Bolt fracture detection method and device for wind generating set

Abstract

The invention relates to the technical field of bolt fracture detection, and provides a wind generating set bolt fracture detection method and device, and the method comprises the steps: obtaining a surrounding magnetic field signal of a to-be-detected bolt; correcting the detected magnetic field signal based on the geomagnetic field to obtain a corrected magnetic field value; on the basis of the current corrected magnetic field value, the magnetic dipole intensity change gradient of the current moment relative to the previous moment is calculated, and when the magnetic dipole intensity change gradient exceeds a set value, it is judged that an abnormal defect occurs; and when the abnormal defect exists, for the current corrected magnetic field value, based on the obtained frequency spectrum energy ratio, calculating fracture feature extraction and diagnosis, and obtaining an abnormal expansion stage of the current defect. By collecting and correcting a magnetic field signal, calculating a magnetic dipole intensity change gradient to identify abnormity, and then combining a frequency spectrum energy ratio to diagnose a fracture stage, early online monitoring of bolt fracture is realized. The method has high sensitivity and strong anti-interference capability, and is suitable for intelligent monitoring in a wind power environment.

Description

Technical Field

[0001] This invention relates to the technical field of bolt fracture detection, specifically to a method and apparatus for detecting bolt fracture in wind turbine generator sets. Background Technology

[0002] The statements in this section are merely background information related to the present invention and do not necessarily constitute prior art.

[0003] In wind power equipment, a large number of high-preload bolts undertake critical structural connection tasks, and their health status directly affects the safety and stability of equipment operation. Because wind power equipment operates under high wind loads, strong vibrations, and complex electromagnetic environments for extended periods, bolts are prone to latent damage such as fatigue cracks and fractures. Especially in the early stages, these damages often lack obvious characteristics, making timely and effective monitoring and diagnosis difficult using conventional methods. Therefore, developing a highly sensitive, interference-resistant, and intelligent detection technology capable of online identification of early bolt fractures has significant engineering application value and is an urgent technical need.

[0004] Currently, commonly used bolt inspection methods mainly include visual inspection, ultrasonic testing, and magnetic particle testing. However, these methods have significant limitations: visual inspection cannot identify microcracks (the detection rate is less than 15% when the crack depth is less than 1 mm); ultrasonic testing relies on coupling agents and is extremely sensitive to bolt surface roughness (the detection effect decreases significantly when the surface roughness Ra is greater than 6.3 μm); magnetic particle testing is only suitable for offline inspection after equipment shutdown and cannot meet the needs of online real-time monitoring. In addition, there is significant magnetic field interference in the wind power equipment operating environment, such as geomagnetic fluctuations (±50 μT) and electromagnetic interference from electrical equipment (exceeding 200 μT), which further limits the accuracy and stability of traditional magnetic testing methods. Summary of the Invention

[0005] To address the aforementioned problems, this invention proposes a method and device for detecting bolt fracture in wind turbine generators. By acquiring and correcting magnetic field signals, calculating the gradient of magnetic dipole intensity changes to identify anomalies, and then combining this with spectral energy ratio analysis to diagnose the fracture stage, early online monitoring of bolt fracture is achieved. This method possesses high sensitivity and strong anti-interference capabilities, making it suitable for intelligent monitoring in wind power environments.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: One or more embodiments provide a method for detecting bolt fracture in a wind turbine generator set, comprising the following steps: According to the set sampling frequency, the magnetic field signal around the bolt to be tested is acquired; The detected magnetic field signal is corrected based on the Earth's magnetic field to obtain the corrected magnetic field value; Based on the current corrected magnetic field value Calculate the gradient of magnetic dipole intensity change at the current moment relative to the previous moment. If the gradient of magnetic dipole intensity change exceeds the set value, it is determined that an abnormal defect has occurred. When an abnormal defect exists, based on the obtained spectral energy ratio, fracture feature extraction and diagnosis are performed for the current corrected magnetic field value to obtain the abnormal expansion stage of the current defect.

