Terahertz analysis-based wind turbine blade fault detection method

Abstract

The application discloses a wind turbine blade fault detection method based on terahertz analysis, relates to the technical field of blade fault detection, and is used for solving the problem that internal defects such as micro cracks or delamination are difficult to be found in time. The running state is evaluated by monitoring the rotating speed of a wind turbine blade. The reflection signal and the incident signal of each blade are collected when the wind turbine blade is running at a low speed, and the terahertz reflectance is calculated. The blade abnormality is judged according to the reflectance, and the marked blade is detected. The reflection delay and the penetration rate are used to generate structure evaluation characteristics. The marked blade is divided into different regions. The absorption peak shift and the peak width value of each divided region are detected. The peak width change trend is evaluated according to the peak width value. The dielectric disturbance coefficient of the divided region is calculated by combining the absorption peak shift, so as to judge whether the divided region has a fault and generate an alarm signal. The high penetration characteristic of the terahertz wave is used to realize high-precision detection of the surface and internal defects of the blade.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of blade fault detection, and more particularly to a wind turbine blade fault detection method based on terahertz analysis. BACKGROUND

[0002] As an important part of renewable energy, the operation safety and power generation efficiency of wind power generation directly depend on the integrity of wind turbine blades. However, wind turbine blades are prone to cracks, delamination, cavities and other structural defects due to wind load, environmental factors and material aging during long-term operation. Traditional blade detection methods mainly include ultrasonic detection and acoustic emission detection.

[0003] The prior art has the following disadvantages:

[0004] Currently, the prior art mainly relies on visual inspection, ultrasonic detection and acoustic emission detection, which cannot quickly and comprehensively obtain blade surface and internal defect information under the running state of the wind turbine blade, lacks high-precision quantitative evaluation means for blade structural abnormalities, and thus it is difficult to discover internal defects such as micro-cracks or delamination in time, and the blade fault diagnosis efficiency is low. Therefore, a wind turbine blade fault detection method based on terahertz analysis is proposed.

[0005] The above information disclosed in the background section is only used to enhance the understanding of the background of the present disclosure, and therefore it can include information that does not constitute prior art known to those of ordinary skill in the art. SUMMARY

[0006] In order to overcome the above-mentioned defects of the prior art, the embodiments of the present application provide a wind turbine blade fault detection method based on terahertz analysis, which collects terahertz reflection signals on the surface and inside of the blade under the low-speed running state of the wind turbine blade, and calculates the dielectric disturbance coefficient by combining reflection delay time, penetration rate, absorption peak shift and peak width value to solve the problems proposed in the above background.

[0007] To achieve the above-mentioned purpose, the present application provides the following technical scheme, a wind turbine blade fault detection method based on terahertz analysis, comprising the following steps:

[0008] Step S1: Monitor the blade speed of the wind turbine generator, evaluate the running state of the wind turbine generator according to the blade speed, and scan the surface of the blade by a terahertz emission device when the running state is low-speed running, and collect the reflection signal intensity and incident signal intensity of each blade;

[0009] Step S2: Calculate the terahertz reflectance by integrating the reflection signal intensity and the incident signal intensity, determine whether the blade is abnormal according to the terahertz reflectance, mark the blade with abnormality, and detect the reflection delay time length of the marked blade;

[0010] Step S3: Obtain the penetration rate of the terahertz emission device, generate the structural evaluation characteristics of the marked blade in combination with the reflection delay time length, divide the marked blade into different regions based on the structural evaluation characteristics, and detect the absorption peak shift and peak width value of each divided region;

[0011] Step S4: Evaluate the peak width trend according to the peak width value, calculate the dielectric disturbance coefficient of the divided region in combination with the absorption peak shift, and determine whether the divided region has a fault and generate an alarm signal according to the dielectric disturbance coefficient.

