3D printing anomaly detection and early warning method based on machine vision

The method leverages machine vision and historical torque analysis to enhance real-time anomaly detection and adjustment in 3D printing concrete, addressing inefficiencies in manual control and single detection methods, thus improving printing quality and reducing waste.

CN119067976AActive Publication Date: 2024-12-03QINGDAO AGRI UNIV

Patent Information

Application Number
CN202411562066.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-05
Publication Date
2024-12-03
Estimated Expiration
2044-11-05

AI Technical Summary

Technical Problem

Existing 3D printing concrete technologies face challenges in efficiently and accurately detecting anomalies in real-time due to the limitations of manual quality control and the inflexibility of single detection methods, leading to reduced printing quality and increased material waste.

Method used

A method utilizing machine vision for anomaly detection, involving image preprocessing, crack and width analysis, and real-time parameter adjustment based on a comprehensive error function, combined with historical torque sequence analysis to update detection intervals.

Benefits of technology

Enhances the accuracy and real-time performance of anomaly detection, allowing for timely adjustments to printing parameters, thereby improving the quality and efficiency of 3D printing processes.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119067976A_ABST
    Figure CN119067976A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of 3D printing anomaly detection, in particular to a 3D printing anomaly detection and early warning method based on machine vision. The method comprises the following steps: processing a to-be-detected image to respectively obtain the width and the crack area of a concrete strip, and further obtaining a crack degree parameter; whether printing is abnormal or not is judged according to the width of the concrete strip and the crack area; if printing is abnormal, a comprehensive error function is obtained based on the crack degree parameter and the width of the concrete strip; parameters of the 3D printing equipment are adjusted through the comprehensive error function; torque values in the current preset time period are collected to form a current torque sequence; forming a historical torque sequence by using the historical torque values when printing is abnormal; calculating the similarity between the current torque sequence and the historical torque sequence; updating an initial detection interval of image detection according to the similarity; and performing subsequent detection according to the updated initial detection interval. The concrete 3D printing quality can be improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of 3D printing anomaly detection, and in particular to a 3D printing anomaly detection and early warning method based on machine vision. Background Art

[0002] In recent years, 3D printing technology, as a cutting-edge manufacturing technology, has shown great application potential in many fields such as aerospace, medical devices and construction with its unique layer-by-layer stacking molding method. In particular, 3D printing concrete technology, as an innovative method in the field of construction, has received widespread attention from academia and industry in recent years. However, in 3D printing concrete technology, the difficulty of the printing system to independently judge and adjust the printing quality in time has always been one of the main challenges facing this technology. In the actual printing process, due to various reasons such as material properties, equipment accuracy, and environmental factors, abnormal problems such as inter-layer misalignment, uneven filling, and support structure failure often occur. These problems not only reduce the yield and quality of printed parts, but also increase the cost and time of post-processing.

[0003] Traditional quality control methods mainly rely on manual observation and empirical judgment, which is not only inefficient, but also prone to material waste and unstable printing quality. Specifically, when the printed strip is too wide or too narrow, cracked, etc., manual intervention is often required to adjust the printing parameters. This lagging adjustment method is difficult to meet the needs of large-scale applications. In addition, with the development of 3D printing concrete technology, the increase in printing speed and complexity makes manual detection more difficult and impractical. At the same time, with the development of machine vision technology, people began to explore the application of machine vision technology in 3D printing concrete technology anomaly detection. Machine vision, as a technology that uses computers and image processing technology to simulate human visual functions, has the advantages of non-contact, high precision, and strong real-time performance. By introducing machine vision technology, the whole process of 3D printing concrete technology can be monitored, and potential abnormal problems can be discovered and warned in time. It has also become a trend to use machine vision to detect defects in 3D printing concrete technology. However, when using machine vision for detection, there is usually only one detection method, which lacks flexibility and may lead to inaccurate detection. Summary of the invention

