Bearing quality detection system based on artificial intelligence
Through the artificial intelligence detection system, speed sensors and deep learning algorithms are used to automatically detect bearing quality, solving the high subjectivity and speed problems of traditional inspections, and achieving efficient and accurate quality evaluation.
Patent Information
- Application Number
- CN202411433818.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-15
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-10-15
AI Technical Summary
Traditional bearing quality inspection has high subjectivity and misjudgment rate, and the inspection speed is slow, which cannot meet the needs of mass production.
Using an artificial intelligence-based detection system, vibration signals are collected through speed sensors and denoising, combined with deep learning algorithms, features are extracted from bearing surface images, and bearing quality qualified values are calculated to achieve automated detection.
It improves the accuracy and efficiency of bearing quality inspection, meets the needs of mass production, and reduces manual misjudgment.
Smart Images

Figure CN118940126A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of quality inspection technology, and more specifically, to an artificial intelligence-based bearing quality inspection system. Background Technology
[0002] As an important component of rotating machinery, bearings play a vital role in everything from power, chemical metallurgy, military, aerospace, and transportation industries to everyday furniture, appliances, and tools. With the continuous development of industrial manufacturing technology, the quality of bearings, as a key component in mechanical equipment, directly affects the operating efficiency and safety of the entire equipment.
[0003] In recent years, the rapid development of artificial intelligence technology has provided new solutions for bearing quality inspection. By introducing artificial intelligence algorithms and advanced sensor technology, it is possible to accurately acquire and analyze bearing vibration signals and surface images, thereby accurately determining the quality condition of the bearing.
[0004] Traditional bearing quality inspection still has some drawbacks in practical use. For example, traditional bearing quality inspection mainly relies on the experience of workers, which has a high degree of subjectivity and misjudgment rate. In addition, manual inspection is slow and cannot meet the needs of mass production. Summary of the Invention
[0005] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an artificial intelligence-based bearing quality inspection system to address the problems mentioned in the background section.
[0006] To achieve the above objectives, the present invention provides the following technical solution: Data acquisition module: Used to acquire vibration signals of the bearing using a speed sensor, and transmit the acquired vibration data to the data preprocessing module; Data preprocessing module: used to denoise the collected vibration signals and transmit the processed vibration signals to the sound feature value calculation module; Sound feature value calculation module: used to calculate the sound feature value from the processed vibration signal, and transmit the calculated sound feature value to the bearing quality qualification value calculation module; Bearing image acquisition and processing module: used to acquire bearing surface images using a scanner, process the acquired bearing surface images, and transmit the bearing surface images to the feature extraction module; Feature extraction module: Used to extract key features from bearing surface images using deep learning algorithms, calculate the roundness value of the bearing surface using the extracted key features, and transmit the calculated bearing roundness value to the bearing quality qualification value calculation module; Bearing quality qualification value calculation module: It is used to calculate the bearing quality qualification value by combining the sound characteristic value and the bearing roundness, and then compare the calculated bearing quality qualification value with the preset bearing quality qualification value.
[0007] Preferably, in the data acquisition module, the speed sensor is a speed sensor, which is mounted on the bearing housing using a fixing device, and the output terminal of the speed sensor is connected to the data acquisition system computer, and then the sampling rate is set. Let the highest frequency of the vibration signal be... According to the Nyquist theorem, the required minimum sampling rate is Start the data acquisition equipment to begin recording vibration signals, monitor signal quality in real time, and collect vibration data.
[0008] Preferably, in the data preprocessing module, the collected vibration signal is filtered by minimum mean square error to remove noise.
[0009] Preferably, in the sound feature value calculation module, the processed vibration signal is calculated to obtain sound feature values, and the specific method for calculating sound feature values is as follows: Where S represents the sound feature value, Represented as vibration frequency, Represented as sound wave value, Represented as sound index values; The calculated sound characteristic values are transmitted to the bearing quality qualification value calculation module.
[0010] Preferably, in the bearing image acquisition module, the scanner is fixed on the bracket, the focus and exposure parameters are adjusted, the bearing is illuminated with a ring light source, and then the bearing image is scanned 360 degrees by the scanner. The scanned bearing image is subjected to mean filtering to remove noise, and then the mean-filtered and denoised image is subjected to grayscale linear transformation to enhance the image.