[0012] In a preferred embodiment, in step S1, the blade rotation speed of the wind turbine is monitored in real time by using a blade rotation speed monitoring device installed at the main shaft of the wind turbine;

[0013] The running state of the wind turbine is determined according to the blade rotation speed:

[0014] When the blade rotation speed is greater than the preset rotation speed threshold, it is determined that the wind turbine is in a high-speed running state;

[0015] When the blade rotation speed is less than or equal to the preset rotation speed threshold, it is determined that the wind turbine is in a low-speed running state.

[0016] In a preferred embodiment, in step S1, when the running state is detected to be a low-speed running state, the surface of the blade is scanned by a terahertz emission device, and the reflection signal intensity and the incident signal intensity of each scanning point on the blade are collected;

[0017] The incident signal intensity is the signal intensity formed by the terahertz electromagnetic wave emitted by the terahertz emission device vertically irradiating the surface of the wind turbine blade;

[0018] The reflection signal intensity is the signal intensity received by the terahertz receiving probe after the terahertz electromagnetic wave is reflected at the surface and internal interface of the blade.

[0019] In a preferred embodiment, in step S2, the ratio of the incident signal intensity and the reflection signal intensity of each scanning point is calculated to obtain the terahertz reflectance of each scanning point;

[0020] The terahertz reflectance of each scanning point is compared with the preset standard reflectance, and the absolute value of the difference is obtained to obtain the reflectance deviation index;

[0021] When the terahertz reflectance of each scanning point is greater than or equal to the preset deviation threshold, it is determined that the scanning point has an abnormal signal;

[0022] When the terahertz reflectance of each scanning point is less than a preset deviation threshold, it is determined that the scanning point does not exist an abnormal signal.

[0023] In a preferred embodiment, in step S2, the number of scanning points with abnormal signals on each leaf is counted to obtain an abnormal scanning point number;

[0024] The average of the abnormal scanning point numbers of all the leaves is calculated as an abnormal leaf determination threshold;

[0025] The leaf with the number of abnormal signals greater than the abnormal leaf determination threshold is determined to exist an abnormality and is marked, otherwise, it is not marked;

[0026] The reflection delay duration of the marked leaf is calculated by recording the difference between the time of receiving the reflected signal and the time of emitting the signal through the terahertz receiving probe.

[0027] In a preferred embodiment, in step S3, the penetration rate of the terahertz emitting device is obtained through the terahertz parameter database;

[0028] The penetration rate and the reflection delay duration are respectively standardized to obtain a penetration rate coefficient and a reflection delay duration coefficient;

[0029] The structure evaluation feature is calculated by integrating the penetration rate coefficient and the reflection delay duration coefficient;

[0030] If the structure evaluation feature is greater than a preset structure evaluation threshold, it is determined to divide the marked leaf into regions;

[0031] Otherwise, it is determined not to divide the marked leaf into regions.

[0032] In a preferred embodiment, in step S3, when the marked leaf is divided into regions, the marked leaf is divided into regions according to a preset size area;

[0033] The terahertz time domain signal of each divided region is obtained by emitting the terahertz emitting device along each divided region, and the frequency domain absorption spectrum curve is extracted by performing Fourier transform on the terahertz time domain signal;

[0034] In the frequency domain absorption spectrum curve, the morphological parameters of the absorption peak are identified by using a peak detection algorithm, and the morphological parameters include the absorption peak center frequency and the peak width value;

[0035] The difference between the absorption peak center frequency and a preset reference absorption peak frequency is taken as the absorption peak shift.

[0036] In a preferred embodiment, in step S4, the peak width values of the divided regions are arranged in turn according to the spatial order of the leaf scanning path, and the peak width change trend is obtained by subtracting the peak width values of adjacent divided regions.

[0037] After standardizing the peak width variation trend and absorption peak shift respectively, the peak width variation trend coefficient and absorption peak shift coefficient are obtained.

[0038] The dielectric perturbation coefficient is obtained by subtracting the peak width variation trend coefficient from the absorption peak offset coefficient.