[0004] In order to solve the above technical problems, the purpose of the present invention is to provide a 3D printing anomaly detection and early warning method based on machine vision. The technical solutions adopted are as follows: An embodiment of the present invention provides a 3D printing anomaly detection and early warning method based on machine vision, the method comprising: Setting the initial detection interval of image detection; obtaining image data of concrete strips during concrete printing and preprocessing to obtain an image to be detected; The image to be detected is processed to obtain the width of the concrete strip and the crack area respectively; the crack degree parameter is obtained according to the crack area; and whether the printing is abnormal is determined according to the width of the concrete strip and the crack area; If printing is abnormal, an early warning is issued and a comprehensive error function is obtained based on the crack degree parameter and the width of the concrete strip; the parameters of the 3D printing equipment are adjusted using the comprehensive error function; Collect the torque values ​​of the current preset period to form a current torque sequence; use the historical torque values ​​when printing abnormalities to form a historical torque sequence; calculate the similarity between the current torque sequence and the historical torque sequence; The initial detection interval of the image detection is updated according to the similarity; and subsequent detection is performed according to the updated initial detection interval.

[0005] Preferably, the image data of the concrete strip during the concrete printing process is obtained and preprocessed to obtain the image to be detected, including: Bilateral filtering is used to reduce the noise of image data to obtain a denoised image; the denoised image is processed using histogram equalization to obtain an image to be detected.

[0006] Preferably, the image to be detected is processed to obtain the width of the concrete strip and the crack area respectively, including: The image to be detected is processed using the Canny edge detection algorithm to extract the edge of the concrete strip; the width of the concrete strip is calculated according to the edge of the concrete strip, wherein the width of the concrete strip includes the maximum value and the minimum value of the width; The image to be detected is analyzed by morphological operation and connected domain analysis to obtain the crack area; the sum of the areas of all crack regions is the crack area.

[0007] Preferably, judging whether printing is abnormal based on the width of the concrete strip and the crack area includes: A first width threshold, a second width threshold and an area threshold are set; when the maximum value of the width of the concrete strip is greater than the first width threshold or the minimum value of the width of the concrete strip is less than the second width threshold, the concrete printing is abnormal; when the crack area of ​​the concrete strip is greater than or equal to the area threshold, the concrete printing is abnormal.

[0008] Preferably, the comprehensive error function is specifically: , in, represents the comprehensive error function; t represents the time t; represents the crack area at time t; Indicates the maximum or minimum value of the width of the concrete strip at time t; Indicates the target width of the concrete strip; represents the crack error weight coefficient, Represents the width error weight coefficient.

[0009] Preferably, the parameters of the 3D printing device are adjusted using the comprehensive error function, including: A parameter adjustment equation group is constructed using the comprehensive error function, and the parameters of the 3D printing device are adjusted using the parameter adjustment equation group, wherein the parameter adjustment equation group includes a print head movement speed adjustment function and a stirring motor stirring speed adjustment function.

[0010] Preferably, the parameter adjustment equation group is specifically: , in, and They represent the print head moving speed adjustment function and the stirring motor stirring speed adjustment function respectively; Indicates the base print head moving speed; Indicates the stirring speed of the base stirring motor; and They represent the print head moving speed adjustment coefficient and the stirring speed adjustment coefficient respectively; represents the comprehensive error function.

[0011] Preferably, calculating the similarity between the current torque sequence and the historical torque sequence includes: The dynamic time warping algorithm is used to calculate the distance between the current torque sequence and the historical torque sequence, which is recorded as the first distance; the mean value, standard deviation, skewness and kurtosis of the current torque sequence are obtained to form the current feature vector; the mean value, standard deviation, skewness and kurtosis of the historical torque sequence are obtained to form the historical feature vector; the distance between the current feature vector and the historical feature vector is calculated, which is recorded as the second distance; the similarity between the current torque sequence and the historical torque sequence is obtained based on the first distance and the second distance.

