Wind driven generator blade damage detection method based on sweep frequency audio signals

Abstract

The invention relates to the field of fan blade defect detection, and discloses a wind driven generator blade damage detection method based on a frequency sweep audio signal, and the method comprises the steps: installing an audio transmitter with a frequency sweep function in a cavity of a fan blade, and transmitting a specified frequency sweep signal s (t); a tower footing control system is installed on a tower at the bottom of the draught fan, and a received signal r (t) is obtained; a signal synchronization mechanism between the audio transmitter and the tower footing control system is constructed, so that the sweep frequency signal s (t) and the received signal r (t) are synchronized, and meanwhile, a filter is switched in advance for subsequent use; performing digital processing and filtering on the signal r (t) to obtain a filtered signal y [n]; calculating signal energy or a peak value through the filtered signal y [n] so as to judge whether the fan blade is damaged or not; according to the invention, the health condition of the wind driven generator blade can be accurately and efficiently detected.

Description

Technical Field

[0001] The present invention relates to the field of wind turbine blade defect detection, and in particular to a wind turbine blade damage detection method based on a swept-frequency audio signal. Background Art

[0002] With the growing global demand for renewable energy, wind power, as a clean, renewable energy source, is becoming increasingly important. As the core equipment in wind power generation systems, the operating efficiency and safety of wind turbines are directly related to the economic benefits and ecological environment of the entire wind farm. However, during operation, wind turbine blades, as key components for capturing wind energy, are exposed to harsh natural environments such as strong winds, dust, and salt spray, making them extremely susceptible to damage.

[0003] Blade damage not only reduces wind turbine power generation efficiency and operational stability, but in severe cases can even cause safety accidents, threatening the safety of people and property. Therefore, real-time monitoring and effective testing of the health of wind turbine blades are crucial for ensuring the normal operation of wind farms and extending the service life of wind turbines.

[0004] Currently, traditional blade inspection methods mostly rely on manual inspections, using visual inspections or tapping and listening to determine if blades are damaged. However, this method is not only inefficient but also significantly affected by weather and environmental factors, making real-time monitoring and accurate judgment difficult. Furthermore, with the continuous advancement of wind power technology, the length and weight of wind turbine blades are increasing, making manual inspections more difficult and dangerous. To overcome these limitations, it is crucial to develop an efficient, accurate, and non-contact blade damage detection method. Summary of the Invention

[0005] The purpose of the present invention is to propose a wind turbine blade damage detection method based on a swept frequency audio signal to solve the technical problem that most of the existing wind turbine blade defect detection is still performed manually, resulting in low detection efficiency and low accuracy.

[0006] Specifically, the present invention provides a method for detecting wind turbine blade damage based on a swept frequency audio signal, comprising the following steps: S1. Install an audio transmitter with sweep frequency function in the cavity of the fan blade to send out a specified sweep frequency signal. s ( t ); S2. Install the tower base control system at the bottom of the wind turbine tower to obtain the received signal r ( t ); S3. Build a signal synchronization mechanism between the audio transmitter and the tower base control system so that the sweep frequency signal s ( t ) and receiving signals r ( t ) synchronization, and pre-switching the filter for subsequent use; S4, signal r ( t ) to perform digital processing and filtering to obtain the filtered signal y [ n ]; S5, the signal after filtering y [ n ], calculate the signal energy or peak value to determine whether the fan blade is damaged.

[0007] A storage device stores instructions and data for implementing a wind turbine blade damage detection method based on a swept-frequency audio signal.

[0008] A wind turbine blade damage detection device based on a swept-frequency audio signal comprises: a processor and a storage device; the processor loads and executes instructions and data in the storage device to implement a wind turbine blade damage detection method based on a swept-frequency audio signal.