[0039] In a preferred embodiment, in step S4, a preset dielectric disturbance threshold is compared with the dielectric disturbance coefficient to determine whether a fault exists in the divided area and an alarm signal is generated.

[0040] If the dielectric perturbation coefficient is greater than the preset dielectric perturbation threshold, a fault is determined and an alarm signal is generated.

[0041] Conversely, if the condition is not met, it is determined that there is no fault in the region division.

[0042] The technical effects and advantages of this invention are as follows:

[0043] This invention assesses the operating status of wind turbines by monitoring the blade rotation speed. When operating at low speed, it collects the reflected and incident signal intensities of each blade and calculates the terahertz emission ratio. Based on the terahertz reflection ratio, it determines whether there are any abnormalities in the blades. Blades with abnormalities are marked, and the reflection delay time and penetration rate of the terahertz emission device of the marked blades are detected to generate structural evaluation characteristics. The marked blades are divided into different regions, and the absorption peak shift and peak width values ​​of each region are detected. The peak width change trend is evaluated based on the peak width value, and the dielectric disturbance coefficient of the divided region is calculated in combination with the absorption peak shift to determine whether there is a fault in the divided region and generate an alarm signal. Utilizing the high penetrability of terahertz waves, high-precision detection of surface and internal defects of the blades is achieved, providing reliable data support for fault detection and maintenance of wind turbine blades and improving the power generation efficiency of wind turbine units. Attached Figure Description

[0044] Fig. 1 This is a flowchart illustrating the implementation of the wind turbine blade fault detection method based on terahertz analysis according to the present invention.

[0045] Fig. 2 This is a schematic diagram illustrating the steps of the wind turbine blade fault detection method based on terahertz analysis of the present invention. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0047] This invention assesses the operating status of wind turbines by monitoring the blade rotation speed. When operating at low speed, it collects the reflected and incident signal intensities of each blade and calculates the terahertz emission ratio. Based on the terahertz reflection ratio, it determines whether there are any abnormalities in the blades. Blades with abnormalities are marked, and the reflection delay time and penetration rate of the terahertz emission device of the marked blades are detected to generate structural evaluation characteristics of the marked blades. The marked blades are divided into different regions, and the absorption peak shift and peak width of each region are detected. The peak width change trend is evaluated based on the peak width value, and the dielectric disturbance coefficient of the divided region is calculated in combination with the absorption peak shift to determine whether there is a fault in the divided region and generate an alarm signal. By utilizing the high penetration characteristics of terahertz waves, high-precision detection of surface and internal defects of blades is achieved, providing reliable data support for fault detection and maintenance of wind turbine blades.

[0048] Example 1, such as Figs. 1-2 As shown, the wind turbine blade fault detection method based on terahertz analysis includes the following steps:

[0049] Step S1: Monitor the blade rotation speed of the wind turbine, assess the operating status of the wind turbine based on the blade rotation speed, and scan the blade surface through a terahertz transmitter when the operating status is low speed, and collect the reflected signal intensity and incident signal intensity of each blade.

[0050] Step S2: Calculate the terahertz reflectivity by combining the intensity of the reflected signal and the intensity of the incident signal. Determine whether there is any abnormality in the blade based on the terahertz reflectivity. Mark the blades with abnormalities and detect the reflection delay of the marked blades.

[0051] Step S3: Obtain the penetration rate of the terahertz transmitter, generate the structural evaluation features of the marked blade by combining the reflection delay time, divide the marked blade into different regions based on the structural evaluation features, and detect the absorption peak shift and peak width of each region.

[0052] Step S4: Evaluate the peak width change trend based on the peak width value, calculate the dielectric disturbance coefficient of the divided region based on the absorption peak offset, determine whether there is a fault in the divided region based on the dielectric disturbance coefficient, and generate an alarm signal.