[0007] One or more embodiments provide a wind turbine generator bolt fracture detection device, comprising: The magnetic sensing module is used to acquire the three-dimensional magnetic field signal around the bolt in real time. A wireless monitoring network is used to achieve stable communication between each detection node and between the detection node and the host computer; The data processing unit is configured to execute the steps of the above-described method for detecting bolt fracture in a wind turbine generator set, thereby enabling the detection of bolt crack defects. One or more embodiments provide a wind turbine generator bolt fracture detection device, comprising: The acquisition module is configured to acquire the surrounding magnetic field signal of the bolt under test according to a set sampling frequency; The correction module is configured to correct the detected magnetic field signal based on the geomagnetic field to obtain the corrected magnetic field value; The anomaly detection module is configured to base its detection on the current corrected magnetic field value. Calculate the gradient of magnetic dipole intensity change at the current moment relative to the previous moment. If the gradient of magnetic dipole intensity change exceeds the set value, it is determined that an abnormal defect has occurred. The stage identification module is configured to, when an abnormal defect exists, calculate fracture feature extraction and diagnosis based on the acquired spectral energy ratio for the current corrected magnetic field value, and obtain the abnormal expansion stage of the current defect.

[0008] A computer-readable storage medium having a computer program / instructions stored thereon, which, when executed by a processor, implements the steps of the above-described method for detecting bolt fracture in a wind turbine generator set.

[0009] A computer program product includes a computer program / instructions that, when executed by a processor, implement the steps of the aforementioned method for detecting bolt fracture in a wind turbine generator set.

[0010] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention acquires real-time magnetic field signals around bolts during wind turbine operation, ensuring temporal continuity through a set sampling frequency. Due to geomagnetic fluctuations and electromagnetic interference from electrical equipment at wind power sites, a geomagnetic field calibration method is used to correct the acquired signals, thereby obtaining true magnetic field characteristic values ​​reflecting the bolt's condition. By calculating the gradient of magnetic dipole intensity changes over continuous time intervals, minute changes in the bolt's internal structure can be identified. Once this rate of change exceeds an empirically set threshold, it is judged as a potential anomaly. This detection method possesses high sensitivity, strong anti-interference capabilities, and intelligent judgment capabilities, enabling online monitoring of early-stage hidden damage to bolts, making it particularly suitable for the long-term operational needs of wind power equipment in complex electromagnetic environments. Compared to traditional methods, it does not rely on contact coupling or shutdown status, is unaffected by significant surface roughness, and overcomes the limitations of visual, ultrasonic, and magnetic particle inspection technologies in wind power applications. Furthermore, the joint judgment mechanism based on spectral energy ratio and magnetic dipole gradient enhances the accuracy of early defect signal identification and the ability to judge development trends, effectively improving the operational safety and maintenance efficiency of wind turbines.

[0011] The advantages of the present invention, as well as its additional advantages, will be described in detail in the following specific embodiments. Attached Figure Description

[0012] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute a limitation thereof.

[0013] Figure 1 This is a flowchart of the detection method of Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the process of correcting the detected magnetic field signal based on the geomagnetic field in Embodiment 1 of the present invention; Detailed Implementation The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0014] It should be noted that the following detailed descriptions are exemplary and intended to provide further illustration of the invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0015] It should be noted that the terminology used herein is for describing particular embodiments only and is not intended to limit the exemplary embodiments of the present invention. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof. It should be noted that, without conflict, the various embodiments and features within those embodiments can be combined with each other. The embodiments will now be described in detail with reference to the accompanying drawings.

