Intelligent monitoring system and method for machining defects of inner wall of wind power tower based on cloud computing

Abstract

The invention relates to the technical field of defect detection, and discloses an intelligent monitoring system and method for machining defects of the inner wall of a wind power tower based on cloud computing. According to the method, wind power monitoring devices are installed at the two ends of a wind power tower, sensors are evenly installed on the inner wall of the wind power tower, meanwhile, wind power data are collected in real time based on the installed sensors, the wind power data collected in real time are analyzed in a cloud computing mode, and when it is monitored that machining defects exist in the inner wall of the wind power tower, the machining defects are detected. Laser data of the inner wall of the wind power tower are collected in real time in a laser calibration mode, the laser data collected in real time are analyzed in a cloud computing mode, the machining defect position of the inner wall of the wind power tower is determined, and after the machining defect position of the inner wall of the wind power tower is determined, ultrasonic detection is conducted on the two sides of the machining defect position, and the machining defect position is determined. Ultrasonic detection data are collected in real time and processed and analyzed to determine the size of the machining defect of the inner wall of the wind power tower, so that the intelligence of monitoring the machining defect of the inner wall of the wind power tower is improved.

Description

Technical Field

[0001] The present invention relates to the field of defect detection technology, and in particular to a cloud computing-based intelligent monitoring system and method for machining defects on the inner wall of a wind tower. Background Art

[0002] For the detection of machining defects on the inner wall of wind towers, traditional methods generally use ultrasonic or radiographic testing to inspect the weld area during processing. However, manual inspection is required after the wind tower tube is formed. This has low inspection efficiency and high labor intensity. In addition, due to the problem of manual inspection accuracy, the re-inspection rate of machining defects on the inner wall of wind towers is high, which greatly reduces the service life of wind towers.

[0003] The existing patent application CN111551565A describes a method that uses a motion control system to rotate the tower. Simultaneously, an image acquisition system collects image data of the tower welds during the rotation process. The image acquisition system then pre-processes the image data and transmits it to a defect detection system. The defect detection system then analyzes the processed image data and outputs the tower weld defect type and data. However, since this method relies solely on image acquisition for tower weld defect detection, it overlooks potential defects within the tower welds, resulting in incomplete detection and certain limitations. Summary of the Invention

[0004] (1) Technical problems solved

[0005] In response to the shortcomings of the existing technology, the present invention provides a cloud computing-based intelligent monitoring system and method for wind tower inner wall processing defects, which has the advantages of accuracy, efficiency, and real-time, and solves the problem of high re-inspection rate of wind tower inner wall processing defects.

[0006] (2) Technical solution

[0007] In order to solve the above-mentioned technical problem of high re-inspection rate of defects in the inner wall of the wind tower, the present invention provides the following technical solutions:

[0008] The present invention discloses a cloud computing-based intelligent monitoring method for machining defects on the inner wall of a wind tower, which specifically includes the following steps:

[0009] S1. Install wind monitoring equipment at both ends of the wind tower and evenly install sensors on the inner wall of the wind tower;

[0010] S2. Real-time wind data is collected based on installed sensors, and the collected wind data is analyzed through cloud computing to preliminarily determine whether there are any machining defects on the inner wall of the wind tower;

[0011] S3. When a machining defect is detected on the inner wall of the wind tower, laser data of the inner wall of the wind tower is collected in real time through laser calibration, and the real-time collected laser data is analyzed through cloud computing to locate the location of the machining defect on the inner wall of the wind tower;

[0012] S4. After determining the location of the machining defect on the inner wall of the wind tower, ultrasonic testing is performed on both sides of the machining defect location, and ultrasonic testing data is collected and processed in real time to obtain processed ultrasonic testing data;

[0013] S5. Analyze the processed ultrasonic detection data obtained through cloud computing to determine the size of the processing defects on the inner wall of the wind tower.

[0014] The present invention installs wind monitoring equipment at both ends of a wind tower and evenly installs sensors on the inner wall of the wind tower. Wind data is collected in real time based on the installed sensors, and the real-time collected wind data is analyzed by cloud computing. When processing defects on the inner wall of the wind tower are detected, laser data of the inner wall of the wind tower is collected in real time by laser calibration, and the real-time collected laser data is analyzed by cloud computing to determine the location of the processing defect on the inner wall of the wind tower. After the location of the processing defect on the inner wall of the wind tower is determined, ultrasonic detection is performed on both sides of the processing defect location, and the ultrasonic detection data is collected in real time, processed, and analyzed to determine the size of the processing defect on the inner wall of the wind tower, thereby improving the intelligence of the monitoring of the processing defects on the inner wall of the wind tower.