[0053] The specific implementation is as follows:

[0054] In step S1, the blade speed of the wind turbine is monitored in real time using a blade speed monitoring device installed on the main shaft of the wind turbine.

[0055] The blade speed monitoring device is a Hall effect speed sensor, used to accurately measure and output data on the rotation speed of wind turbine blades. It is fixedly installed on the outside of the main shaft housing of the wind turbine, with its sensing surface maintaining a fixed distance from the end of the main shaft. Several magnetic trigger blocks are evenly installed on the end face of the main shaft. When the main shaft drives the blades to rotate, the magnetic trigger blocks periodically pass through the magnetic induction area of ​​the Hall effect speed sensor and output the corresponding pulse signal sequence.

[0056] The Hall effect speed sensor counts the total number of pulse signals within a fixed sampling period. Each time the main shaft rotates one revolution, a magnetic trigger block activates the Hall effect speed sensor to output a pulse signal frequency. By combining these parameters, the number of revolutions the blade makes per unit time is calculated, thus yielding the blade speed. The calculation formula is as follows:

[0057] ;

[0058] in, For blade rotation speed, The total number of pulse signals. This refers to the pulse signal frequency, which is the number of pulses generated per revolution of the spindle. 60 represents the duration of the sampling period, and 60 is a time unit conversion constant.

[0059] Determining the operating status of a wind turbine based on its blade rotation speed:

[0060] When the blade rotation speed is greater than the preset speed threshold, the wind turbine is determined to be in a high-speed operating state.

[0061] When the blade rotation speed is less than or equal to the preset rotation speed threshold, the wind turbine is determined to be in a low-speed operation state.

[0062] It should be noted that the speed threshold is a parameter used to distinguish the operating status of wind turbine generators. During the process from the start-up of the wind turbine generator set to normal operation, the blade speed change data is continuously collected. When the blade speed gradually increases, the average speed value of the blade in the early stage of operation is recorded and set as the speed threshold.

[0063] When the operating state is detected as low speed, the blade surface is scanned by a terahertz transmitter, and the reflected signal intensity and incident signal intensity of each scanning point on each blade are collected.

[0064] The terahertz transmitting device includes a terahertz wave emitting source and a terahertz receiving probe. When scanning the blade surface, the terahertz wave emitting source emits continuous terahertz electromagnetic waves, and the incident signal intensity is the signal intensity formed when the terahertz electromagnetic waves are perpendicularly irradiated onto the surface of the wind turbine blade.

[0065] When terahertz electromagnetic waves irradiate the blade surface, due to the physical properties of the blade material such as dielectric constant and surface roughness, the terahertz electromagnetic waves are reflected at the blade surface and internal interface and received by the terahertz receiving probe. The intensity of the reflected signal is the same as the intensity of the signal received by the terahertz receiving probe.

[0066] The terahertz transmitter performs a linear scan along the blade length with a fixed step distance, and simultaneously acquires the incident signal intensity and the reflected signal intensity at each scan point;

[0067] It should be noted that terahertz refers to electromagnetic waves with a frequency range between 0.1THz and 10THz, which are electromagnetic wavebands located between microwaves and infrared light. Terahertz waves combine the penetrating characteristics of microwaves with the high-resolution characteristics of light waves. They have physical characteristics such as non-ionization, low energy, and high sensitivity, and can penetrate a variety of non-metallic materials without damaging their internal structure.

[0068] In step S2, the incident signal intensity and reflected signal intensity at each scanning point are processed to calculate the terahertz reflectance at each scanning point. The calculation formula is as follows:

[0069] ;

[0070] in, Terahertz reflectivity The intensity of the reflected signal. The intensity of the incident signal.

[0071] Terahertz reflectance reflects the ability of blade material to reflect terahertz waves at the scanning point. The larger the value, the stronger the reflection of terahertz waves by the material or interface at that location, which may indicate structural anomalies such as delamination, cracks, or bubbles. The smaller the value, the stronger the absorption or scattering ability of the material at that location, which may indicate wetting, absorption anomalies, or surface damage.