[0009] The present invention provides the following beneficial effects: by transmitting an audio signal of a specific frequency within the blade cavity, receiving the leakage signal at the tower base, and filtering out background noise using a digitally adjustable narrowband filter, blade damage can be detected with a high signal-to-noise ratio. This method utilizes a synchronized center frequency to ensure effective signal detection, enabling real-time monitoring of blade health. Overall, the present invention can accurately and efficiently detect the health of wind turbine blades. BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Figure 1 1. It is a flow chart of a wind turbine blade damage detection method based on a swept-frequency audio signal according to 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 1The present invention provides a wind turbine blade damage detection method based on a swept frequency audio signal, comprising the following steps: S1. Install an audio transmitter with sweep frequency function in the cavity of the fan blade to send out a specified sweep frequency signal. s ( t ); It should be noted that the present invention installs an audio transmitter with a frequency sweep function in the cavity of the wind turbine blade, and the transmitter can emit audio covering a certain frequency range. audio signal.

[0014] Set the center frequency of the sweep signal to , the sweep bandwidth is , then the frequency sweep signal can be expressed as:

[0015] in, is the signal amplitude, is a frequency function that varies with time, usually using a linear frequency sweep:

[0016] is the frequency sweep period.

[0017] The purpose of frequency sweeping in the present invention is as follows: Covering a wider frequency range: Sweeping frequency can cover a wider frequency range, which helps detect defects that are sensitive to specific frequencies.

[0018] Improve detection sensitivity: By changing the frequency, the response differences of materials at different frequencies can be discovered, thereby improving the accuracy of detection.

[0019] Adapting to different material properties: Different materials have different absorption and reflection characteristics for sound waves of different frequencies. Sweeping frequency ensures that the detection process can adapt to these changes, making the test results more reliable.

[0020] Avoid resonance effect: Frequency sweeping can prevent the system from operating at a specific frequency and causing resonance, which may lead to misjudgment or unstable detection results.

[0021] S2. Install the tower base control system at the bottom of the wind turbine tower to obtain the received signal r ( t ); It should be noted that when the blade is in good condition, the audio signal The audio signal will hardly penetrate the blade material to reach the outside environment. If the blade is damaged or cracked, the audio signal will leak out at the damaged part and spread to the outside space.

[0022] A broadband sound sensor (microphone) is installed on the bottom tower of the wind turbine to receive the audio signal leaking from the blade cavity. Includes background noise and target signal , which can be expressed as: .

[0023] S3. Build a signal synchronization mechanism between the audio transmitter and the tower base control system so that the sweep frequency signal s ( t ) and receiving signals r ( t ) synchronization, and pre-switching the filter for subsequent use; It should be noted that step S3 is specifically as follows: S31. Obtain the periodic frequency sweep sequence of the audio transmitter , where N is the total number of frequency points, each frequency point Corresponding to a fixed sweep bandwidth ; S32. The audio transmitter transmits one of the frequency points in each time window and sends the number of the current frequency point to the tower base control system through coded communication. The tower base control system includes: a microcontroller unit or a dedicated integrated circuit unit, a pickup, and a digitally adjustable narrowband filter module; S33. In the tower base control system, the microcontroller unit or dedicated integrated circuit receives the synchronous control signal from the audio transmitter and parses it to obtain the number of the frequency point where the transmitter is currently located. , and find the pre-built frequency-filter parameter comparison table according to the number to obtain the corresponding center frequency and swept bandwidth ; S34, loading the corresponding filter coefficients into the digital adjustable narrowband filter module so that it has a passband characteristic that matches the transmitted signal; S35. Load the corresponding filter coefficients into the digitally adjustable narrowband filter module so that it has a passband characteristic that matches the transmitted signal.

[0024] As an example, in order to ensure that the tower base end pickup can accurately identify and extract the audio signal leaking from the blade cavity while effectively suppressing environmental noise interference, a high-precision synchronization mechanism must be established between the transmitter and receiver.

[0025] This mechanism is one of the key technologies for realizing non-contact, high-sensitivity blade damage detection in the present invention.