[0015] Furthermore, installing wind monitoring equipment at both ends of the wind tower and evenly installing sensors on the inner wall of the wind tower includes the following steps:

[0016] Measure the length and diameter of the wind tower, set the measurement interval, evenly install sensors based on the set measurement interval, and record the number and location information of each sensor.

[0017] Preferably, the real-time collection of wind data based on installed sensors and analysis of the real-time collected wind data by cloud computing to preliminarily determine whether there are machining defects on the inner wall of the wind tower include the following steps:

[0018] S21. Select 10 groups of defect-free wind tower inner walls, perform wind force tests on the selected 10 groups of defect-free wind tower inner walls, and output wind speed attenuation curves;

[0019] S22. Based on the output wind speed attenuation curve, the real-time collected wind attenuation data is analyzed through cloud computing to preliminarily determine whether there are any processing defects on the inner wall of the wind tower.

[0020] Preferably, the 10 groups of defect-free wind tower inner walls are selected, and the selected 10 groups of defect-free wind tower inner walls are subjected to wind force test, and the output wind speed attenuation curve comprises the following steps:

[0021] The wind force test size is fixed, and the wind speed attenuation data passing through the defect-free wind tower inner wall is collected based on the installed sensors;

[0022] The collected 10 groups of wind speed attenuation data are fitted to determine the wind speed attenuation curve;

[0023] The Gaussian fitting expression is as follows:

[0024]

[0025] Wherein, n represents the order of Gaussian fitting, the number of fitted data is 10, a i represents the height of curve i, b i represents the center position coordinate of curve i on the x-axis, g i represents the width of curve i, and f(x) represents the wind speed attenuation curve fitted with respect to the x-axis.

[0026] Preferably, based on the output wind speed attenuation curve, the real-time collected wind force attenuation data are analyzed by cloud computing to preliminarily judge whether the wind tower inner wall has processing defects, comprising the following steps:

[0027] The real-time collected wind force attenuation data are input into the cloud platform, and based on each interval wind force attenuation data, the wind force attenuation curve is constructed in real time by cloud computing;

[0028] The expression of the wind force attenuation curve constructed by cloud computing is as follows:

[0029]

[0030] Wherein, f(x1) represents the constructed wind force attenuation curve, d j represents the wind force attenuation data detected by the jth sensor, and m represents the number of sensors;

[0031] The real-time constructed wind force attenuation curve is compared with the fitted wind speed attenuation curve, a comparison fluctuation threshold is set, and when the comparison fluctuation threshold of the real-time constructed wind force attenuation curve and the fitted wind speed attenuation curve is greater than the set comparison fluctuation threshold, it indicates that the current wind tower inner wall has processing defects, otherwise it does not have processing defects.

[0032] The present invention selects 10 groups of wind data of defect-free wind tower inner walls as references, fits the reference wind data, determines the wind data standard, and compares the determined wind data standard with the real-time collected wind data to preliminarily judge whether there are processing defects on the wind tower inner wall, thereby improving the inspection efficiency of the wind tower inner wall.

[0033] Preferably, when a machining defect is detected on the inner wall of the wind tower, real-time laser data of the inner wall of the wind tower is collected by laser calibration, and the real-time collected laser data is analyzed by cloud computing to locate the position of the machining defect on the inner wall of the wind tower, which includes the following steps:

[0034] S31. Real-time collection of laser data on the inner wall of the wind tower through laser calibration;

[0035] A set of lasers with fixed intensity is emitted from the wind tower, and the installed sensors collect the laser reflection data from the inner wall of the wind tower;

[0036] The incident laser reflection is set to include Fresnel reflection and Rayleigh scattering;

[0037] It is assumed that when there are no processing defects in the reflection area of ​​the inner wall of the wind tower, the Fresnel reflection power is greater than the Rayleigh scattering; when there are processing defects in the reflection area of ​​the inner wall of the wind tower, the Fresnel reflection power is less than the Rayleigh scattering;

[0038] S32. Analyze the real-time collected laser data through cloud computing to locate the processing defects on the inner wall of the wind tower.