[0072] The reflectance deviation index is obtained by subtracting the terahertz reflectance of each scanning point from the preset standard reflectance and taking the absolute value.

[0073] The reflectance deviation index reflects the degree to which the terahertz reflection characteristics of the scanning point deviate from the normal reference state. The larger the value, the more obvious the difference between the electromagnetic response of the scanning point and the normal blade, and the higher the degree of abnormality. The smaller the value, the closer the reflection characteristics of the scanning point are to the normal state, and the more complete the structure.

[0074] The reflectance deviation index of each scanning point is compared with the preset deviation threshold:

[0075] When the terahertz reflectance of each scanning point is greater than or equal to the preset deviation threshold, it is determined that there is an abnormal signal at that scanning point.

[0076] When the terahertz reflectance of each scanning point is less than the preset deviation threshold, it is determined that there is no abnormal signal at that scanning point.

[0077] It should be noted that the standard reflectance refers to the terahertz reflectance reference value corresponding to the normal state of the blade. It is used to reflect the electromagnetic reflection characteristics of the blade material under the conditions of no damage, uniformity and integrity. It is obtained by performing terahertz scanning measurements on multiple normal blades and calculating the average reflectance of each scanning point. The deviation threshold is the boundary value used to determine whether there is an abnormal signal at the scanning point after subtracting the terahertz reflectance of each scanning point from the standard reflectance and taking the absolute value. It is obtained by performing full-surface scanning on several blades in good condition, obtaining the standard reflectance of the blade, calculating the standard deviation of the standard reflectance of several blades, and using twice the standard deviation as the deviation threshold.

[0078] After determining the abnormal signals at all scan points on the blade, the number of scan points with abnormal signals on each blade is counted to obtain the number of abnormal scan points. The average number of abnormal scan points across all blades is then calculated, and this average is used as the threshold for determining abnormal blades. The calculation formula is as follows:

[0079] ;

[0080] in, The threshold for identifying abnormal leaves. Let be the number of abnormal signals on the j-th leaf. This represents the total number of blades in a wind turbine. This is the index value of the blade.

[0081] Leaves with an abnormal signal count exceeding the abnormal leaf determination threshold are identified as having an abnormality and are marked accordingly;

[0082] Conversely, no marking is performed.

[0083] Furthermore, the reflection delay time of the marked blade is obtained, and the reflection delay time is calculated by recording the difference between the time of receiving the reflected signal and the time of transmitting the signal using a terahertz receiving probe.

[0084] The reflection delay time reflects the time consumed by the terahertz signal to propagate and reflect inside the blade. The larger the value, the longer the propagation path at that location, which may indicate an increase in interlayer gaps, delamination, or changes in material thickness. The smaller the value, the shorter the propagation path and the better the integrity of the material structure.

[0085] In step S3, the penetration rate of the terahertz emitting device is obtained through the terahertz parameter database. The penetration rate refers to the propagation speed of the terahertz wave inside the material being tested.

[0086] The penetration rate and reflection delay time are standardized to obtain the penetration rate coefficient and the reflection delay time coefficient, respectively.

[0087] The structural evaluation features are generated by combining the penetration rate coefficient and the reflection delay time coefficient. ,in, and As a preset weighting factor, This is the penetration rate coefficient. This is the reflection delay time factor. For structural evaluation characteristics.

[0088] It should be explained that the preset weighting factor is used to determine the influence of the penetration rate coefficient and the reflection delay time coefficient on the structural evaluation characteristics. The value range is from 0 to 1 and can be set according to the sensitivity of the terahertz detection device, the range of the material dielectric constant, and the distribution pattern of historical defect samples.

[0089] A higher penetration rate indicates a weaker absorption or scattering effect of terahertz waves inside the marked blade material, a more uniform dielectric layer structure, and a smaller structural evaluation feature; a longer reflection delay indicates an increased terahertz wave propagation path or the presence of multi-layer interface reflections, and a larger structural evaluation feature.