[0026] In the present invention, the audio transmitter is installed inside the wind turbine blade. The audio signal it outputs is not randomly swept, but periodically scanned according to a set of preset discrete center frequency point sequences. This sequence is recorded as:

[0027] Where N is the total number of frequency points, each frequency point Corresponding to a fixed sweep bandwidth ,Right now: Center frequency and swept bandwidth It is a fixed combination with one-to-one correspondence.

[0028] The transmitter transmits one of the frequency points in each time window and sends the number of the current frequency point to the tower base control system through coded communication. In addition, the transmitter can also dynamically adjust the following parameters according to the current working status: transmission power level P, sweep frequency period T; Therefore, the synchronization control signal contains at least the following information: Current frequency point number (Used for table lookup and ; Current transmit power level P; Current sweep period T; To improve anti-interference capabilities, communication links can use shielded cables or optoelectronic conversion modules for signal transmission to ensure that data can still be transmitted stably and reliably in complex electromagnetic environments.

[0029] At the tower base, the microcontroller unit (MCU) or application-specific integrated circuit (ASIC) receives the synchronization control signal from the transmitter and parses it to determine the frequency point number of the transmitter. Then, the system searches the pre-built frequency-filter parameter comparison table according to the number and obtains the corresponding center frequency. and swept bandwidth And other key parameters.

[0030] Since the transmitter frequency and bandwidth are both preset fixed combinations, the receiver can quickly obtain the corresponding f center ( i ) and B ( i ), and set the passband characteristics of the narrowband filter accordingly.

[0031] Once the current frequency point number is determined , the system loads the corresponding filter coefficients into the digital narrowband filter module, giving it a passband characteristic that matches the transmitted signal. The filter can use FIR or IIR structures, supporting fast switching and low-latency updates.

[0032] It should be noted that to address possible communication delays, temperature drift, or local oscillator errors, the following auxiliary mechanisms can also be introduced: Time domain delay compensation: By measuring the basic transmission delay and combining it with temperature sensor data, the synchronization time difference is corrected; Frequency domain feature matching: Perform FFT analysis on the received signal and compare it with the standard template to verify whether it is a valid transmitted signal; Phase-locked loop-assisted tracking: Enables the digital phase-locked loop (DPLL) mechanism when necessary to further improve synchronization accuracy.

[0033] S4, signal r ( t ) to perform digital processing and filtering to obtain the filtered signal y [ n ]; It should be noted that the signal received by the pickup is converted into a digital signal through digital processing. A digital adjustable narrowband filter is used to filter out background noise and retain the target signal. The center frequency of the narrowband filter is Synchronized to the center frequency of the audio transmitter. The digital narrowband filter can be expressed as:

[0034] in: : Bandwidth control factor, : normalized center frequency, 、 :Passband ripple coefficient (preset constant) 0.5dB; : System sampling rate (preset constant), 16kHz; Its parameters are dynamically associated, as follows: Center frequency synchronization: obtained through synchronization mechanism ,calculate .

[0035] Bandwidth adaptation: according to the sweep bandwidth B ( i ) Adjust the bandwidth control factor As follows:

[0036] As an embodiment, the present invention provides an example: when =1kHz, B ( i ) =30Hz: =0.3927rad, =0.127, generating coefficient: .

[0037] S5, the signal after filtering y [ n ], calculate the signal energy or peak value to determine whether the fan blade is damaged.

[0038] Step S5 is as follows: If the calculated energy or peak Exceeded the set threshold or , it indicates that the leaf may be damaged.

[0039] Specifically, the signal processed by the narrowband filter It can be expressed as:

[0040] By calculating the energy or peak value of the signal, it is determined whether the blade is damaged:

[0041] or:

[0042] In order to accurately determine whether the blade is damaged, a reasonable threshold needs to be set. The threshold setting is usually based on the following steps: Historical data analysis: For a known intact blade, collect signal data over a period of time and calculate its average energy and standard deviation .