[0039] Preferably, the analyzing of the real-time collected laser data by cloud computing to locate the position of the machining defect on the inner wall of the wind tower comprises the following steps:

[0040] The laser data collected by the sensor in real time is uploaded to the cloud platform, where it is calculated and compared, and the location of machining defects on the inner wall of the wind tower is located based on the comparison results;

[0041] The calculation formula of laser Fresnel reflection power is as follows:

[0042] P r (z) = S × P j ×e -2μ ;

[0043] Where S is the backscatter coefficient, P j represents the reflected power detected by the jth sensor, e is a natural constant, μ is the Fresnel reflection coefficient, P r (z) represents the Fresnel reflection power at point z on the inner wall of the wind tower, P r represents the Fresnel reflection power;

[0044] The calculation formula of laser Rayleigh scattering power is as follows:

[0045]

[0046] Where S is the backscatter coefficient, P j represents the scattered power detected by the jth sensor, e is a natural constant, θ represents the Rayleigh scattering coefficient, η represents the refractive index, τ represents the optical pulse width, c is the speed of light, P b (z) represents the Rayleigh scattering coefficient at point z on the inner wall of the wind tower.

[0047] The present invention collects laser data of the inner wall of the wind tower in real time through laser calibration, and analyzes the real-time collected laser data through cloud computing. When the wind data test monitors the existence of processing defects on the inner wall of the wind tower, the position of the processing defect on the inner wall of the wind tower is determined by comparing the power of Fresnel reflection and Rayleigh scattering on the inner wall of the wind tower, thereby improving the accuracy of locating the position of the processing defect on the inner wall of the wind tower.

[0048] Preferably, after determining the location of the machining defect on the inner wall of the wind tower, ultrasonic testing is performed on both sides of the machining defect location, and ultrasonic testing data is collected and processed in real time to obtain the processed ultrasonic testing data, which includes the following steps:

[0049] S41. Acquire an ultrasonic image to determine the location of machining defects on the inner wall of the wind tower through ultrasonic testing;

[0050] S42, processing the real-time collected ultrasonic image to obtain a processed ultrasonic image;

[0051] S43. Separate the processed ultrasound image using a background separation algorithm.

[0052] Preferably, the step of processing the real-time acquired ultrasound image to obtain the processed ultrasound image comprises the following steps:

[0053] By adopting the wavelet threshold algorithm, the real-time collected ultrasound image data is subjected to noise reduction processing by setting the threshold;

[0054] The real-time collected ultrasound image is transformed by a wavelet transform function to obtain a corresponding ultrasound image;

[0055] The wavelet transform function is as follows:

[0056]

[0057] Among them, h is the scaling variable, τ is the translation variable, r represents the time-frequency of the waveform, ψ represents the wavelet transform function, and R represents a real number;

[0058] The ultrasound image is divided by setting a threshold, and the ultrasound image area smaller than the threshold is removed;

[0059] The ultrasound image after aggregation and division is set as the processed ultrasound image.

[0060] Preferably, separating the processed ultrasound image obtained by using a background separation algorithm comprises the following steps:

[0061] An initial grayscale threshold k is selected to classify all pixels in the processed ultrasound image into two categories C1 and C2;

[0062] Set C1 to be the pixel category that is less than or equal to the grayscale threshold k, and C2 to be the pixel category that is greater than the grayscale threshold k;

[0063] Set the grayscale mean of pixel category C1 to w1, the grayscale mean of pixel category C2 to w2, and the global grayscale mean to w3;

[0064] Assume that the probability of a pixel in the processed ultrasound image belonging to pixel category C1 is q1, and the probability of a pixel belonging to pixel category C2 is p2;

[0065] The binarization formula is as follows:

[0066] w3=w1×q1+w2×q2;

[0067]

[0068] Among them, θ represents the binarization threshold;

[0069] The grayscale values ​​greater than the binarization threshold are set to 255, and the grayscale values ​​less than or equal to the binarization threshold are set to 0;

[0070] Summarize the pixels in the binarized fire image to obtain the separated ultrasonic image;

[0071] Performing contour extraction on the separated ultrasonic image to obtain the contour of the ultrasonic image after background separation;

[0072] The contour of the ultrasound image after background separation is set as the processed ultrasound detection data.