[0090] A preset structural evaluation threshold is compared with structural evaluation features to determine whether to divide the marked blades into different regions:

[0091] If the structural evaluation characteristics are greater than the preset structural evaluation threshold, then the marked blades are divided into regions.

[0092] Conversely, it is determined that the marked blades are not divided into regions.

[0093] When dividing the marking blade into regions, the marking blade is divided into regions according to the preset size area.

[0094] It should be noted that the terahertz parameter database is used to store the calibration parameters of the terahertz transmitter. In this embodiment, it is used to obtain the penetration rate of the terahertz transmitter. The standardization processing method includes, but is not limited to, standard linear transformation based on interval scaling, Z-Score standardization method based on statistics, or normalization method based on nonlinear mapping function. The application method of standardization processing will not be elaborated here. The preset structure evaluation threshold is a critical value used to determine whether there are abnormalities in the internal structure of the marked blade. It can be set according to the blade material characteristics and the statistical results of historical test samples. For example, the median of the distribution of structural evaluation characteristics of historical normal blade samples can be used as the preset structure evaluation threshold. The preset size area is used to limit the detection coverage of each divided region and is set according to the geometric characteristics of the blade.

[0095] The terahertz transmitter transmits along each divided region to acquire the terahertz time-domain signal of the divided region, and performs Fourier transform on the terahertz time-domain signal to extract the frequency domain absorption spectrum curve.

[0096] In the frequency domain absorption spectrum curve, the peak detection algorithm is used to identify the morphological parameters of the absorption peak, including the center frequency and peak width of the absorption peak.

[0097] The absorption peak center frequency refers to the frequency point at which the absorption intensity of the absorption peak reaches its maximum value in the frequency domain absorption spectrum curve.

[0098] Absorption peak offset refers to the degree of deviation of the center frequency of the absorption peak in the divided region relative to the preset reference absorption peak frequency. The difference between the center frequency of the absorption peak and the preset reference absorption peak frequency is used as the absorption peak offset.

[0099] When the absorption peak shift is negative, it indicates that the absorption peak has shifted towards lower frequencies, which means that the propagation speed of the terahertz wave in the divided region has slowed down and the internal structure of the material in the divided region is abnormal. When the absorption peak shift is positive, it means that the absorption peak has shifted towards higher frequencies, which means that the internal structure of the material in the divided region is normal.

[0100] Peak width refers to the frequency width of the absorption peak at half peak height, which reflects the concentration of energy distribution in the absorption spectrum.

[0101] It should be noted that the peak detection algorithm refers to a signal analysis algorithm used to identify the center position and morphological parameters of absorption peaks in the frequency domain absorption spectrum curve; the preset reference absorption peak frequency is used to evaluate whether the absorption peaks in the divided region have undergone frequency shift, and is obtained by statistically analyzing the average absorption peak positions of normal leaf samples detected in history.

[0102] In step S4, the peak width values ​​of the divided regions are arranged sequentially according to the spatial order of the blade scanning path, and the peak width values ​​of adjacent divided regions are subtracted to obtain the peak width variation trend.

[0103] If the peak width change trend is greater than 0, it indicates that the peak width value of the later segmented region is greater than that of the previous segmented region, the absorption peak shows a broadening trend, the non-uniformity of the dielectric environment inside the blade material is enhanced, and there may be abnormal blade structure; if the peak width change trend is less than 0, it indicates that the peak width value of the later segmented region is less than that of the previous segmented region, and the structure of the segmented region is normal.

[0104] It should be explained that the spatial order of the blade scanning path refers to the trajectory order of the terahertz detection device when scanning the marked blade's divided area;

[0105] After standardizing the peak width variation trend and absorption peak offset respectively, the peak width variation trend coefficient and absorption peak offset coefficient are obtained. The dielectric perturbation coefficient is obtained by subtracting the peak width variation trend coefficient and absorption peak offset coefficient.