[0016] Example 1 In one or more of the technical solutions disclosed in the implementation methods, such as Figures 1 to 2 As shown, a method for detecting bolt fracture in a wind turbine generator set includes the following steps: Step 1: Obtain the surrounding magnetic field signal of the bolt to be tested according to the set sampling frequency; Step 2: Correct the detected magnetic field signal based on the Earth's magnetic field to obtain the corrected magnetic field value; Step 3: Based on the current corrected magnetic field value Calculate the gradient of magnetic dipole intensity change at the current moment relative to the previous moment. If the gradient of magnetic dipole intensity change exceeds the set value, it is determined that an abnormal defect has occurred. Step 4: When an abnormal defect exists, for the current corrected magnetic field value, calculate the fracture feature extraction and diagnosis based on the obtained spectral energy ratio to obtain the abnormal expansion stage of the current defect.

[0017] This method combines magnetic detection principles with signal processing algorithms. It collects magnetic field signals around bolts in real time during wind turbine operation, ensuring temporal continuity using a set sampling frequency. Due to geomagnetic fluctuations and electromagnetic interference from electrical equipment at wind power sites, a geomagnetic field calibration method is used to correct the collected signals, thereby obtaining the true magnetic field characteristics reflecting the bolt's condition. By calculating the gradient of magnetic dipole intensity changes over continuous time intervals, minute changes in the bolt's internal structure can be identified. Once this rate of change exceeds an empirically set threshold, it is judged as a potential anomaly. Furthermore, fracture feature information is extracted using spectral energy ratio, and pattern recognition methods are combined to analyze the evolution stages of the abnormal signal, thus achieving a phased diagnosis of bolt fracture development.

[0018] This embodiment of the detection method possesses high sensitivity, strong anti-interference capabilities, and intelligent judgment capabilities, enabling online monitoring of early-stage hidden damage from bolt fracture. It is particularly suitable for the long-term operational needs of wind power equipment in complex electromagnetic environments. Compared to traditional methods, it does not rely on contact coupling or shutdown conditions and is unaffected by significant surface roughness, overcoming the limitations of visual, ultrasonic, and magnetic particle inspection technologies in wind power applications. Furthermore, the joint judgment mechanism based on spectral energy ratio and magnetic dipole gradient enhances the accuracy of early defect signal identification and the ability to judge development trends, effectively improving the operational safety and maintenance efficiency of wind turbine units.

[0019] In step 1, magnetic signal acquisition and preprocessing are performed. Specifically, a triaxial magnetoresistive sensor (HMC5883L) is used to acquire the magnetic field around the bolt. Optionally, by setting a fixed sampling frequency, such as adjustable from 10Hz to 100Hz, a high-sensitivity triaxial magnetoresistive sensor is placed around the bolt to be tested to collect its three-dimensional magnetic field signal in real time, including the X-axis, Y-axis, and Z-axis, with the unit being μT; Optionally, the triaxial magnetoresistive sensor is installed at a set angle to the axis of the bolt to be measured in order to optimize the response to changes in the magnetic field; Optionally, the triaxial magnetoresistive sensor is installed at a set angle of 90° or 45° to the axis of the bolt to be measured.

[0020] In magnetic field detection, the magnetic field data output by sensors (especially triaxial magnetoresistive sensors) is often affected by both their own hardware errors (such as proportional coefficient differences, zero offset, and axial non-orthogonality) and external magnetic interference, resulting in variations in the original magnetic field value. Deviation from the actual geomagnetic field This affects the accuracy of the judgment.

[0021] To address this, this embodiment proposes a geomagnetic compensation model based on a three-dimensional ellipsoid fitting algorithm, which automatically calibrates the nonlinear error of the sensor and dynamically eliminates interfering magnetic fields, thereby achieving accurate extraction of magnetic signals.