[0012] Preferably, the similarity between the current torque sequence and the historical torque sequence is specifically: , in, Represents the similarity between the current torque sequence and the i-th historical torque sequence; represents the first distance between the current torque sequence and the i-th historical torque sequence; represents the second distance between the current torque sequence and the i-th historical torque sequence; and denote the first weight and the second weight respectively; Represents an exponential function with a natural constant as its base.

[0013] Preferably, updating the initial detection interval of the image detection according to the similarity comprises: Set the similarity threshold, obtain the minimum similarity between the current torque sequence and each historical torque sequence, and when the minimum similarity is greater than or equal to the similarity threshold, update the initial detection interval of the image detection. The updated image detection interval is Detect once every minute. is the minimum value of similarity, and the value of x is the preset value.

[0014] The embodiments of the present invention have at least the following beneficial effects: the present invention obtains the image data of the concrete strip during the concrete printing process and performs preprocessing to obtain the image to be detected, thereby improving the quality of the image and thus improving the accuracy of subsequent printing quality detection; further, if the printing is abnormal, an early warning is issued and a comprehensive error function is obtained based on the crack degree parameter and the width of the concrete strip; the parameters of the 3D printing device are adjusted using the comprehensive error function, and the parameters of the printing device can be adjusted in real time, thereby improving the subsequent printing quality; finally, the historical torque value when printing is abnormal is used to form a historical torque sequence, and the similarity between the current torque sequence and the historical torque sequence is calculated, and the initial detection interval of the image detection is updated according to the similarity, and the detection interval of the image detection can be continuously updated, thereby improving the real-time performance of subsequent detection, and providing construction personnel with time to modify the parameters of the printing device. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0016] Figure 1 A method flow chart of a 3D printing anomaly detection and early warning method based on machine vision provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0017] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the specific implementation method, structure, features and effects of a 3D printing anomaly detection and early warning method based on machine vision proposed by the present invention in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures, or characteristics in one or more embodiments may be combined in any suitable form.

[0018] Unless defined otherwise, 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 belongs.

[0019] The following is a detailed description of a specific scheme of a 3D printing anomaly detection and early warning method based on machine vision provided by the present invention in conjunction with the accompanying drawings.

[0020] Example: The main application scenarios of the present invention are: the existing 3D printing anomaly detection methods based on machine vision have certain limitations. The singleness of the detection mode may lead to reduced accuracy and real-time performance of the detection, which in turn affects the subsequent timely adjustment of the printing machine, thereby affecting the printing quality. And due to the lack of real-time, accurate monitoring and automatic adjustment mechanisms, quality fluctuations in the printing process are difficult to be effectively controlled in a timely manner, which directly affects the structural performance and durability of the final component.

[0021] See also Figure 1 , which shows a flow chart of a 3D printing anomaly detection and early warning method based on machine vision provided by an embodiment of the present invention, the method comprising the following steps: Step S1, setting an initial detection interval for image detection; acquiring image data of concrete strips during concrete printing and performing preprocessing to obtain an image to be detected.

[0022] First, it is necessary to set the initial detection interval of image detection. Preferably, in the present invention, the initial detection interval is once every m seconds. It should be noted that a complete detection is from the beginning of image acquisition to the acquisition of the detection result. Preferably, in the present invention, m is taken as 20, that is, a machine vision detection is performed every 20 seconds to obtain the detection result. It should be noted that during the concrete 3D printing process, the movement of the nozzle is generally linear or circular. In the present invention, linear or circular movement is used as an example for explanation. The image captured by the camera is a section of concrete strip during the concrete 3D printing process. The time when the image detection starts is the same as the time when the printing starts.

[0023] During the printing process of 3D printing concrete technology, factors such as the viscosity of the material, the movement speed of the print nozzle, etc. will cause defects in the width of the printed concrete strip and the surface of the concrete strip. The strip will become wider or narrower, and cracks will appear on the surface.