[0106] The dielectric perturbation coefficient is used to reflect whether the energy absorption and propagation characteristics of the divided region are stable under the action of terahertz waves;

[0107] The preset dielectric disturbance threshold is compared with the dielectric disturbance coefficient to determine whether a fault exists in the divided area and generate an alarm signal.

[0108] If the dielectric perturbation coefficient is greater than the preset dielectric perturbation threshold, a fault is determined and an alarm signal is generated.

[0109] Conversely, if the condition is not met, it is determined that there is no fault in the region division.

[0110] It should be noted that the preset dielectric disturbance threshold is a critical value used to determine whether there is a fault in the divided area. It can be set based on historical test samples or equipment calibration results. For example, the average value can be taken as the preset dielectric disturbance threshold based on the statistical distribution of dielectric disturbance coefficients of multiple normal divided areas.

[0111] Finally, it should be noted that in this paper, relational terms such as first and second are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations.

[0112] Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0113] In this document, the singular forms “a,” “an,” and “the” may also include the plural forms unless the context clearly indicates otherwise. It should also be understood that terms such as “comprising / including” or “having” specify the presence of the stated features, integrals, steps, operations, components, parts, or combinations thereof, but do not preclude the possibility of the presence or addition of one or more other features, integrals, steps, operations, components, parts, or combinations thereof. Meanwhile, the term “and / or” as used in this specification includes any and all combinations of the associated listed items.

[0114] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0115] The above description of the disclosed embodiments will enable those skilled in the art to make or use various modifications to these embodiments. It will be readily apparent to those skilled in the art that the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for detecting wind turbine blade faults based on terahertz analysis, characterized in that: Includes the following steps: Step S1: Monitor the blade rotation speed of the wind turbine generator set, assess the operating status of the wind turbine generator set based on the blade rotation speed, and scan the blade surface through a terahertz transmitter when the operating status is low speed, and collect the reflected signal intensity and incident signal intensity of each blade. Step S2: Calculate the terahertz reflectivity by combining the intensity of the reflected signal and the intensity of the incident signal. Determine whether there is any abnormality in the blade based on the terahertz reflectivity. Mark the blades with abnormalities and detect the reflection delay of the marked blades. In step S2, the reflection delay of the marked blade is calculated by recording the difference between the time of receiving the reflected signal and the time of transmitting the signal using a terahertz receiving probe. Step S3: Obtain the penetration rate of the terahertz transmitter, generate the structural evaluation features of the marked blade by combining the reflection delay time, divide the marked blade into different regions based on the structural evaluation features, and detect the absorption peak shift and peak width of each region. In step S3, the penetration rate of the terahertz emitting device is obtained through the terahertz parameter database. The penetration rate refers to the propagation speed of the terahertz wave inside the material being tested. The penetration rate and reflection delay time are standardized to obtain the penetration rate coefficient and the reflection delay time coefficient, respectively. The structural evaluation features are generated by combining the penetration rate coefficient and the reflection delay time coefficient. ,in, and As a preset weighting factor, This is the penetration rate coefficient. This is the reflection delay time factor. For structural evaluation characteristics; Step S4: Evaluate the peak width change trend based on the peak width value, calculate the dielectric disturbance coefficient of the divided region in combination with the absorption peak offset, determine whether there is a fault in the divided region based on the dielectric disturbance coefficient, and generate an alarm signal. In step S4, the peak width values ​​of the divided regions are arranged sequentially according to the spatial order of the blade scanning path, and the peak width values ​​of adjacent divided regions are subtracted to obtain the peak width variation trend. After standardizing the peak width variation trend and absorption peak shift respectively, the peak width variation trend coefficient and absorption peak shift coefficient are obtained. The dielectric perturbation coefficient is obtained by subtracting the peak width variation trend coefficient from the absorption peak offset coefficient.

