A wind turbine blade damage detection method based on a swept-frequency audio signal

Abstract

The application relates to the field of fan blade defect detection and discloses a wind turbine blade damage detection method based on a swept-frequency audio signal. s ( t ) is emitted; a tower base control system is installed at the bottom of a fan tower to acquire the received signal r ( t ); a signal synchronization mechanism between the audio transmitter and the tower base control system is constructed to synchronize the swept-frequency signal s ( t ) with the received signal r ( t ), and a filter is switched in advance for subsequent use; the signal r ( t ) is subjected to digital processing and filtering to obtain a filtered signal y [ n ]; the signal energy or peak value is calculated through the filtered signal y [ n ], so that whether the fan blade is damaged or not is judged; and the application can accurately and efficiently detect the health condition of the wind turbine blade.

Description

Technical Field

[0001] This invention relates to the field of wind turbine blade defect detection, and in particular to a method for detecting wind turbine blade damage based on swept frequency audio signals. Background Technology

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

[0003] Damage to wind turbine blades not only reduces the power generation efficiency and operational stability of wind turbines, but can also lead to safety accidents in severe cases, threatening the safety of people and property. Therefore, real-time monitoring and effective detection of the health status of wind turbine blades are of great significance 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 inspection, using visual checks or tapping to listen for sounds to determine if blades are damaged. However, this method is not only inefficient but also highly susceptible to weather and environmental factors, making real-time monitoring and accurate assessment difficult. Furthermore, with the continuous development of wind power technology, the length and weight of wind turbine blades are constantly increasing, further increasing the difficulty and danger of manual inspection. To overcome these limitations, developing an efficient, accurate, and non-contact blade damage detection method is particularly important. Summary of the Invention

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

[0006] Specifically, this invention provides a method for detecting damage to wind turbine blades based on swept-frequency audio signals, comprising the following steps:

[0007] S1. Install an audio transmitter with frequency sweep function inside the cavity of the wind turbine blade to emit a specified frequency sweep signal. s ( t );

[0008] S2. Install a tower base control system at the bottom of the wind turbine to acquire the received signals. r ( t );

[0009] S3. Construct a signal synchronization mechanism between the audio transmitter and the tower base control system to enable the sweep frequency signal s ( t ) and received signal r ( t Synchronization, while pre-switching filters for later use;

[0010] S4, for the signal r ( t The signal is then digitized and filtered to obtain the filtered signal. y [ n ];

[0011] S5. The filtered signal y [ n The signal energy or peak value is calculated to determine whether the wind turbine blades are damaged.

[0012] A storage device that stores instructions and data for implementing a wind turbine blade damage detection method based on swept frequency audio signals.

[0013] A wind turbine blade damage detection device based on swept-frequency audio signals includes: 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 swept-frequency audio signals.

[0014] The beneficial effects provided by this invention are as follows: by emitting an audio signal of a specific frequency within the blade cavity and receiving the leakage signal at the tower base, and using a digitally adjustable narrowband filter to filter out background noise, the presence of blade damage can be detected with a high signal-to-noise ratio. This method utilizes a synchronized center frequency to ensure effective signal detection, thereby achieving real-time monitoring of blade health. In summary, this invention can accurately and efficiently detect the health status of wind turbine blades. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the wind turbine blade damage detection method based on swept frequency audio signals according to the present invention;

[0016] Figure 2 This is a schematic diagram of the hardware device used in this application. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the present invention clearer, the embodiments of the present invention will be further described below with reference to the accompanying drawings.

[0018] Before formally describing the present invention, a general description of the solution of the present invention will be given first to facilitate understanding.

[0019] Please refer to Figure 1 The present invention provides a method for detecting damage to wind turbine blades based on swept-frequency audio signals, comprising the following steps:

[0020] S1. Install an audio transmitter with frequency sweep function inside the cavity of the wind turbine blade to emit a specified frequency sweep signal. s ( t );

[0021] It should be noted that the present invention installs an audio transmitter with frequency sweeping function inside the cavity of the wind turbine blade. This transmitter is capable of transmitting frequencies covering a certain frequency range. The audio signal.

[0022] Set the center frequency of the sweep signal to be The sweep bandwidth is Then the frequency sweep signal can be expressed as:

[0023] in, It is the signal amplitude. It is a frequency function that varies with time, and is usually expressed using a linear frequency sweep:

[0024]

[0025] It is the frequency sweep cycle.