[0024] An industrial high-speed camera is installed at an appropriate position near the 3D printing nozzle to ensure that the image data of the newly printed concrete strip can be captured clearly and in real time; further, the image data of the concrete strip is preprocessed by image enhancement and noise reduction algorithms, specifically, the image data of the concrete strip is denoised by using noise reduction algorithms such as mean filtering, median filtering, and Gaussian filtering to obtain noise-reduced image data, and further, the noise-reduced image data is enhanced, and in a specific embodiment of the present invention, the noise-reduced image is processed by using histogram equalization to obtain the image to be detected. In this way, the image to be detected at the current moment can be obtained, and subsequent analysis can be performed.

[0025] It should be noted that the present invention adopts a bilateral filtering method to reduce noise on the image data while maintaining edge information, so that the subsequent detection of the width is more accurate.

[0026] The calculation formula for histogram equalization is: , Among them, g represents the gray value after image enhancement, L represents the gray level, f represents the gray value of the original image data of the concrete strip, and CDF represents the cumulative distribution function. Thus, the image to be detected can be obtained.

[0027] Step S2, processing the image to be detected to obtain the width and crack area of ​​the concrete strip respectively; obtaining the crack degree parameter according to the crack area; and judging whether the printing is abnormal according to the width and crack area of ​​the concrete strip.

[0028] In step S1, the image to be detected of the concrete strip is obtained. Further, the image to be detected is processed using the Canny edge detection algorithm to extract the edge of the concrete strip, and then the width of the concrete strip is calculated by distance transformation using the extracted edge. The method for obtaining the width of the concrete strip is different depending on the shape of the strip, but usually the width of the strip is a rectangular or annular strip. The width of the concrete strip can be obtained by data calculation, and the width of the concrete strip includes the maximum and minimum values ​​of the width. Thus, the maximum and minimum values ​​of the width of the concrete strip at the current moment can be obtained.

[0029] Furthermore, it is necessary to obtain the crack defects on the concrete strip at the current moment, use morphological operations and connected domain analysis to analyze the image to be detected, obtain the crack area, and then obtain the total area of ​​the crack area, which is recorded as the crack area, where the crack area is the sum of the areas of each crack area, that is, the sum of the number of pixels in each crack area.

[0030] Thus, the maximum value, minimum value and crack area of ​​the width of the concrete strip can be obtained. Furthermore, whether the printing is abnormal is determined according to the width and crack area of ​​the concrete strip, and a first width threshold, a second width threshold and an area threshold S are set, wherein the first width threshold is D+5mm, the second width threshold is D-5mm, and the value of the area threshold is obtained based on statistical analysis. The implementer can adjust it according to different actual conditions and construction conditions. When the maximum value of the width of the concrete strip is greater than the first width threshold or the minimum value of the width of the concrete strip is less than the second width threshold, the concrete printing is abnormal. When the crack area of ​​the concrete strip is greater than or equal to the area threshold, the concrete printing is abnormal.

[0031] Finally, a crack degree parameter is obtained based on the crack area. Specifically, when the crack area of ​​the concrete strip is greater than or equal to the area threshold, the crack degree parameter is a first preset value. When the crack area of ​​the concrete strip is less than the area threshold, the crack degree parameter is a second preset value. Preferably, in an embodiment of the present invention, the first preset value and the second preset value are 1 and 0, respectively.

[0032] In this way, an image detection can be completed, and the width of the concrete strip and the area of ​​the crack region on the strip at the current moment can be obtained, and then it can be determined whether the 3D printing equipment has any abnormality during the printing process. The concrete equipment can be adjusted based on this in the future to improve the printing quality. It should be noted that in the process of image detection, the CUDA parallel computing framework is used to accelerate the image processing and feature extraction process, and the sliding window technology is used to process only the region of interest (ROI) in the image to improve the efficiency of image detection.

[0033] It should also be noted that if the width of the concrete strip is abnormal, only one of the abnormalities may occur, narrowing or widening. This is because the printing material is stable in a short period of time, and the image taken by the camera is a very short strip in the entire printing process. Therefore, within this strip, the width abnormality can only be widening or narrowing.