[0022] In step 2, the detected magnetic field signal is ellipsoidally fitted, and a geomagnetic field compensation model is constructed. The detected magnetic field signal is then corrected based on the geomagnetic field, including the following steps: Step 21: Construct the ellipsoid fitting equation and correct the detected magnetic field signal to obtain the first corrected magnetic field signal; Step 211: Obtain the original magnetic field data, and use the least squares method to fit the data to form an ellipsoid model, i.e., the ellipsoid fitting equation; Place the magnetic sensor in a defect-free device; the sensor model can be HMC5883L. Collect a sufficient number of three-dimensional magnetic field data points by slowly rotating the magnetic sensor in various directions. The number of data sets collected should be no less than 300; these data constitute a point cloud in space, which should theoretically be distributed on a spherical surface. Due to sensor errors, the actual collected data will be distributed on an eccentric, deformed ellipsoid. Specifically, the least squares method is used to fit these magnetic data, and a general three-dimensional ellipsoid fitting equation model is established: ; Solve for parameters A~J to obtain the shape, orientation, and center position of the ellipsoid.

[0023] Step 212: Extract the offset vector and scale matrix from the fitted ellipsoid equation as key calibration parameters. The offset vector Offset represents the offset of the ellipsoid center relative to the origin, corresponding to the zero offset error of the sensor in three directions; The scale matrix represents the radius length (a, b, c) of the ellipsoid along the three principal axes, and is used to correct for inconsistencies in the sensor's sensitivity across the axes. Step 213: Based on the obtained key calibration parameters, correct the detected magnetic field signal to obtain the first corrected magnetic field signal; Zero bias correction: ; Proportion Correction: ; in, This is the final calibrated magnetic field value, i.e., the first corrected magnetic field signal; In this embodiment, ellipsoid fitting effectively eliminates distortion caused by sensor hardware errors, reduces magnetic field measurement error to within ±3μT, and provides a stable and reliable magnetic signal input for subsequent crack identification.

[0024] Step 22: Subtract the geomagnetic reference signal from the obtained first corrected magnetic field signal. and environmental interference signals The corrected magnetic signal is obtained by the following formula: ; in, It represents the geomagnetic reference signal, which can be extracted by averaging under normal or static conditions, and a baseline update strategy can be set to remeasure according to a set time period or every operating cycle. ; To represent environmental interference signals, noise signals obtained by processing the detected magnetic signals can be obtained using methods such as bandpass filtering, wavelet packet decomposition, and outlier removal. Step 3: Based on the current corrected magnetic field value Calculate the gradient of magnetic dipole intensity change at the current moment relative to the previous moment. If the gradient of magnetic dipole intensity change exceeds the set value, it is determined that an abnormal defect has occurred. Among them, the calculation of magnetic dipole strength: Step 31: Set the magnetic field value Decomposed into triaxial magnetic field values Calculate the strength of the magnetic dipole The calculation formula is: ; Step 32: Based on the calculated rate of change of the magnetic field gradient relative to the previous time, compare the rate of change of the gradient with the set value to determine whether there is any abnormal defect. ; This embodiment analyzes the gradient change rate of magnetic field strength to determine in real time whether there are magnetic anomalies such as cracks in the bolt structure, thereby achieving rapid identification of early damage. This method has high detection sensitivity, accurately identifying early defects with crack depths of not less than 0.5 mm, and a response time of less than 30 seconds, meeting the rapid early warning requirements of wind power equipment. Because this method is based on non-contact detection of magnetic field changes, it does not require direct contact with the tested structure, making it particularly suitable for complex operating environments such as offshore wind power, characterized by high humidity, high corrosion, and high electromagnetic interference.