[0026] The purpose of frequency sweeping in this invention is as follows:

[0027] Wider frequency coverage: Frequency sweeping can cover a wider frequency range, which helps to detect defects that are sensitive to specific frequencies.

[0028] Improve detection sensitivity: By changing the frequency, the difference in the material's reaction at different frequencies can be detected, thereby improving the accuracy of the detection.

[0029] Adapting to different material properties: Different materials have different absorption and reflection characteristics for sound waves of different frequencies. Frequency sweeping can ensure that the detection process can adapt to these changes, making the detection results more reliable.

[0030] Avoid resonance effect: Frequency sweeping can prevent the system from resonating at a specific frequency, which could lead to misjudgment or unstable detection results.

[0031] S2. Install a tower base control system at the bottom of the wind turbine to acquire the received signals. r ( t );

[0032] It should be noted that when the blades are in good condition, the audio signal... It will almost never penetrate the blade material to reach the external environment. If the blade is damaged or cracked, the audio signal will leak out at the point of damage and propagate into the outside space.

[0033] A broadband sound sensor (pickup unit) is installed on the base of the wind turbine to receive audio signals leaking from the blade cavity. The signal received by the pickup unit... Includes background noise and target signal , can be represented as: .

[0034] S3. Construct a signal synchronization mechanism between the audio transmitter and the tower base control system to enable the sweep frequency signal s ( t ) and received signal r ( t Synchronization, while pre-switching filters for later use;

[0035] It should be noted that step S3 is as follows:

[0036] S31. Obtain the periodic sweep sequence of the audio transmitter. Where N is the total number of frequency points, and each frequency point Corresponding to a fixed sweep bandwidth ;

[0037] S32. The audio transmitter transmits one frequency point within each time window and sends the current frequency point number to the tower-based control system via coded communication. The tower base control system includes: a microcontroller unit or a dedicated integrated circuit unit, a microphone, and a digitally adjustable narrowband filter module.

[0038] S33. In the tower base control system, the microcontroller unit or application-specific integrated circuit receives the synchronization control signal from the audio transmitter and parses the number of the current frequency point of the transmitter from it. Then, based on the number, a pre-built frequency-filter parameter lookup table is consulted to obtain the corresponding center frequency. and sweep bandwidth ;

[0039] S34. Load the corresponding filter coefficients into the digitally adjustable narrowband filter module to give it passband characteristics that match the transmitted signal;

[0040] S35. Load the corresponding filter coefficients into the digitally adjustable narrowband filter module to give it passband characteristics that match the transmitted signal.

[0041] As one example, in order to ensure that the tower-based microphone can accurately identify and extract the audio signal leaking from the blade cavity, while effectively suppressing environmental noise interference, a high-precision synchronization mechanism between the transmitter and receiver must be established.

[0042] This mechanism is one of the key technologies for achieving non-contact, high-sensitivity blade damage detection in this invention.

[0043] In this invention, the audio transmitter is installed inside the wind turbine blade. Its output audio signal is not randomly swept, but rather periodically scanned according to a preset sequence of discrete center frequency points. This sequence is denoted as:

[0044] Where N is the total number of frequency points, and each frequency point Corresponding to a fixed sweep bandwidth ,Right now:

[0045] Center frequency With sweep bandwidth It is a fixed combination with a one-to-one correspondence.

[0046] The transmitter emits one frequency point within each time window and sends the current frequency point number to the tower-based control system via coded communication. In addition, the transmitter can dynamically adjust the following parameters according to the current operating status: transmit power level P, frequency sweep period T;

[0047] Therefore, the synchronization control signal contains at least the following information:

[0048] Current frequency point number (Used for table lookup) and Current transmit power level P; Current frequency sweep period T;

[0049] To improve anti-interference capabilities, shielded cables or photoelectric conversion modules can be used for signal transmission in communication links to ensure that data can be transmitted stably and reliably even in complex electromagnetic environments.

[0050] At the base of the tower, a microcontroller unit (MCU) or application-specific integrated circuit (ASIC) receives the synchronization control signal from the transmitter and parses the number of the current frequency point of the transmitter from it. Subsequently, the system looks up the corresponding center frequency in a pre-built frequency-filter parameter lookup table based on the given number. and sweep bandwidth Key parameters, etc.