[0034] Step S3: If printing is abnormal, an early warning is issued and a comprehensive error function is obtained based on the crack degree parameter and the width of the concrete strip; the parameters of the 3D printing equipment are adjusted using the comprehensive error function.

[0035] If there is an abnormality in printing, an early warning is first issued and reported to the staff. The adjustment of the printing equipment is automatically processed first, and can also be adjusted manually by the staff; for the abnormality of the strip widening, it is necessary to reduce the speed of the stirring motor that stirs the printing material and increase the moving speed of the printing nozzle; for the abnormality of the strip narrowing or cracks, it is necessary to increase the speed of the stirring motor that stirs the printing material and reduce the moving speed of the printing nozzle. Further, a comprehensive error function is constructed based on the crack degree parameter obtained in step S2 and the width of the concrete strip. The comprehensive error function is specifically: , in, represents the comprehensive error function, and can also represent the comprehensive printing error at time t; t represents time t; represents the crack area at time t; Indicates the maximum or minimum value of the width of the concrete strip at time t. When the width of the strip becomes wider, represents the maximum value of the width of the concrete strip at time t. When the width of the strip becomes narrower, represents the minimum value of the width of the concrete strip at time t; Indicates the target width of the concrete strip; represents the crack error weight coefficient, represents a width error weight coefficient. Preferably, in the embodiment of the present invention, and The values ​​of are 1 and 0.5 respectively.

[0036] Furthermore, the comprehensive error function is used to adjust the parameters of the 3D printing device. Specifically, the comprehensive error function is used to construct a parameter adjustment equation group, and the parameter adjustment equation group is used to adjust the parameters of the 3D printing device, wherein the parameter adjustment equation group includes a print head movement speed adjustment function and a stirring motor stirring speed adjustment function; wherein the parameter adjustment equation group is: , in, and They represent the print head moving speed adjustment function and the stirring motor stirring speed adjustment function respectively, and can also represent the print head moving speed and the stirring motor stirring speed adjusted at time t; Indicates the base print head moving speed, in mm / s; Indicates the stirring speed of the base stirring motor, in r / s; and They represent the print head moving speed adjustment coefficient and the stirring speed adjustment coefficient respectively. Preferably, in the embodiment of the present invention, and The values ​​of are 5 and 0.05 respectively, and implementers can adjust them according to actual conditions. It represents the comprehensive error function, and can also represent the comprehensive printing error at time t.

[0037] Finally, when printing abnormalities occur, the adjusted print head movement speed and stirring motor stirring speed are calculated, and then the parameters of the 3D printing equipment are adjusted to make subsequent printing meet the requirements.

[0038] Step S4, collecting the torque values ​​of the current preset time period to form a current torque sequence; using the historical torque values ​​when printing abnormalities to form a historical torque sequence; calculating the similarity between the current torque sequence and the historical torque sequence.

[0039] During the inspection, the torque changes fed back by the motor of the printing equipment can be used to predict the upcoming printing problems of concrete 3D printing. By establishing a rich database containing historical event data, specific torque patterns related to past problems can be identified. When a similar pattern is detected, it triggers an early warning mechanism, prompting the image detection to shorten its detection interval. New events are constantly recorded and fed back to the historical database, continuously improving the accuracy of the prediction model. It can predict potential anomalies before the problem actually appears, providing operators with time to intervene.

[0040] Specifically, the torque value of a period of time before time t is collected, or more precisely, the torque value from time tn to time t is collected, where t is the current time, time tn is the time before time t, and time tn to time t is recorded as the current preset time period. The torque values ​​of the current preset time period constitute the current torque sequence; the historical torque values ​​when printing anomalies constitute the historical torque sequence. It should be noted that the length of the historical torque sequence may be the same as or different from the length of the current torque sequence. This is because the time when the anomalies occurred in history may be the same or different, and there is more than one historical torque sequence.