[0011] Preferably, in the feature extraction module, the deep learning algorithm selected is the convolutional neural network algorithm. Key features are extracted through the neural network algorithm, including bearing surface roughness, bearing diameter, and bearing width. The specific method for calculating the rounded value of the bearing surface by calculating the key features is as follows: Where Z represents the roundness of the bearing surface, R represents the surface roughness of the bearing, W represents the bearing diameter, and W represents the bearing width.
[0012] Preferably, in the bearing quality qualification value calculation module, the specific method for calculating the bearing quality qualification value is as follows: Where Q represents the bearing quality qualification value, S represents the sound characteristic value, and Z represents the bearing surface roundness value. Represented as the weights of sound feature values, The weight is represented as the rounding value of the bearing surface. The calculated bearing quality qualification value is compared with the preset bearing quality qualification value. If the calculated bearing quality qualification value is greater than or equal to the preset bearing quality qualification value, a bearing quality qualification command is output, and the bearing is determined to be a qualified product and can be used. If the calculated bearing quality qualification value is less than the preset bearing quality qualification value, a bearing quality failure command is output, and the bearing is determined to be a defective product and will be destroyed by professionals.
[0013] The technical effects and advantages of this invention are as follows: This invention uses a speed sensor to collect bearing vibration data and transmits it to a data preprocessing module for noise reduction. Then, the noise-reduced vibration signal is used to calculate sound feature values. Simultaneously, a scanner is used to acquire and process images of the bearing surface. A deep learning algorithm is used to extract key features from the images, and the roundness of the bearing surface is calculated. Finally, the sound feature values and the bearing roundness are combined to calculate the bearing quality pass value, which is then compared with a preset bearing quality pass value, and corresponding instructions are output. This invention improves the accuracy and efficiency of bearing quality inspection by introducing artificial intelligence algorithms and advanced sensor technology, and meets the needs of mass production. Attached Figure Description
[0014] Figure 1 This is a schematic diagram of the system module connections of the present invention. Detailed Implementation
[0015] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0016] Please see Figure 1 As shown, the present invention provides an artificial intelligence-based bearing quality inspection system, including a data acquisition module, a data preprocessing module, a sound feature value calculation module, a bearing image acquisition and processing module, a feature extraction module, and a bearing quality pass value calculation module.
[0017] The data acquisition module is used to collect vibration data of the bearing using a speed sensor and transmit the collected vibration data to the data preprocessing module. In the data acquisition module, the speed sensor is a speed sensor that is mounted on the bearing housing using a fixing device. The output terminal of the speed sensor is connected to the computer of the data acquisition system, and then the sampling rate is set. Let the highest frequency of the vibration signal be... According to the Nyquist theorem, the required minimum sampling rate is Start the data acquisition equipment to begin recording vibration signals, monitor signal quality in real time, and acquire vibration signals; the acquired vibration data includes vibration frequency, sound waves, and vibration signals.
[0018] The data preprocessing module is used to denoise the collected vibration signals and transmit the processed vibration signals to the sound feature value calculation module. In the data preprocessing module, the acquired vibration signal is denoised by minimum mean square error filtering. The denoising method is as follows: The input signal consists of vibration signal and noise. The specific method for calculating the input signal is as follows: Where A represents the input signal, B represents the vibration signal, and C represents the noise; Extract a signal containing the current input signal. The specific method for calculating the input signal window for a given number of points is as follows: Where D represents the input signal window, Let be the signal value at time point a, and T denotes the transpose, converting the row vector within the parentheses into a column vector; Select the filter order, set it to E, and calculate the filter output signal. The specific method for calculating the filter output signal is as follows: ,in, Let D represent the output signal of the filter, D represent the input signal window, and T represent the transpose. It is represented as the filter's weight vector; The error between the desired signal and the filter output is calculated using the following method: Where b represents the error. Represented as the desired signal, This is represented as the output signal of the filter; The desired signal acquisition method can directly measure the signal that is not contaminated by noise as the desired signal; for example, in a laboratory environment, conditions are controlled to reduce the influence of noise. The weights are updated using the LMS algorithm. The specific calculation method for the updated weights is as follows: ,in, Represented as the filter weight vector, This represents the weight before the update. Let b represent the learning rate, b represent the error, and D represent the input signal window. For each new input signal sample 'a', repeat the above continuous calculation process until all data points have been processed. The final generated filter output signal... This is the denoised signal; the denoised signal is visualized and compared with the original signal to verify the denoising effect; if the initial effect is not ideal, the learning rate is adjusted according to the results. The filter order E is adjusted, and the denoising process is rerun.