[0051] Since the transmitting frequency and bandwidth are preset fixed combinations, the receiving end only needs to obtain the corresponding number i to quickly obtain the corresponding value.f center ( i ) and B ( i ), and set the passband characteristics of the narrowband filter accordingly.

[0052] Once the current frequency point number is determined The system then loads the corresponding filter coefficients into the digital narrowband filter module, giving it passband characteristics that match the transmitted signal. The filter can adopt an FIR or IIR structure, supporting fast switching and low-latency updates.

[0053] It should be noted that, to address potential communication delays, temperature drift, or local oscillator errors, the following auxiliary mechanisms can be introduced:

[0054] Time-domain delay compensation: Corrects the synchronization time difference by measuring the basic transmission delay and combining it with temperature sensor data;

[0055] Frequency domain feature matching: Perform FFT analysis on the received signal and compare it with a standard template to verify whether it is a valid transmitted signal;

[0056] Phase-locked loop-assisted tracking: When necessary, the digital phase-locked loop (DPLL) mechanism is activated to further improve synchronization accuracy.

[0057] S4, for the signal r ( t The signal is then digitized and filtered to obtain the filtered signal. y [ n ];

[0058] It should be noted that the signal received by the microphone is processed digitally and converted into a digital signal. A digitally adjustable narrowband filter is used to remove background noise and preserve the target signal. The center frequency of the narrowband filter... Synchronized with the center frequency of the audio transmitter. A digital narrowband filter can be represented as:

[0059]

[0060] in:

[0061] Bandwidth control factor Normalized center frequency, , Passband ripple factor (preset constant) 0.5dB; System sampling rate (preset constant), 16kHz;

[0062] Its parameters are dynamically related, as detailed below:

[0063] Center frequency synchronization: obtained through a synchronization mechanism ,calculate .

[0064] Bandwidth adaptive: based on the swept bandwidth B ( i Adjust the bandwidth control factor As shown in the following formula:

[0065]

[0066] As one embodiment, the present invention provides an example: when =1kHz, B ( i When ) = 30Hz:

[0067] =0.3927 rad, =0.127, generation coefficient: .

[0068] S5. The filtered signal y [ n The signal energy or peak value is calculated to determine whether the wind turbine blades are damaged.

[0069] Step S5 is as follows: If the calculated energy or peak Exceeded the set threshold or This indicates that the leaf may be damaged.

[0070] Specifically, the signal after narrowband filtering It can be represented as:

[0071]

[0072] Determining whether a blade is damaged by calculating the energy or peak value of the signal:

[0073]

[0074] or:

[0075]

[0076] To accurately determine whether a blade is damaged, a reasonable threshold needs to be set. The threshold is typically set based on the following steps:

[0077] Historical data analysis:

[0078] For a known intact blade, collect signal data over a period of time and calculate its average energy. and standard deviation .

[0079] For blades with known damage, signal data is collected and their average energy is calculated. .

[0080] Threshold calculation:

[0081] Based on intact blade data, an energy threshold was set. for:

[0082]

[0083] in, It is an empirical factor, usually taking a value of 2 or 3 to ensure a high level of confidence.

[0084] Similarly, for peak Set threshold for:

[0085]

[0086] in, and These are the mean peak value and standard deviation of intact leaf data, respectively.

[0087] Determine if damaged:

[0088] If the calculated energy or peak Exceeded the set threshold or This indicates that the leaf may be damaged.

[0089] A continuous detection mechanism can be further set up, that is, damage is only confirmed when multiple detections exceed the threshold, in order to reduce the possibility of false alarms.

[0090] Using the above method, the present invention can accurately detect sound signals leaking from blade cracks and pores, and determine whether the blade is damaged by setting a reasonable threshold, thereby realizing real-time monitoring of the health status of wind turbine blades.

[0091] Please see Figure 2 , Figure 2 This is a schematic diagram of the hardware device in operation according to an embodiment of the present invention. The hardware device specifically includes: a wind turbine blade damage detection device 401 based on swept frequency audio signals, a processor 402, and a storage device 403.

[0092] A wind turbine blade damage detection device 401 based on swept frequency audio signals: The wind turbine blade damage detection device 401 based on swept frequency audio signals implements the wind turbine blade damage detection method based on swept frequency audio signals.

[0093] 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.

[0094] 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.