[0041] Furthermore, the similarity between the current torque sequence and each historical torque sequence is calculated. Specifically, the dynamic time warping algorithm (DTW) is used to calculate the distance between the current torque sequence and the historical torque sequence, which is recorded as the first distance; the mean value, standard deviation, skewness and kurtosis of the current torque sequence are obtained to form the current feature vector; the mean value, standard deviation, skewness and kurtosis of the historical torque sequence are obtained to form the historical feature vector; the distance between the current feature vector and the historical feature vector is calculated, which is recorded as the second distance; the similarity between the current torque sequence and the historical torque sequence is obtained according to the first distance and the second distance. The specific calculation formula is: , in, Represents the similarity between the current torque sequence and the i-th historical torque sequence; represents the first distance between the current torque sequence and the i-th historical torque sequence; represents the second distance between the current torque sequence and the i-th historical torque sequence; and denote the first weight and the second weight respectively; Represents an exponential function with a natural constant as the base. The values ​​of the first weight and the second weight can be determined by the actual situation of the implementer. Preferably, in the embodiment of the present invention, the values ​​are 0.7 and 0.3, from which it can be obtained that after obtaining the similarity between the current torque sequence and each historical torque sequence, the interval of image detection can be updated later.

[0042] Step S5, updating the initial detection interval of the image detection according to the similarity; and performing subsequent detection according to the updated initial detection interval.

[0043] Set the similarity threshold, obtain the minimum similarity between the current torque sequence and each historical torque sequence, and when the minimum similarity is greater than or equal to the similarity threshold, update the initial detection interval of the image detection. The updated image detection interval is Detect once per second. is the minimum value of similarity, and the value of x is a preset value. Preferably, in the embodiment of the present invention, the preset value is 5 seconds. It should be noted that the values ​​of m and x can be set by the implementer according to the actual printing situation.

[0044] After obtaining the updated initial detection interval, the updated initial detection interval is used to perform image detection on the concrete print; it should be noted that the detection interval is continuously updated, and the condition for stopping the update is that the detection interval is less than 5 seconds.

[0045] In summary, the present invention utilizes image detection to detect the quality of concrete printing in real time, and analyzes the current torque value and the abnormal torque value in the history, and can continuously update the detection interval of image detection, thereby improving the real-time performance of subsequent detections and providing construction personnel with time to modify the parameters of the printing equipment.

[0046] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The above is a description of a specific embodiment of this specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0047] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0048] 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 principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A 3D printing anomaly detection and early warning method based on machine vision, characterized in that: The method includes: Setting the initial detection interval of image detection; obtaining image data of concrete strips during concrete printing and preprocessing to obtain an image to be detected; The image to be detected is processed to obtain the width of the concrete strip and the crack area respectively; the crack degree parameter is obtained according to the crack area; and whether the printing is abnormal is determined according to the width of the concrete strip and the crack area; If printing is abnormal, an early warning is issued and a comprehensive error function is obtained based on the crack degree parameter and the width of the concrete strip; the parameters of the 3D printing equipment are adjusted using the comprehensive error function; Collect the torque values ​​of the current preset period to form a current torque sequence; use the historical torque values ​​when printing abnormalities to form a historical torque sequence; calculate the similarity between the current torque sequence and the historical torque sequence; The initial detection interval of the image detection is updated according to the similarity; and subsequent detection is performed according to the updated initial detection interval.

2. The method for detecting and warning 3D printing anomalies based on machine vision according to claim 1, characterized in that: The method of obtaining image data of concrete strips during concrete printing and preprocessing to obtain an image to be detected includes: Bilateral filtering is used to reduce the noise of image data to obtain a denoised image; the denoised image is processed using histogram equalization to obtain an image to be detected.

3. The 3D printing anomaly detection and early warning method based on machine vision according to claim 1 is characterized in that: The processing of the image to be detected to obtain the width of the concrete strip and the crack area respectively includes: The image to be detected is processed using the Canny edge detection algorithm to extract the edge of the concrete strip; the width of the concrete strip is calculated according to the edge of the concrete strip, wherein the width of the concrete strip includes the maximum value and the minimum value of the width; The image to be detected is analyzed by morphological operation and connected domain analysis to obtain the crack area; the sum of the areas of all crack regions is the crack area.