[0073] The present invention performs ultrasonic testing on both sides of the machining defect position on the inner wall of the wind tower after determining the machining defect position, collects ultrasonic testing data in real time to determine an ultrasonic image of the machining defect position on the inner wall of the wind tower, and simultaneously filters and separates the background of the real-time collected ultrasonic image to determine the machining defect contour, thereby improving the reliability of determining the machining defect contour of the inner wall of the wind tower.

[0074] Further, based on the obtained processed ultrasonic detection data, the size of the processing defect of the inner wall of the wind tower is determined by analyzing the processed ultrasonic detection data through cloud computing, comprising the following steps:

[0075] Two reference points are set in the obtained processed ultrasonic detection data, the distance between the two reference points in the image data is calculated, and the distance represented by each pixel point is determined based on the ratio of the distance between the image and the wind tower sensor, and the size of the processing defect of the inner wall of the wind tower is determined based on the distance represented by each pixel point.

[0076] The application also discloses a cloud computing-based intelligent monitoring system for processing defects of the inner wall of a wind tower, which is used to realize the cloud computing-based intelligent monitoring method for processing defects of the inner wall of a wind tower, and the system comprises a sensor data acquisition module, a data processing module, a defect positioning module and a cloud computing platform.

[0077] The sensor data acquisition module is used to acquire wind data, laser data and ultrasonic detection data in real time.

[0078] The data processing module is used to process the real-time acquired data.

[0079] The cloud computing platform is used to analyze the real-time acquired wind data, laser data and ultrasonic detection data.

[0080] The defect positioning module is used to position the processing defect of the inner wall of the wind tower according to the analysis result.

[0081] (Three) beneficial effects

[0082] Compared with the prior art, the application provides a cloud computing-based intelligent monitoring system and method for processing defects of the inner wall of a wind tower, which has the following beneficial effects:

[0083] 1. The application installs wind monitoring equipment at both ends of the wind tower, uniformly installs sensors on the inner wall of the wind tower, acquires wind data in real time based on the installed sensors, analyzes the real-time acquired wind data through cloud computing, detects the processing defect of the inner wall of the wind tower through laser calibration, acquires laser data of the inner wall of the wind tower in real time, analyzes the real-time acquired laser data through cloud computing, determines the position of the processing defect of the inner wall of the wind tower, and then detects the processing defect of the inner wall of the wind tower through ultrasonic detection on both sides of the processing defect position, acquires ultrasonic detection data in real time, processes and analyzes the ultrasonic detection data to determine the size of the processing defect of the inner wall of the wind tower, and improves the intelligence of the monitoring of the processing defect of the inner wall of the wind tower.

[0084] 2. The invention selects 10 groups of wind data from defect-free wind tower inner walls as references, fits the reference wind data, determines the wind data standard, and compares the determined wind data standard with the real-time collected wind data to preliminarily determine whether there are processing defects on the inner wall of the wind tower, thereby improving the inspection efficiency of the inner wall of the wind tower.

[0085] 3. This invention collects the laser data of the inner wall of the wind tower in real time through laser calibration, and analyzes the real-time collected laser data through cloud computing. When the wind data test monitors the existence of processing defects on the inner wall of the wind tower, the position of the processing defect on the inner wall of the wind tower is determined by comparing the power of Fresnel reflection and Rayleigh scattering on the inner wall of the wind tower, thereby improving the accuracy of locating the position of the processing defect on the inner wall of the wind tower.

[0086] 4. This invention performs ultrasonic testing on both sides of the processing defect position after determining the processing defect position on the inner wall of the wind tower, collects ultrasonic testing data in real time to determine the ultrasonic image of the processing defect position on the inner wall of the wind tower, and simultaneously filters and separates the background of the real-time collected ultrasonic image to determine the processing defect contour, thereby improving the reliability of determining the processing defect contour of the inner wall of the wind tower. BRIEF DESCRIPTION OF THE DRAWINGS

[0087] Figure 1 It is a structural diagram of the intelligent monitoring process for machining defects on the inner wall of a wind tower according to the present invention. DETAILED DESCRIPTION

[0088] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0089] Example 1

[0090] See also Figure 1 This embodiment discloses a cloud computing-based intelligent monitoring method for wind tower inner wall machining defects, which specifically includes the following steps:

[0091] S1. Install wind monitoring equipment at both ends of the wind tower and evenly install sensors on the inner wall of the wind tower;

[0092] S2. Real-time wind data is collected based on installed sensors, and the collected wind data is analyzed through cloud computing to preliminarily determine whether there are any machining defects on the inner wall of the wind tower;