[0019] The sound feature value calculation module is used to calculate the sound feature value from the processed vibration signal and transmit the calculated sound feature value to the bearing quality qualification value calculation module. In the sound feature value calculation module, the processed vibration signal is calculated to obtain sound feature values. The specific method for calculating sound feature values is as follows: Where S represents the sound feature value, Represented as vibration frequency, Represented as sound wave value, Represented as sound index values; The specific method for calculating the sound index value is as follows: ,in, Represented as sound index values, Represented as the peak value of the vibration signal. Represented as the root mean square value, Expressed as the mean; The specific method for calculating the root mean square value is as follows: ,in, The root mean square value is represented by H, where H represents the number of sampling points. Represented as a vibration signal, ; The specific method for calculating the mean is as follows: ,in, The mean is represented by H, and the number of sampling points is represented by H. Represented as a vibration signal, ; The specific method for calculating the peak value of the vibration signal is as follows: After dividing the vibration signal into g segments, n peak values are found. , ; , This represents the peak value of the vibration signal, and g represents the number of segments in the vibration signal. Represented as the j-th peak; The calculated sound characteristic values are transmitted to the bearing quality qualification value calculation module.
[0020] The bearing image acquisition module is used to acquire bearing surface images using a scanner, process the acquired bearing surface images, and transmit the bearing surface images to the feature extraction module. In the bearing image acquisition module, the scanner is fixed on the bracket, and the focus and exposure parameters are adjusted. The bearing is illuminated by a ring light source, and the bearing image is scanned 360 degrees by the scanner. The scanned bearing image is then subjected to mean filtering to remove noise. The specific method for mean filtering to remove noise is as follows: A1: Filter window set to Size, A2: For each pixel in the image, define a filter window centered on the pixel; this window will cover the pixel and the surrounding pixels. A3: To calculate the average value of pixels within the filter window, the specific method for calculating the average value of pixels within the filter window is as follows: ,in, Let u and v represent the new pixel average, u and v represent the offset relative to the center pixel, and p and q represent a... The filter window, where k represents the value in the vertical direction of the image. Represented as the value in the horizontal direction of the image; A4: Replace the original pixel value at the center of the window with the calculated average value; A5: Repeat steps 2 to 4 for each pixel in the image until the entire image is processed, resulting in the image after noise reduction using mean filtering.
[0021] To enhance an image after mean filtering and denoising, a grayscale linear transformation is applied. The specific method for enhancing an image using grayscale linear transformation is as follows: B1: Determine the minimum gray value of the original image. and maximum grayscale value Minimum grayscale value and maximum grayscale value This is obtained by iterating through all the pixels of the image; B2: Select a target gray level range, i.e., the minimum gray value of the transformed image. and maximum grayscale value For example, for an 8-bit image, the range is 0 to 255. B3: The specific method for calculating the transformed grayscale value is as follows: ,in, Represented as the original image in grayscale value at that location This represents the transformed grayscale value, where m represents the scaling value and r represents the offset. The scaling value is calculated as follows: Where m represents the scaling factor. Represented as the maximum grayscale value of the transformed image. Represented as the minimum grayscale value of the transformed image. Represented as the maximum grayscale value of the original image. Represented as the minimum grayscale value of the original image; The scaling value controls the degree to which the grayscale value is scaled; if Then the range of grayscale values will be expanded, which means increasing the contrast; if Then the range of grayscale values will be compressed, and compression means reducing contrast; if If 1 is selected, the grayscale value remains unchanged and is only affected by r. Where r represents the adjustment factor. Represented as the minimum grayscale value of the transformed image. denoted as the minimum grayscale value of the original image, and m represents the scaling value; Used to adjust the overall brightness of grayscale values. This will increase all grayscale values, making the image brighter; This will reduce all grayscale values, making the image darker; B4: Apply the above transformation to each pixel in the image, and the grayscale processing is complete.
[0022] The feature extraction module is used to extract key features from the bearing surface image using a deep learning algorithm, calculate the roundness value of the bearing surface using the extracted key features, and transmit the calculated bearing roundness value to the bearing quality qualification value calculation module. In the feature extraction module, the deep learning algorithm selected is the convolutional neural network algorithm. Key features are extracted through this neural network algorithm, including bearing surface roughness, bearing diameter, and bearing width. The specific method for calculating the rounded value of the bearing surface from these key features is as follows: Where Z represents the roundness of the bearing surface, R represents the surface roughness of the bearing, W represents the bearing diameter, and W represents the bearing width; The calculated bearing surface rounding value is transmitted to the bearing quality qualification value calculation module.