4. The method for detecting and warning 3D printing anomalies based on machine vision according to claim 1, characterized in that: The step of judging whether printing is abnormal based on the width of the concrete strip and the crack area includes: A first width threshold, a second width threshold and an area threshold are set; when the maximum value of the width of the concrete strip is greater than the first width threshold or the minimum value of the width of the concrete strip is less than the second width threshold, the concrete printing is abnormal; when the crack area of ​​the concrete strip is greater than or equal to the area threshold, the concrete printing is abnormal.

5. The 3D printing anomaly detection and early warning method based on machine vision according to claim 1, characterized in that: The comprehensive error function is specifically: , in, represents the comprehensive error function; t represents the time t; represents the crack area at time t; Indicates the maximum or minimum value of the width of the concrete strip at time t; Indicates the target width of the concrete strip; represents the crack error weight coefficient, Represents the width error weight coefficient.

6. The 3D printing anomaly detection and early warning method based on machine vision according to claim 1, characterized in that: The method of adjusting the parameters of the 3D printing device by using the comprehensive error function includes: A parameter adjustment equation group is constructed using the comprehensive error function, and the parameters of the 3D printing device are adjusted using the parameter adjustment equation group, wherein the parameter adjustment equation group includes a print head movement speed adjustment function and a stirring motor stirring speed adjustment function.

7. The method for detecting and warning 3D printing anomalies based on machine vision according to claim 6, characterized in that: The parameter adjustment equation group is specifically: , in, and They represent the print head moving speed adjustment function and the stirring motor stirring speed adjustment function respectively; Indicates the base print head moving speed; Indicates the stirring speed of the base stirring motor; and They represent the print head moving speed adjustment coefficient and the stirring speed adjustment coefficient respectively; represents the comprehensive error function.

8. The method for detecting and warning 3D printing anomalies based on machine vision according to claim 1, characterized in that: The calculating the similarity between the current torque sequence and the historical torque sequence includes: The dynamic time warping algorithm is used to calculate the distance between the current torque sequence and the historical torque sequence, which is recorded as the first distance; the mean value, standard deviation, skewness and kurtosis of the current torque sequence are obtained to form the current feature vector; the mean value, standard deviation, skewness and kurtosis of the historical torque sequence are obtained to form the historical feature vector; the distance between the current feature vector and the historical feature vector is calculated, which is recorded as the second distance; the similarity between the current torque sequence and the historical torque sequence is obtained based on the first distance and the second distance.

9. The method for detecting and warning 3D printing anomalies based on machine vision according to claim 8, characterized in that: The similarity between the current torque sequence and the historical torque sequence is specifically: , in, Represents the similarity between the current torque sequence and the i-th historical torque sequence; represents the first distance between the current torque sequence and the i-th historical torque sequence; represents the second distance between the current torque sequence and the i-th historical torque sequence; and denote the first weight and the second weight respectively; Represents an exponential function with a natural constant as its base.

10. The method for detecting and warning 3D printing anomalies based on machine vision according to claim 1, characterized in that: The updating of the initial detection interval of the image detection according to the similarity comprises: Set the similarity threshold, obtain the minimum similarity between the current torque sequence and each historical torque sequence, and when the minimum similarity is greater than or equal to the similarity threshold, update the initial detection interval of the image detection. The updated image detection interval is Detect once every minute. is the minimum value of similarity, and the value of x is the preset value.

Citation Information

Patent Citations

  • 3D printing concrete column surface crack detection method based on point cloud segmentation registration

    CN117115140A

  • Real-time quality detection, regulation and control method and device for 3D printing concrete

    CN118469926A

  • Concrete structure crack inspection device and crack inspection method

    WO2006049174A1

  • Systems and methods for 3D printing

    WO2022174322A1

Cited By

  • Printing model flaw detection method and system in 3D printing process

    CN119477915A