[0093] S3. When a machining defect is detected on the inner wall of the wind tower, laser data of the inner wall of the wind tower is collected in real time through laser calibration, and the real-time collected laser data is analyzed through cloud computing to locate the location of the machining defect on the inner wall of the wind tower;

[0094] S4. After determining the location of the machining defect on the inner wall of the wind tower, ultrasonic testing is performed on both sides of the machining defect location, and ultrasonic testing data is collected and processed in real time to obtain processed ultrasonic testing data;

[0095] S5. Analyzing the processed ultrasonic testing data obtained by cloud computing to determine the size of the machining defects on the inner wall of the wind tower;

[0096] Further, see Figure 1 , installing wind monitoring equipment at both ends of the wind tower and evenly installing sensors on the inner wall of the wind tower includes the following steps:

[0097] Measure the length and diameter of the wind tower, set the measurement interval, evenly install sensors based on the set measurement interval, and record the number and location information of each sensor;

[0098] Further, see Figure 1 , based on the installed sensors to collect wind data in real time, and analyze the real-time collected wind data through cloud computing, the preliminary judgment of whether there are processing defects on the inner wall of the wind tower includes the following steps:

[0099] S21. Select 10 groups of defect-free wind tower inner walls, perform wind force tests on the selected 10 groups of defect-free wind tower inner walls, and output wind speed attenuation curves;

[0100] The wind test size is set to be fixed, and the installed sensors collect wind speed attenuation data through the inner wall of the wind tower without defects;

[0101] Fit the 10 sets of wind speed attenuation data collected to determine the wind speed attenuation curve;

[0102] The Gaussian fitting expression is as follows:

[0103]

[0104] Among them, n represents the order of Gaussian fitting, and the number of fitted data is 10, a i represents the height of curve i, b i Indicates the coordinates of the center position of curve i on the x-axis, g i represents the width of curve i, and f(x) represents the wind speed attenuation curve after fitting about the x-axis;

[0105] S22. Based on the output wind speed attenuation curve, analyze the real-time collected wind attenuation data through cloud computing to preliminarily determine whether there are any processing defects on the inner wall of the wind tower;

[0106] The real-time collected wind attenuation data is input into the cloud platform, and based on the wind attenuation data of each interval, a wind attenuation curve is constructed in real time through cloud computing;

[0107] The expression of wind attenuation curve constructed by cloud computing is as follows:

[0108]

[0109] Among them, f(x1) represents the constructed wind attenuation curve, d j represents the wind attenuation data detected by the jth sensor, and m represents the number of sensors;

[0110] The real-time constructed wind attenuation curve is compared with the fitted wind speed attenuation curve, and a comparison fluctuation threshold is set. When the comparison fluctuation threshold between the real-time constructed wind attenuation curve and the fitted wind speed attenuation curve is greater than the set comparison fluctuation threshold, it indicates that there is a processing defect in the inner wall of the current wind tower; otherwise, there is no processing defect;

[0111] Further, see Figure 1 When a machining defect is detected on the inner wall of a wind tower, laser data of the inner wall of the wind tower is collected in real time through laser calibration. The real-time collected laser data is analyzed through cloud computing to locate the machining defect on the inner wall of the wind tower. The following steps are included:

[0112] S31. Real-time collection of laser data on the inner wall of the wind tower through laser calibration;

[0113] A set of lasers with fixed intensity is emitted from the wind tower, and the installed sensors collect the laser reflection data from the inner wall of the wind tower;

[0114] The incident laser reflection is set to include Fresnel reflection and Rayleigh scattering;

[0115] It is assumed that when there are no processing defects in the reflection area of ​​the inner wall of the wind tower, the Fresnel reflection power is greater than the Rayleigh scattering; when there are processing defects in the reflection area of ​​the inner wall of the wind tower, the Fresnel reflection power is less than the Rayleigh scattering;

[0116] S32. Analyze the real-time collected laser data through cloud computing to locate the position of machining defects on the inner wall of the wind tower;

[0117] The laser data collected by the sensor in real time is uploaded to the cloud platform, where it is calculated and compared, and the location of machining defects on the inner wall of the wind tower is located based on the comparison results;

[0118] The calculation formula of laser Fresnel reflection power is as follows:

[0119] P r (z) = S × P j ×e -2μ ;