2. The method for wind turbine blade fault detection based on terahertz analysis according to claim 1, characterized in that: In step S1, the blade speed of the wind turbine is monitored in real time using a blade speed monitoring device installed on the main shaft of the wind turbine. Determining the operating status of a wind turbine based on its blade rotation speed: When the blade rotation speed is greater than the preset speed threshold, the wind turbine is determined to be in a high-speed operating state. When the blade rotation speed is less than or equal to the preset rotation speed threshold, the wind turbine is determined to be in a low-speed operation state.

3. The method for wind turbine blade fault detection based on terahertz analysis according to claim 1, characterized in that: In step S1, when the operating state is detected to be low speed, the blade surface is scanned by the terahertz transmitter to collect the reflected signal intensity and incident signal intensity at each scanning point on the blade. The incident signal strength is the signal strength formed when the terahertz electromagnetic waves emitted by the terahertz transmitter are perpendicularly irradiated onto the surface of the wind turbine blade. The reflected signal intensity is the signal intensity received by the terahertz receiving probe after the terahertz electromagnetic wave is reflected at the blade surface and internal interface.

4. The wind turbine blade fault detection method based on terahertz analysis according to claim 3, characterized in that: In step S2, the ratio of the incident signal intensity to the reflected signal intensity at each scanning point is calculated to obtain the terahertz reflection ratio at each scanning point; The reflectance deviation index is obtained by subtracting the terahertz reflectance of each scanning point from the preset standard reflectance and taking the absolute value. When the terahertz reflectance of each scanning point is greater than or equal to the preset deviation threshold, it is determined that there is an abnormal signal at that scanning point. When the terahertz reflectance of each scanning point is less than the preset deviation threshold, it is determined that there is no abnormal signal at that scanning point.

5. The wind turbine blade fault detection method based on terahertz analysis according to claim 4, characterized in that: In step S2, the number of abnormal scan points is obtained by counting the number of scan points with abnormal signals on each blade; The average number of abnormal scan points for all leaves is calculated as the threshold for determining abnormal leaves. Leaves with more abnormal signals than the abnormal leaf determination threshold are identified as having abnormalities and are marked; otherwise, they are not marked.

6. The method for wind turbine blade fault detection based on terahertz analysis according to claim 1, characterized in that: In step S3, the penetration rate of the terahertz transmitting device is obtained through the terahertz parameter database; The penetration rate and reflection delay time are standardized to obtain the penetration rate coefficient and the reflection delay time coefficient, respectively. The structural evaluation characteristics are obtained by combining the penetration rate coefficient and the reflection delay time coefficient. If the structural evaluation characteristics are greater than the preset structural evaluation threshold, then the marked blades are divided into regions. Conversely, it is determined that the marked blades are not divided into regions.

7. The method for wind turbine blade fault detection based on terahertz analysis according to claim 6, characterized in that: In step S3, when dividing the marking blade into regions, the marking blade is divided into regions according to a preset size area; The terahertz transmitter transmits along each divided region to acquire the terahertz time-domain signal of the divided region, and performs Fourier transform on the terahertz time-domain signal to extract the frequency domain absorption spectrum curve. In the frequency domain absorption spectrum curve, the peak detection algorithm is used to identify the morphological parameters of the absorption peak, including the center frequency and peak width of the absorption peak. The difference between the center frequency of the absorption peak and the preset reference absorption peak frequency is used as the absorption peak offset.

8. The method for wind turbine blade fault detection based on terahertz analysis according to claim 1, characterized in that: In step S4, a preset dielectric disturbance threshold is compared with the dielectric disturbance coefficient to determine whether a fault exists in the divided area and an alarm signal is generated. If the dielectric perturbation coefficient is greater than the preset dielectric perturbation threshold, a fault is determined and an alarm signal is generated. Conversely, if the condition is not met, it is determined that there is no fault in the region division.

Images