[0025] Step 4: When an abnormal defect exists, based on the obtained spectral energy ratio and the current corrected magnetic field value, fracture feature extraction and diagnosis are calculated to obtain the abnormal expansion stage of the current defect. The method for determining the abnormal expansion stage of the current defect includes the following steps: Step 41: For the current corrected magnetic field value, use the sym8 wavelet function to perform five-layer wavelet threshold denoising to eliminate background noise interference and enhance the feature signal; Specifically, the noise reduction expression is: ; Step 42: Perform frequency domain analysis on the denoised signal. Analyze the spectrum of the denoised signal using Fast Fourier Transform (FFT) to extract high-frequency energy. With low-frequency energy And calculate the spectral energy ratio; Among them, high-frequency energy It can reflect the intensity of crack excitation signals and the energy in the low-frequency range. It can reflect changes in the background magnetic field; The Band Energy Ratio (BER) of the two is calculated as follows: ; in, To prevent extremely small positive numbers from causing numerical anomalies when the denominator is zero, such as ; Step 43: Based on the degree of increase in the spectral energy ratio compared to the normal state, determine that the current magnetic anomaly is in the crack defect propagation stage of the bolt crack; Step 431: Construct a reference model for the spectrum energy ratio under normal conditions and determine the statistical mean of the spectrum energy ratio under normal conditions. This is taken as a normal reference value; Under crack-free conditions, magnetic field data of multiple bolt operating cycles are collected for a set duration. Wavelet denoising and spectral energy ratio analysis were performed on the collected data to obtain the statistical mean of the spectral energy ratio under normal conditions. This is taken as a normal reference value; Step 432: Compare the BER value at the current detection time with the normal reference value. Compare the results and calculate the growth rate; ; Step 423: Determine the crack propagation stage based on the set crack stage division based on the growth ratio; A specific method for dividing crack stages, specifically: When the growth factor is less than 1.5, the crack stage is within the normal fluctuation range and is classified as a low-risk level. When the growth factor is between 1.5 and 3, the crack stage is the early stage of crack initiation and is classified as a medium-risk level. When the growth factor is greater than 3, the crack stage is the crack propagation stage, which is a high-risk stage, and it is recommended to shut down the machine. When a high-risk phase is detected, the system triggers an alarm signal and records the abnormal event, uploading it to the structural health monitoring platform for maintenance personnel to refer to.

[0026] Furthermore, by combining the SN fatigue curves of the bolt material, it is possible to further estimate the remaining life and classify the risk level.

[0027] This scheme, by introducing statistical modeling and a multiple-based discrimination mechanism, achieves refined identification of the bolt fracture evolution process, enhancing the monitoring system's sensitivity to damage development and its quantitative diagnostic capabilities. Compared to traditional methods that only determine whether something is abnormal, this method can identify the specific stage of the crack, facilitating maintenance personnel to implement tiered response measures. This optimizes maintenance resource allocation, reduces the risk of sudden failures, and improves the safety and economy of system operation. Furthermore, the energy ratio modeling approach exhibits good repeatability and scalability, making it applicable to various bolt types and complex working environments.

[0028] Example 2 Based on Embodiment 1, this embodiment provides a bolt fracture detection device for wind turbine generator sets, suitable for long-term condition monitoring and early defect warning of structural connection parts of offshore or onshore wind turbine generator sets. The device includes: a magnetic sensing module, a wireless monitoring network, a data processing unit, and an energy management module. The magnetic sensing module is used to acquire the three-dimensional magnetic field signal around the bolt in real time. A wireless monitoring network is used to achieve stable communication between each detection node and between the detection node and the host computer; The data processing unit is configured to execute the steps of the bolt fracture detection method for wind turbine generator sets described in Embodiment 1, realizing bolt crack defect detection. It can employ edge MCU platforms such as STM32 and ESP32, possessing floating-point arithmetic and DSP instruction sets to meet signal filtering and matrix transformation requirements. The embedded platform has a built-in fault identification model, enabling anomaly judgment and early warning decisions to be completed at the node end, reducing communication load.

[0029] Optionally, for the probe structure of the magnetic sensing module, it is recommended to adopt a multi-layer shielding structure, with an outer layer of permalloy shell and an inner layer of μ-metal (high magnetic permeability material) shielding layer, to effectively shield surrounding electromagnetic interference and improve the purity of magnetic signal acquisition.