[0120] Where S is the backscatter coefficient, P j represents the reflected power detected by the jth sensor, e is a natural constant, μ is the Fresnel reflection coefficient, P r (z) represents the Fresnel reflection power at point z on the inner wall of the wind tower, P r represents the Fresnel reflection power;

[0121] The calculation formula of laser Rayleigh scattering power is as follows:

[0122]

[0123] Where S is the backscatter coefficient, P j represents the scattered power detected by the jth sensor, e is a natural constant, θ represents the Rayleigh scattering coefficient, η represents the refractive index, τ represents the optical pulse width, c is the speed of light, P b (z) represents the Rayleigh scattering coefficient at point z on the inner wall of the wind tower;

[0124] Further, see Figure 1 After determining the location of the machining defect on the inner wall of the wind tower, ultrasonic testing is performed on both sides of the machining defect location, and ultrasonic testing data is collected and processed in real time to obtain processed ultrasonic testing data;

[0125] S41. Acquire an ultrasonic image to determine the location of machining defects on the inner wall of the wind tower through ultrasonic testing;

[0126] S42, processing the real-time collected ultrasonic image to obtain a processed ultrasonic image;

[0127] By adopting the wavelet threshold algorithm, the real-time collected ultrasound image data is subjected to noise reduction processing by setting the threshold;

[0128] The real-time collected ultrasound image is transformed by a wavelet transform function to obtain a corresponding ultrasound image;

[0129] The wavelet transform function is as follows:

[0130]

[0131] Among them, h is the scaling variable, τ is the translation variable, r represents the time-frequency of the waveform, ψ represents the wavelet transform function, and R represents a real number;

[0132] Further, the ultrasound image is divided by setting a threshold value, and an ultrasound image region smaller than the threshold value is removed;

[0133] Further, the ultrasound image after division is set as a processed ultrasound image;

[0134] S43, the obtained processed ultrasound image is separated by a background separation algorithm;

[0135] An initial gray threshold value k is selected to divide all pixels in the processed ultrasound image into two categories C1 and C2;

[0136] C1 is set as a pixel category less than or equal to the gray threshold value k, and C2 is set as a pixel category greater than the gray threshold value k;

[0137] The mean gray value of the pixel category C1 is set as w1, the mean gray value of the pixel category C2 is set as w2, and the global mean gray value is set as w3;

[0138] The probability that a pixel in the processed ultrasound image belongs to the pixel category C1 is set as q1, and the probability that a pixel belongs to the pixel category C2 is set as p2;

[0139] The binarization processing formula is as follows:

[0140] w3 = w1 x q1 + w2 x q2;

[0141]

[0142] Where θ represents the binarization threshold value;

[0143] The gray value greater than the binarization threshold value is set as 255, and the gray value less than or equal to the binarization threshold value is set as 0;

[0144] The pixels in the fire image after binarization processing are summarized to obtain a separated ultrasound image;

[0145] The contour of the separated ultrasound image is extracted to obtain the contour of the background-separated ultrasound image;

[0146] The contour of the background-separated ultrasound image is set as the processed ultrasound detection data;

[0147] Further, please refer to Figure 1 , based on the processed ultrasound detection data obtained by cloud computing, the size of the wind turbine tower inner wall processing defect is determined, including the following steps:

[0148] Two reference points are set in the processed ultrasonic test data, and the distance between the two reference points in the image data is calculated. At the same time, based on the ratio of the distance between the image and the wind tower sensor, the distance represented by each pixel is determined, and the size of the machining defect on the inner wall of the wind tower is determined based on the determined distance represented by each pixel;

[0149] Example 2

[0150] See also Figure 1 This embodiment also discloses a cloud computing-based intelligent monitoring system for wind tower inner wall machining defects, which is used to implement a cloud computing-based intelligent monitoring method for wind tower inner wall machining defects. The system includes: a sensor data acquisition module, a data processing module, a defect location module, and a cloud computing platform;

[0151] The sensor data acquisition module is used to collect wind data, laser data and ultrasonic detection data in real time;

[0152] The data processing module is used to process the data collected in real time;

[0153] The cloud computing platform is used to analyze the real-time collected wind data, laser data and ultrasonic detection data;

[0154] The defect locating module is used to locate machining defects on the inner wall of the wind tower according to the analysis results.