[0023] The bearing quality qualification value calculation module is used to calculate the bearing quality qualification value by combining the sound characteristic value and the bearing roundness, and to compare the calculated bearing quality qualification value with the preset bearing quality qualification value. The specific method for calculating the bearing quality qualification value in the bearing quality qualification value calculation module is as follows: Where Q represents the bearing quality qualification value, S represents the sound characteristic value, and Z represents the bearing surface roundness value. Represented as the weights of sound feature values, The weight is represented as the rounding value of the bearing surface. The calculated bearing quality qualification value is compared with the preset bearing quality qualification value. If the calculated bearing quality qualification value is greater than or equal to the preset bearing quality qualification value, a bearing quality qualification command is output, and the bearing is determined to be a qualified product and can be used. If the calculated bearing quality qualification value is less than the preset bearing quality qualification value, a bearing quality failure command is output, and the bearing is determined to be a defective product and will be destroyed by professionals.
[0024] The thresholds set in this invention are all set by those skilled in the art and are not limited herein.
[0025] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A bearing quality detection system based on artificial intelligence, characterized in that: include: Data acquisition module: used to collect the vibration signal of the bearing using a speed sensor, and transmit the collected vibration data to the data preprocessing module; Data preprocessing module: used to perform denoising on the collected vibration signal and transmit the processed vibration signal to the sound eigenvalue calculation module; Sound characteristic value calculation module: used to calculate the sound characteristic value from the processed vibration signal, and transmit the calculated sound characteristic value to the bearing quality qualified value calculation module; Bearing image acquisition and processing module: used to acquire bearing surface images using a scanner, process the acquired bearing surface images, and transmit the bearing surface images to the feature extraction module; Feature extraction module: used to extract key features from the bearing surface image using a deep learning algorithm, calculate the extracted key features to obtain the bearing surface roundness value, and transmit the calculated bearing roundness to the bearing quality qualified value calculation module; Bearing quality qualified value calculation module: used to calculate the sound characteristic value and the bearing roundness to obtain the bearing quality qualified value, and to judge the calculated bearing quality qualified value with the preset bearing quality qualified value.
2. The bearing quality detection system based on artificial intelligence according to claim 1 is characterized in that: In the data acquisition module, the speed sensor is mounted on the bearing housing using a fixing device, and the output end of the speed sensor is connected to the data acquisition system computer, and then the sampling rate is set. , assuming the highest frequency of the vibration signal is According to the Nyquist theorem, the minimum sampling rate required is ; Start the data acquisition equipment, start recording vibration signals, monitor signal quality in real time and collect vibration data.
3. The bearing quality detection system based on artificial intelligence according to claim 1 is characterized in that: In the data preprocessing module, the collected vibration signal is subjected to minimum mean square error filtering for denoising. The denoising method is specifically as follows: The input signal consists of vibration signal and noise, so the calculation method of the input signal is as follows: , where A represents the input signal, B represents the vibration signal, and C represents the noise; Extract a containing The calculation method of the input signal window is as follows: , where D represents the input signal window, It is represented as the signal value at the ath time point, and T is represented as the transpose, which converts the row vector in the brackets into a column vector; Select the order of the filter, set the order of the filter to E, and calculate the output signal of the filter. The specific method for calculating the output signal of the filter is: ,in, It represents the output signal of the filter, D represents the input signal window, T represents the transpose, Represented as the weight vector of the filter; Calculate the error between the expected signal and the filter output. The error calculation method is as follows: , where b represents the error, is represented as the expected signal, Represented as the output signal of the filter; Use the LMS algorithm to update the weights. The calculation method for updating the weights is as follows: ,in, Represented as the weight vector of the filter, Represents the weight before updating, It is represented as learning rate, b is represented as error, and D is represented as input signal window; Repeat the above continuous calculation process for each new input signal sample a until all data points are processed; the final filter output signal is is the denoised signal.