[0043] For blades with known damage, signal data is also collected and the average energy is calculated. .

[0044] Threshold calculation: Set energy threshold based on intact blade data for:

[0045] in, is an empirical factor, usually taken as 2 or 3 to ensure a high confidence level.

[0046] Similarly, for the peak , set the threshold for:

[0047] in, and are the mean peak value and standard deviation of intact leaf data, respectively.

[0048] Determine damage: If the calculated energy or peak Exceeded the set threshold or , it indicates that the leaf may be damaged.

[0049] A continuous detection mechanism can be further set up, that is, multiple detections must exceed the threshold before it is confirmed to be damaged, so as to reduce the possibility of false alarms.

[0050] Through the above method, the present invention can accurately detect the sound signals leaked from the cracks and pores of the blades, and determine whether the blades are damaged by setting a reasonable threshold, thereby realizing real-time monitoring of the health status of the wind turbine blades.

[0051] 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 wind turbine blade damage detection device 401 based on a swept frequency audio signal, a processor 402 and a storage device 403.

[0052] A wind turbine blade damage detection device 401 based on a swept-frequency audio signal: The wind turbine blade damage detection device 401 based on a swept-frequency audio signal implements the wind turbine blade damage detection method based on a swept-frequency audio signal.

[0053] Processor 402: The processor 402 loads and executes the instructions and data in the storage device 403 to implement the wind turbine blade damage detection method based on swept frequency audio signals.

[0054] Storage device 403: The storage device 403 stores instructions and data; the storage device 403 is used to implement the wind turbine blade damage detection method based on swept frequency audio signals.

[0055] The key points of the present invention are: Sweep-Frequency Audio Signal Transmission Technology: This invention utilizes a sweep-frequency audio transmitter within the wind turbine blade cavity for the first time. By transmitting audio signals covering a specific frequency range, it not only improves detection sensitivity but also adapts to different material properties, effectively avoiding resonance effects. This sweep-frequency technology ensures that more information relevant to blade health is captured during the detection process, significantly improving detection accuracy and reliability.

[0056] Synchronization Mechanism and Narrowband Filtering Technology: This invention incorporates a sophisticated synchronization mechanism to ensure accurate synchronization between the audio transmitter and the microphone on the tower base. Using wired transmission and optoelectronic conversion technology, the transmitter's center frequency and sweep parameters are transmitted to the tower base in real time, dynamically adjusting the narrowband filter parameters. This synchronization mechanism, combined with narrowband filtering technology, effectively filters out background noise while preserving the target signal, improving signal processing efficiency and accuracy.

[0057] Intelligent Threshold Setting and Judgment Mechanism: This invention proposes an intelligent threshold setting method based on historical data analysis. This method collects signal data from known intact and damaged blades, calculates their average energy and standard deviation, and uses this data to set appropriate thresholds. Furthermore, a continuous detection mechanism is implemented to reduce the possibility of false alarms. This intelligent threshold setting and judgment mechanism accurately determines whether a blade is damaged, improving detection accuracy and reliability while reducing false alarm rates.

[0058] In summary, the present invention achieves the following beneficial effects: by transmitting an audio signal of a specific frequency within the blade cavity, receiving the leakage signal at the tower base, and filtering out background noise using a digitally adjustable narrowband filter, blade damage can be detected with a high signal-to-noise ratio. This method utilizes a synchronized center frequency to ensure effective signal detection, enabling real-time monitoring of blade health. Overall, the present invention can accurately and efficiently detect the health of wind turbine blades.