[0095] The key point of this invention is:

[0096] Frequency-sweeping audio signal transmission technology: This invention is the first to apply a frequency-sweeping audio transmitter inside the cavity of a wind turbine blade. By transmitting audio signals covering a certain frequency range, not only is the detection sensitivity improved, but it can also adapt to different material properties and effectively avoid resonance effects. The frequency-sweeping technology ensures that more information related to the blade's health condition is captured during the detection process, significantly improving the accuracy and reliability of the detection.

[0097] Synchronization Mechanism and Narrowband Filtering Technology: This invention designs a precise synchronization mechanism to ensure accurate synchronization between the audio transmitter and the pickup on the tower base. Through wired transmission and photoelectric conversion technology, the center frequency and sweep parameters of the transmitter are transmitted to the tower base in real time, and the parameters of the narrowband filter are dynamically adjusted. This synchronization mechanism, combined with narrowband filtering technology, effectively filters out background noise, preserves the target signal, and improves the efficiency and accuracy of signal processing.

[0098] Intelligent Threshold Setting and Judgment Mechanism: This invention proposes an intelligent threshold setting method based on historical data analysis. By collecting signal data from known intact and damaged blades, the average energy and standard deviation are calculated, and a reasonable threshold is set accordingly. Simultaneously, a continuous detection mechanism is implemented to reduce the possibility of false alarms. This intelligent threshold setting and judgment mechanism can accurately determine whether a blade is damaged, improving the accuracy and reliability of detection and reducing the false alarm rate.

[0099] In summary, the beneficial effects of this invention are: by emitting an audio signal of a specific frequency within the blade cavity and receiving the leakage signal at the tower base, and using a digitally adjustable narrowband filter to filter out background noise, the presence of blade damage can be detected with a high signal-to-noise ratio. This method utilizes a synchronized center frequency to ensure effective signal detection, thereby achieving real-time monitoring of blade health. Overall, this invention can accurately and efficiently detect the health status of wind turbine blades.

[0100] 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 within the protection scope of the present invention.

Claims

1. A method for detecting damage to wind turbine blades based on swept-frequency audio signals, characterized in that: The method includes the following steps: S1. Install an audio transmitter with frequency sweep function inside the cavity of the wind turbine blade to emit a specified frequency sweep signal. s ( t ); S2. Install a tower base control system at the bottom of the wind turbine to acquire the received signals. r ( t ); S3. Construct a signal synchronization mechanism between the audio transmitter and the tower base control system to enable the sweep frequency signal s ( t ) and received signal r ( t Synchronization, while pre-switching filters for later use; S4, for the signal r ( t The signal is then digitized and filtered to obtain the filtered signal. y [ n ]; S5. The filtered signal y [ n ] Calculate the signal energy or peak value to determine whether the wind turbine blades are damaged; In step S3, the signal synchronization mechanism can also be adapted to perform signal compensation synchronization in the form of time-domain delay compensation, frequency characteristic matching, or phase-locked loop assisted tracking. Step S5 is as follows: If the calculated energy or peak Exceeded the set threshold or This indicates that the leaf may be damaged; The blade is confirmed as damaged only when the average value of multiple tests exceeds a threshold, through a continuous multi-test mechanism.

2. The method for detecting wind turbine blade damage based on swept-frequency audio signals as described in claim 1, characterized in that: The signal received in step S2 r ( t Includes background noise n ( t ) and target sweep frequency signal s ( t The specific formula is as follows: .

3. The method for detecting wind turbine blade damage based on swept-frequency audio signals as described in claim 1, characterized in that: Step S4 is as follows: S41, will receive signal r ( t Digital processing yields digital signals. ; S42, For digital signals A digitally adjustable narrowband filter is used to remove background noise, resulting in a filtered signal. y [ n ]: in, H ( z () is a digitally adjustable narrowband filter. Indicates bandwidth control factor, Represents the normalized center frequency. According to the center frequency and system sampling rate Adjustment, parameters , It is dynamically related, as shown in the following formula: Among them, the sweep bandwidth is B ( i ), This is the passband ripple factor.

4. A storage device, characterized in that: The storage device stores instructions and data to implement the wind turbine blade damage detection method based on swept frequency audio signals as described in any one of claims 1 to 3.

5. A wind turbine blade damage detection device based on swept-frequency audio signals, characterized in that: include: A processor and a storage device; the processor loads and executes instructions and data in the storage device to implement the wind turbine blade damage detection method based on swept frequency audio signals as described in any one of claims 1 to 3.

Images