4. The bearing quality detection system based on artificial intelligence according to claim 1 is characterized in that: In the sound characteristic value calculation module, the processed vibration signal is calculated to obtain the sound characteristic value, and the sound characteristic value calculation method is specifically as follows: , where S represents the sound feature value, Expressed as the vibration frequency, Expressed as a sound wave value, Expressed as sound index value; The calculated sound characteristic values are transmitted to the bearing quality qualified value calculation module.
5. The bearing quality detection system based on artificial intelligence according to claim 4 is characterized in that: The specific method for calculating the sound index value is as follows: ,in, Expressed as a sound index value, Expressed as the peak value of the vibration signal, Expressed as root mean square value, Expressed as mean.
6. The bearing quality detection system based on artificial intelligence according to claim 1 is characterized in that: In the bearing image acquisition module, the scanner is fixed on the bracket, the focal length and exposure parameters are adjusted, the bearing is illuminated with a ring light source, and then the scanner is used to scan the bearing image 360 degrees; The scanned bearing image is subjected to mean filtering to remove noise, and then the image after mean filtering and denoising is subjected to grayscale linear transformation to enhance the image.
7. The bearing quality detection system based on artificial intelligence according to claim 1 is characterized in that: The specific method of image enhancement by grayscale linear transformation is: B1: Determine the minimum grayscale value of the original image and the maximum gray value , minimum gray value and the maximum gray value It is obtained by traversing all pixels of the image; B2: Select a target grayscale range, which is the minimum grayscale value of the transformed image. and the maximum gray value ; B3: The specific method for calculating the gray value after transformation is: ,in, Represented as the original image in The gray value at It is represented as the gray value after transformation, m is represented as the scaling value, and r is represented as the offset; The calculation method of the scaling value is as follows: , where m is the scaling factor, Represented as the maximum gray value of the transformed image, Represents the minimum gray value of the transformed image, Represented as the maximum gray value of the original image, Represented as the minimum gray value of the original image; The scaling value controls the degree of scaling of the grayscale value; if , the range of grayscale values will be expanded, and expansion means increasing the contrast; if , the range of grayscale values will be compressed, and compression means reducing the contrast; if 1, the gray value remains unchanged and is only affected by r; , where r is the adjustment coefficient, Represents the minimum gray value of the transformed image, It represents the minimum gray value of the original image, and m represents the scaling value; Used to adjust the overall brightness of the grayscale value. When it increases all gray values, it makes the image brighter; When it is turned on, it will reduce all gray values and make the image darker; B4: Apply the above transformation to each pixel in the image, and the grayscale processing is completed.
8. The bearing quality detection system based on artificial intelligence according to claim 1 is characterized in that: In the feature extraction module, the deep learning algorithm selects the convolutional neural network algorithm, and extracts key features through the neural network algorithm. The key features include the surface roughness of the bearing, the bearing diameter, and the bearing width. The calculation method for calculating the roundness value of the bearing surface by calculating the key features is specifically as follows: , where Z represents the roundness of the bearing surface, R represents the roughness of the bearing surface, W represents the bearing diameter, and W represents the bearing width; The calculated bearing surface roundness value is transmitted to the bearing quality qualified value calculation module.
9. The bearing quality detection system based on artificial intelligence according to claim 1 is characterized in that: In the bearing quality qualified value calculation module, the calculation method of the bearing quality qualified value is specifically as follows: , where Q represents the bearing quality qualification value, S represents the sound characteristic value, and Z represents the bearing surface roundness value. Expressed as the weight of the sound feature value, Expressed as the weight of the rounded value of the bearing surface; The calculated bearing quality qualified value is compared with the preset bearing quality qualified value; if the calculated bearing quality qualified value is greater than or equal to the preset bearing quality qualified value, a bearing quality qualified instruction is output, and the bearing is judged to be qualified and can be used; if the calculated bearing quality qualified value is less than the preset bearing quality qualified value, a bearing quality unqualified instruction is output, and the bearing is judged to be unqualified and destroyed by professionals.
Citation Information
Patent Citations
Bearing ball surface flatness detection apparatus based on sound waves
CN106595540A
Bearing fault detection method and system for intelligent main shaft
CN112697434A
Fan blade fault diagnosis method based on improved MFCC algorithm
CN114841193A
Bearing fault diagnosis method based on GADF-GAN-AVOA-CNN
CN115575125A
Printing quality detection method, system and equipment and storage medium
CN117698288A
Cited By
Axle box processing quality detection method based on artificial intelligence
CN120746931A