[0155] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A cloud computing-based intelligent monitoring method for wind tower inner wall machining defects, characterized in that: The following steps are involved: S1. Install wind monitoring equipment at both ends of the wind tower and evenly install sensors on the inner wall of the wind tower; S2. Real-time wind data is collected based on installed sensors, and the collected wind data is analyzed through cloud computing to preliminarily determine whether there are any machining defects on the inner wall of the wind tower; S3. When a machining defect is detected on the inner wall of the wind tower, laser data of the inner wall of the wind tower is collected in real time through laser calibration, and the real-time collected laser data is analyzed through cloud computing to locate the location of the machining defect on the inner wall of the wind tower; S4. After determining the location of the machining defect on the inner wall of the wind tower, ultrasonic testing is performed on both sides of the machining defect location, and ultrasonic testing data is collected and processed in real time to obtain processed ultrasonic testing data; S5. Analyze the processed ultrasonic detection data obtained through cloud computing to determine the size of the processing defects on the inner wall of the wind tower.

2. The method for intelligently monitoring machining defects of wind tower inner walls based on cloud computing according to claim 1, characterized in that: The real-time collection of wind data based on the installed sensors and analysis of the real-time collected wind data by cloud computing to preliminarily determine whether there are machining defects on the inner wall of the wind tower include the following steps: S21. Select 10 groups of defect-free wind tower inner walls, perform wind force tests on the selected 10 groups of defect-free wind tower inner walls, and output wind speed attenuation curves; S22. Based on the output wind speed attenuation curve, the real-time collected wind attenuation data is analyzed through cloud computing to preliminarily determine whether there are any processing defects on the inner wall of the wind tower.

3. The method for intelligently monitoring machining defects of wind tower inner wall based on cloud computing according to claim 2, characterized in that: The step of selecting 10 groups of defect-free wind tower inner walls, performing wind force tests on the selected 10 groups of defect-free wind tower inner walls, and outputting wind speed attenuation curves comprises the following steps: The wind test size is set to be fixed, and the installed sensors collect wind speed attenuation data through the inner wall of the wind tower without defects; Fit the 10 sets of wind speed attenuation data collected to determine the wind speed attenuation curve; The Gaussian fitting expression is as follows: Among them, n represents the order of Gaussian fitting, and the number of fitted data is 10, a i represents the height of curve i, b i Indicates the coordinates of the center position of curve i on the x-axis, g i represents the width of curve i, and f(x) represents the wind speed attenuation curve after fitting about the x-axis.

4. The method for intelligently monitoring machining defects of wind tower inner walls based on cloud computing according to claim 2, characterized in that: The method of analyzing the real-time collected wind attenuation data based on the output wind speed attenuation curve by cloud computing to preliminarily determine whether there are machining defects on the inner wall of the wind tower includes the following steps: The real-time collected wind attenuation data is input into the cloud platform, and based on the wind attenuation data of each interval, a wind attenuation curve is constructed in real time through cloud computing; The expression of wind attenuation curve constructed by cloud computing is as follows: Among them, f(x1) represents the constructed wind attenuation curve, d j represents the wind attenuation data detected by the jth sensor, and m represents the number of sensors; The real-time constructed wind attenuation curve is compared with the fitted wind speed attenuation curve, and a comparison fluctuation threshold is set. When the comparison fluctuation threshold between the real-time constructed wind attenuation curve and the fitted wind speed attenuation curve is greater than the set comparison fluctuation threshold, it indicates that there is a processing defect in the inner wall of the current wind tower, otherwise there is no processing defect.

5. The method for intelligently monitoring machining defects of wind tower inner wall based on cloud computing according to claim 1, characterized in that: When a machining defect is detected on the inner wall of a wind tower, laser data of the inner wall of the wind tower is collected in real time by laser calibration, and the real-time collected laser data is analyzed by cloud computing to locate the position of the machining defect on the inner wall of the wind tower, including the following steps: S31. Real-time collection of laser data on the inner wall of the wind tower through laser calibration; A set of lasers with fixed intensity is emitted from the wind tower, and the installed sensors collect the laser reflection data from the inner wall of the wind tower; The incident laser reflection is set to include Fresnel reflection and Rayleigh scattering; It is assumed that when there are no processing defects in the reflection area of ​​the inner wall of the wind tower, the Fresnel reflection power is greater than the Rayleigh scattering; when there are processing defects in the reflection area of ​​the inner wall of the wind tower, the Fresnel reflection power is less than the Rayleigh scattering; S32. Analyze the real-time collected laser data through cloud computing to locate the processing defects on the inner wall of the wind tower.