[0030] A further technical solution includes a pneumatic push rod structure mounted on the probe mounting base. This structure automatically adjusts the detection distance based on minute changes in the bolt surface, with an adjustment accuracy of ±0.05mm. This ensures that the sensor always maintains the optimal detection position, improving signal stability and repeatability.

[0031] The energy management module is used to provide a stable and sustainable energy supply for the sensor nodes; To verify the practical application effect of this device, it was deployed in the following wind power scenario: The test unit was a GW155-4.5MW wind turbine generator; the monitoring object was the tower flange connection bolts (M36 specification); the deployment scheme was as follows: 8 sensor nodes were evenly arranged on each flange connection surface, the sampling frequency was set to 1kHz, the triaxial magnetic field signal was recorded in real time and processed locally; abnormal conditions were uploaded to the central diagnostic platform via LoRa. Through detection, the system can reliably identify early defects with crack depth ≥0.5mm, meeting the requirements for identifying latent fractures in offshore wind turbine bolts. Under operating conditions with wind speeds reaching 12m / s, the system's signal-to-noise ratio remains at ≥35dB, and the sensing signal is not significantly affected by distortion. Through modular design and batch deployment, this system reduces the cost of single bolt monitoring by approximately 60% compared to traditional sensing and manual inspection solutions, improving economic efficiency and engineering promotion value.

[0032] This embodiment verifies the feasibility and practicality of using the device to perform non-contact, anti-interference, and low-power monitoring of high-preload bolts in wind power equipment. It is applicable to key structural areas such as towers, impeller connections, and main shaft flanges, providing a cost-effective and highly sensitive technical solution for wind power structural health monitoring.

[0033] Example 3 Based on Embodiment 1, this embodiment provides a wind turbine generator bolt fracture detection device, comprising: The acquisition module is configured to acquire the surrounding magnetic field signal of the bolt under test according to a set sampling frequency; The correction module is configured to correct the detected magnetic field signal based on the geomagnetic field to obtain the corrected magnetic field value; The anomaly detection module is configured to base its detection on the current corrected magnetic field value. Calculate the gradient of magnetic dipole intensity change at the current moment relative to the previous moment. If the gradient of magnetic dipole intensity change exceeds the set value, it is determined that an abnormal defect has occurred. The stage identification module is configured to, when an abnormal defect exists, calculate fracture feature extraction and diagnosis based on the acquired spectral energy ratio for the current corrected magnetic field value, and obtain the abnormal expansion stage of the current defect.

[0034] It should be noted that each module in this embodiment corresponds one-to-one with each step in embodiment 1, and their specific implementation process is the same, so it will not be repeated here.

[0035] Example 4 This embodiment provides an electronic device, including a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When the computer instructions are executed by the processor, they complete the steps in the wind turbine generator bolt fracture detection method of Embodiment 1.

[0036] Example 5 This embodiment provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by a processor, they complete the steps in the wind turbine generator bolt fracture detection method of Embodiment 1.

[0037] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

[0038] While the specific embodiments of the present invention have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of the present invention are still within the scope of protection of the present invention.

Claims

1. A wind turbine bolt breakage detection method, characterized by, The method comprises the following steps: acquiring the magnetic field signal around the bolt to be detected according to a set sampling frequency; correcting the detected magnetic field signal based on the geomagnetic field to obtain a corrected magnetic field value; based on the current corrected magnetic field value , a magnetic dipole strength change gradient relative to the previous time is calculated, and when the magnetic dipole strength change gradient exceeds a set value, it is determined that an abnormal defect has occurred; when there is an abnormal defect, calculating the fracture feature extraction and diagnosis based on the obtained spectrum energy ratio for the current corrected magnetic field value to obtain the abnormal expansion stage of the current defect.