[0059] 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 wind turbine blade damage detection method based on a swept frequency audio signal, characterized by: The method comprises the following steps: S1. Install an audio transmitter with sweep frequency function in the cavity of the fan blade to send out a specified sweep frequency signal. s ( t ); S2. Install the tower base control system at the bottom of the wind turbine tower to obtain the received signal r ( t ); S3. Build a signal synchronization mechanism between the audio transmitter and the tower base control system so that the sweep frequency signal s ( t ) and receiving signals r ( t ) synchronization, and pre-switching the filter for subsequent use; S4, signal r ( t ) to perform digital processing and filtering to obtain the filtered signal y [ n ]; S5, the signal after filtering y [ n ], calculate the signal energy or peak value to determine whether the fan blade is damaged.

2. The wind turbine blade damage detection method based on a swept frequency audio signal according to claim 1, characterized in that: The signal received in step S2 r ( t ) including background noise n ( t ) and target sweep signal s ( t ), specifically as follows: .

3. The wind turbine blade damage detection method based on a swept frequency audio signal according to claim 1, characterized in that: Step S3 is as follows: S31. Obtain the periodic frequency sweep sequence of the audio transmitter , where N is the total number of frequency points, each frequency point Corresponding to a fixed sweep bandwidth ; S32. The audio transmitter transmits one of the frequency points in each time window and sends the number of the current frequency point to the tower base control system through coded communication. The tower base control system includes: a microcontroller unit or a dedicated integrated circuit unit, a pickup, and a digitally adjustable narrowband filter module; S33. In the tower base control system, the microcontroller unit or dedicated integrated circuit receives the synchronous control signal from the audio transmitter and parses it to obtain the number of the frequency point where the transmitter is currently located. , and find the pre-built frequency-filter parameter comparison table according to the number to obtain the corresponding center frequency and swept bandwidth ; S34, loading the corresponding filter coefficients into the digital adjustable narrowband filter module so that it has a passband characteristic that matches the transmitted signal; S35. Load the corresponding filter coefficients into the digitally adjustable narrowband filter module so that it has a passband characteristic that matches the transmitted signal.

4. The wind turbine blade damage detection method based on a swept frequency audio signal according to claim 3, characterized in that: In step S35, the digitally adjustable narrowband filter adopts an FIR or IIR structure to support fast switching and low-latency updating.

5. The wind turbine blade damage detection method based on a swept frequency audio signal according to claim 1, characterized in that: The signal synchronization mechanism in step S3 may also be implemented in the form of time domain delay compensation, frequency feature matching, or phase-locked loop assisted tracking to perform adaptive signal compensation synchronization.

6. The wind turbine blade damage detection method based on a swept frequency audio signal according to claim 1, characterized in that: Step S4 is specifically as follows: S41, sweep frequency signal s ( t ) digital processing to obtain digital signal ; S42, digital signal A digital adjustable narrowband filter is used to filter out background noise and obtain the filtered signal. y [ n ]: in, H ( z ) is a digitally adjustable narrowband filter, represents the bandwidth control factor, represents the normalized center frequency, , according to the center frequency and system sampling rate Adjustment, parameters 、 For dynamic association, the specific formula is as follows: The sweep bandwidth is B ( i ), is the passband ripple coefficient.

7. The wind turbine blade damage detection method based on a swept frequency audio signal according to claim 1, characterized in that: Step S5 is as follows: If the calculated energy or peak Exceeded the set threshold or , it indicates that the leaf may be damaged.

8. The wind turbine blade damage detection method based on a swept frequency audio signal according to claim 7, characterized in that: Through a continuous multiple detection mechanism, the blade is confirmed to be damaged only when the average value of multiple detections exceeds the threshold.

9. A storage device, characterized in that: The storage device stores instructions and data for implementing a wind turbine blade damage detection method based on a swept frequency audio signal as described in any one of claims 1 to 8.

10. A wind turbine blade damage detection device based on a swept frequency audio signal, characterized by: include: A processor and a storage device; the processor loads and executes instructions and data in the storage device to implement a wind turbine blade damage detection method based on a swept frequency audio signal as described in any one of claims 1 to 8.

Images