6. The method for intelligently monitoring machining defects of wind tower inner wall based on cloud computing according to claim 5, characterized in that: The analysis of the real-time collected laser data by cloud computing to locate the position of the machining defect on the inner wall of the wind tower includes the following steps: The laser data collected by the sensor in real time is uploaded to the cloud platform, where it is calculated and compared, and the location of machining defects on the inner wall of the wind tower is located based on the comparison results; The calculation formula of laser Fresnel reflection power is as follows: P r (z)=S×P j ×e -2μ ; Where S is the backscatter coefficient, P j represents the reflected power detected by the jth sensor, e is a natural constant, μ is the Fresnel reflection coefficient, P r (z) represents the Fresnel reflection power at point z on the inner wall of the wind tower, P r represents the Fresnel reflection power; The calculation formula of laser Rayleigh scattering power is as follows: Where S is the backscatter coefficient, P j represents the scattered power detected by the jth sensor, e is a natural constant, θ is the Rayleigh scattering coefficient, η is the refractive index, τ is the optical pulse width, c is the speed of light, P b (z) represents the Rayleigh scattering coefficient at point z on the inner wall of the wind tower.

7. The method for intelligently monitoring machining defects of wind tower inner wall based on cloud computing according to claim 1, characterized in that: After the machining defect position of the inner wall of the wind tower is determined, ultrasonic testing is performed on both sides of the machining defect position, and ultrasonic testing data is collected and processed in real time to obtain the processed ultrasonic testing data, which includes the following steps: S41. Acquire an ultrasonic image to determine the location of machining defects on the inner wall of the wind tower through ultrasonic testing; S42, processing the real-time collected ultrasonic image to obtain a processed ultrasonic image; S43. Separate the processed ultrasound image using a background separation algorithm.

8. The method for intelligently monitoring machining defects of wind tower inner walls based on cloud computing according to claim 7, characterized in that: Processing the real-time collected ultrasound image to obtain the processed ultrasound image comprises the following steps: By adopting the wavelet threshold algorithm, the real-time collected ultrasound image data is subjected to noise reduction processing by setting the threshold; The real-time collected ultrasound image is transformed by a wavelet transform function to obtain a corresponding ultrasound image; The wavelet transform function is as follows: Among them, h is the scaling variable, τ is the translation variable, r represents the time-frequency of the waveform, ψ represents the wavelet transform function, and R represents a real number; The ultrasound image is divided by setting a threshold, and the ultrasound image area smaller than the threshold is removed; The ultrasound image after aggregation and division is set as the processed ultrasound image.

9. The method for intelligently monitoring machining defects of wind tower inner walls based on cloud computing according to claim 7, characterized in that: The process of separating the processed ultrasound image obtained by the background separation algorithm comprises the following steps: An initial grayscale threshold k is selected to classify all pixels in the processed ultrasound image into two categories C1 and C2; Set C1 to be the pixel category that is less than or equal to the grayscale threshold k, and C2 to be the pixel category that is greater than the grayscale threshold k; Set the grayscale mean of pixel category C1 to w1, the grayscale mean of pixel category C2 to w2, and the global grayscale mean to w3; Assume that the probability of a pixel in the processed ultrasound image belonging to pixel category C1 is q1, and the probability of a pixel belonging to pixel category C2 is p2; The binarization formula is as follows: w3=w1×q1+w2×q2; Among them, θ represents the binarization threshold; The grayscale values ​​greater than the binarization threshold are set to 255, and the grayscale values ​​less than or equal to the binarization threshold are set to 0; Summarize the pixels in the binarized fire image to obtain the separated ultrasonic image; Performing contour extraction on the separated ultrasonic image to obtain the contour of the ultrasonic image after background separation; The contour of the ultrasound image after background separation is set as the processed ultrasound detection data.

10. A system for implementing the cloud computing-based intelligent monitoring method for wind tower inner wall machining defects according to any one of claims 1 to 9, characterized in that: It includes sensor data acquisition module, data processing module, defect location module and cloud computing platform; The sensor data acquisition module is used to collect wind data, laser data and ultrasonic detection data in real time; The data processing module is used to process the data collected in real time; The cloud computing platform is used to analyze the real-time collected wind data, laser data and ultrasonic detection data; The defect locating module is used to locate machining defects on the inner wall of the wind tower according to the analysis results.

Images