2. The bolt fracture detection method of a wind turbine generator set according to claim 1, characterized in that: correcting the detected magnetic field signal based on the geomagnetic field comprises the following steps: constructing an ellipsoid fitting equation to correct the detected magnetic field signal to obtain a first corrected magnetic field signal; subtracting the geomagnetic reference signal and the environmental interference signal from the obtained first corrected magnetic field signal to obtain a corrected magnetic signal.

3. The bolt fracture detection method of a wind turbine generator set according to claim 2, characterized in that: constructing an ellipsoid fitting equation to correct the detected magnetic field signal to obtain a first corrected magnetic field signal comprises the following steps: acquiring the original magnetic field data, fitting the data by using the least square method to obtain an ellipsoid fitting equation; extracting the offset vector and the scale matrix from the obtained ellipsoid fitting equation as key calibration parameters; correcting the detected magnetic field signal based on the obtained key calibration parameters to obtain a first corrected magnetic field signal.

4. The bolt fracture detection method of a wind turbine generator set according to claim 1, characterized in that: the abnormal defect judgment method comprises the following steps: decomposing the magnetic field value into three axial magnetic field values decomposing the magnetic field value into three axial magnetic field values calculating the magnetic dipole strength ; comparing the gradient change rate with a set value to determine whether there is an abnormal defect based on the calculation of the magnetic field gradient change rate at the current time relative to the last time.

5. The bolt fracture detection method of a wind turbine generator set according to claim 1, characterized in that: the determination method of the abnormal expansion stage of the current defect comprises the following steps: using a sym8 wavelet function to perform five-layer wavelet threshold denoising on the current corrected magnetic field value to eliminate background noise interference and enhance the feature signal; The signal after noise reduction processing is subjected to frequency domain analysis, the de-noised signal is subjected to frequency spectrum analysis through fast Fourier transform, high frequency band energy is extracted respectively and low frequency band energy , and the spectrum energy ratio is calculated; judging the current magnetic anomaly to be the crack defect expansion stage of the bolt crack according to the growth degree of the spectrum energy ratio value compared with the normal state.

6. The bolt fracture detection method of a wind turbine generator set according to claim 1, characterized in that: constructing a spectrum energy ratio reference model of a normal state to determine a statistical mean value of the spectrum energy ratio in the normal state as a normal reference value , as a normal reference value; The BER value at the current detection time is compared with the normal reference value The growth multiple is calculated; the expansion stage of the crack is determined according to the set crack stage division based on the growth multiple.

7. A wind turbine generator unit bolt breakage detection device characterized by, It comprises: a magnetic sensing module for real-time acquisition of three-dimensional magnetic field signals around the bolt; a wireless monitoring network for stable communication between detection nodes and between detection nodes and the upper computer; a data processing unit configured to perform the steps of the bolt fracture detection method of a wind turbine generator set according to any one of claims 1-6 to realize crack defect detection of the bolt.

8. A wind turbine generator unit bolt breakage detection device characterized by, It comprises: an acquisition module configured to acquire the magnetic field signal around the bolt to be detected according to a set sampling frequency; a correction module configured to correct the detected magnetic field signal based on the geomagnetic field to obtain a corrected magnetic field value; Anomaly determination module configured to calculate a magnetic dipole strength change gradient of a current time relative to a previous time based on the current corrected magnetic field value , and determine that an abnormal defect occurs when the magnetic dipole strength change gradient exceeds a set value. a stage identification module configured to calculate the fracture feature extraction and diagnosis based on the obtained spectrum energy ratio for the current corrected magnetic field value when there is an abnormal defect to obtain the abnormal expansion stage of the current defect.

9. A computer readable storage medium having stored thereon computer programs / instructions, characterized in that, The computer program / instructions, when executed by the processor, implement the steps of the wind turbine bolt fracture detection method according to any one of claims 1-6.

10. A computer program product comprising computer programs / instructions, characterized in that, The computer program / instructions, when executed by the processor, implement the steps of the wind turbine bolt fracture detection method according to any one of claims 1-6.

Images