled into discrete time series ,in is the sampling point index. The sampling frequency is set to , then the sampling interval is .

[0015] The digitized signal is represented as: .

[0016] S3, using a speed sensor to eliminate the time difference of acoustic signals of different blades to obtain a correction signal; Step S3 is as follows: S31. Obtain the real-time speed of the fan according to the speed sensor and calculate the fan rotation period; S32, determining the blade delay according to the fan rotation period; Delay in step S32 The calculation formula is as follows:

[0017] in, i is the number of the fan blade, t i For the i The time point at which the wind turbine blade pickup collects the sound, t 1 is the acquisition time point of the first fan blade, which is also the reference time point. r ( t 1) Real-time speed information of the fan S33. Correct the signals generated by different blades according to the calculated time delay to obtain corrected signals.

[0018] The signal after correction in step S33 is as follows:

[0019] in is the corrected signal; n Sampling point index As an embodiment, the speed information is obtained as follows: the speed sensor provides the real-time speed information of the fan Or the number of rotations per unit time. Calculate the fan rotation period based on the speed information , .

[0020] For the case of three blades, the theoretical time interval for each blade to pass through the pickup is .

[0021] Time difference calculation: Assume that the time points at which three blades pass through the pickup are ,in The difference between the actual time point and the theoretical time point is calculated to determine the delay : The delay of the second blade :

[0022] The delay of the third blade :

[0023] Delay correction: The signal is corrected based on the calculated delay so that all signals can be compared at the same rotation angle. The corrected signal is expressed as:

[0024]

[0025] S4, extracting features from each correction signal to obtain a feature signal; It should be noted that step S4 is specifically as follows: S41, perform discrete wavelet transform on the corrected signal to obtain different scales l The signal below; S42, for different scales l The feature extraction is performed on the signal under the wavelet to obtain a feature signal, wherein the feature signal includes: wavelet coefficients, energy distribution and peak position.

[0026] As an embodiment, for each corrected noise signal Perform discrete wavelet transform (DWT) to obtain multi-level wavelet coefficients Is the decomposition level. Use appropriate wavelet basis functions, such as Haar wavelet, Daubechies wavelet, etc., to perform multi-scale analysis. Discrete wavelet transform can decompose the signal into a series of approximate coefficients and detail coefficients, which correspond to different scales. The following signal characteristics.

[0027]

[0028] in , is the minimum wavelet scale.

[0029] Feature extraction: Extract signal features at different scales, such as energy distribution, peak position, etc.

[0030] For discrete wavelet transform, the wavelet coefficients at each level can be Extract features. The energy distribution at each scale can be calculated:

[0031] The peak position at each scale can also be calculated:

[0032] These features can define the local characteristics of the signal at different scales.

[0033] in: It is mainly used to represent the position index in the wavelet coefficient sequence and to calculate the correlation between wavelet coefficients, energy distribution and peak position.

[0034] S5. Obtaining correlation coefficients between characteristic signals; It should be noted that, in step S5 , a cross-correlation function is used to calculate the correlation coefficient, and the correlation coefficient is further standardized.

[0035] As an example, for each scale , calculate the correlation between the wavelet coefficients of the three blade noise signals.

[0036] Correlation can be estimated by the cross-correlation function:

[0037] in yes The complex conjugate of .

[0038] in, This is also the number of the leaf. The leaves and The correlation coefficient of the leaves.

[0039] To eliminate the effect of amplitude differences, the normalized correlation can be calculated:

[0040] in and They are and The autocorrelation function of .

[0041] For each scale , calculate the correlation between the energy distributions of the three blade noise signals.

[0042]

[0043] To eliminate the effect of amplitude differences, the normalized correlation can be calculated:

[0044] in is the autocorrelation function of the energy distribution.

[0045] For each scale , calculate the correlation between the peak positions of the three blade noise signals.

[0046]

[0047] To eliminate the effect of amplitude differences, the normalized correlation can be calculated:

[0048] in is the autocorrelation function of the peak position.

[0049] S6. Detect fan blade defects based on the correlation coefficient and generate an alarm.

[0050] Step S6 is specifically as follows: setting a threshold value according to the wavelet coefficient characteristics of the normal leaf signal, for each scale , set the corresponding threshold , if a leaf signal is on the scale l If the correlation coefficient is lower than the set threshold, it is considered that the blade may have defects and an alarm is issued.

[0051] As an embodiment, the threshold is set: the threshold is set according to the wavelet coefficient characteristics of the normal leaf signal. , set the corresponding threshold .

[0052] Abnormal judgment: If the characteristics of a blade signal at a certain scale exceed the set threshold, it is considered that the blade may have defects.

[0053] For example, if the correlation of the wavelet coefficients, energy distribution, or peak position of a blade at a certain scale is significantly lower than that of other blades, it may indicate that there is a problem with the blade.

[0054] Setting alarm thresholds , an alarm is triggered when the correlation falls below this threshold: For any wavelet scale , , ,

[0055] Rule No. or The fan blade is defective.

[0056] See Figure 2 , Figure 2 4 is a schematic diagram of the working of the hardware device of an embodiment of the present invention, wherein the hardware device specifically comprises: a wavelet multi-scale wind turbine blade defect detection device 401 based on speed synchronization time difference correction, a processor 402 and a storage device 403.

[0057] A wavelet multi-scale fan blade defect detection device 401 based on speed synchronization time difference correction: The wavelet multi-scale fan blade defect detection device 401 based on speed synchronization time difference correction implements the wavelet multi-scale fan blade defect detection method based on speed synchronization time difference correction.

[0058] Processor 402: The processor 402 loads and executes instructions and data in the storage device 403 to implement the wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction.

[0059] Storage device 403: The storage device 403 stores instructions and data; the storage device 403 is used to implement the wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction.

[0060] The key points of the present invention are: 1. High-Precision Detection: Signal time difference is corrected using real-time rotational speed information, ensuring consistency when comparing signals at the same rotation angle, thereby improving detection accuracy and reliability. This approach overcomes the signal time difference issue caused by rotational speed fluctuations in traditional methods, improving the accuracy and reliability of signal comparison. Furthermore, the use of wavelet transforms to extract signal features at different scales provides a more comprehensive picture of the blade's state across different frequency ranges, enabling identification of defects of varying types and sizes. This approach can capture subtle changes in the blade at different frequencies, enhancing detection sensitivity.

[0061] 2. Real-time Monitoring: Utilizing the real-time speed information provided by the speed sensor, the signal can be corrected in real time under dynamic conditions, enabling instant monitoring of blade status. This method promptly captures any abnormal changes in the blade during operation, ensuring real-time detection results. Furthermore, through multi-scale feature extraction and correlation calculation, potential blade problems can be promptly identified and responded to quickly. This method can quickly identify and locate defects, improving the system's response speed and detection efficiency.

[0062] 3. Affordability and Ease of Use: Compared to high-cost technologies like fiber Bragg grating (FBG), the hardware required by this method is relatively simple and easy to deploy and maintain. This approach reduces the initial investment and maintenance costs of the system, improving economic efficiency. Furthermore, through standardized correlation calculations, the data analysis process is simplified, reducing the need for specialized personnel and improving the system's usability. This approach makes data analysis simpler and faster, lowering the technical barriers to entry.

[0063] 4. Adaptability to Complex Environments: Environmentally robust: Even in complex and changing natural environments, time difference correction and multi-scale analysis maintain stable detection results. This method can adapt to various harsh environmental conditions, ensuring the stability and reliability of detection results. It also offers diverse applicability: it is suitable for different types of wind turbines and diverse wind farm environments, demonstrating its broad applicability. This method can be widely applied to various types of wind turbines, enhancing the versatility and practicality of the technology.

[0064] In summary, the present invention significantly improves wind turbine blade health monitoring, reduces operation and maintenance costs, and ensures safe and stable wind turbine operation. This technology not only overcomes the limitations of traditional monitoring methods but also provides a highly accurate, real-time, and cost-effective solution. It is applicable to various complex environments and different types of wind turbines, and has broad application prospects.

[0065] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. 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 wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction, characterized by: The method comprises the following steps: S1. Install a pickup at the root of the fan blade; install a speed sensor on the fan shaft; S2, collecting and digitizing the acoustic signals of the fan blades; S3, using a speed sensor to eliminate the time difference of acoustic signals of different blades to obtain a correction signal; S4, extracting features from each correction signal to obtain a feature signal; S5. Obtaining correlation coefficients between characteristic signals; S6. Detect fan blade defects based on the correlation coefficient and generate an alarm.

2. The wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction according to claim 1 is characterized in that: Step S3 is as follows: S31. Obtain the real-time speed of the fan according to the speed sensor and calculate the fan rotation period; S32, determining the blade delay according to the fan rotation period; S33. Correct the signals generated by different blades according to the calculated time delay to obtain corrected signals.

3. The wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction according to claim 2 is characterized in that: Delay in step S32 The calculation formula is as follows: in, i is the number of the fan blade, t i For the i The time point at which the wind turbine blade pickup collects the sound, t 1 is the acquisition time point of the first fan blade, which is also the reference time point. r ( t 1) Real-time speed information of the fan.

4. The wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction according to claim 3 is characterized in that: The signal after correction in step S33 is as follows: in is the corrected signal; n The sampling point index.

5. The wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction according to claim 1 is characterized in that: Step S4 is specifically as follows: S41, perform discrete wavelet transform on the corrected signal to obtain different scales l The signal below; S42, for different scales l The feature extraction is performed on the signal under the wavelet to obtain a feature signal, wherein the feature signal includes: wavelet coefficients, energy distribution and peak position.

6. The wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction according to claim 1 is characterized in that: In step S5 , a cross-correlation function is used to calculate the correlation coefficient, and the correlation coefficient is further standardized.

7. The wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction according to claim 5, characterized in that: Step S6 is specifically as follows: setting a threshold value according to the wavelet coefficient characteristics of the normal leaf signal, for each scale , set the corresponding threshold ,If the correlation coefficient of a blade signal on scale l is lower than the set threshold, it is considered that the blade may have defects and an alarm is issued.

8. A storage device, characterized in that: The storage device stores instructions and data for implementing a wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction as described in any one of claims 1 to 7.

9. A wavelet multi-scale wind turbine blade defect detection device based on speed synchronization time difference correction, characterized by: include: Processor and storage device; the processor loads and executes instructions and data in the storage device to implement a wavelet multi-scale wind turbine blade defect detection method based on speed synchronization time difference correction as described in any one of claims 1 to 7.

Images