Method and system for real-time intelligent monitoring of quality of long-span bridge tower fair-faced concrete

Abstract

A long bridge tower clear water concrete quality real-time intelligent monitoring method and system, comprising: real-time acquisition of concrete resistivity, temperature and pressure, and concrete image information in the concrete pouring process; according to the concrete image information, real-time calculation of the maximum diameter, minimum diameter, number and pixel area of the bubbles in the concrete, and prediction of the resistivity, temperature and pressure of the concrete after molding according to the real-time acquisition of the concrete resistivity, temperature and pressure; in the current control cycle, comparison of the deviation between the predicted values of the resistivity, temperature and pressure of the concrete after molding and the expected resistivity, temperature and pressure, and the real-time calculated maximum diameter, minimum diameter, number and pixel area of the bubbles and the expected maximum diameter, minimum diameter, number and pixel area of the bubbles; according to the deviation, control of the vibrator to compact the concrete until the deviation meets the set requirements.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of civil engineering and information technology, and relates to an intelligent construction method and an artificial intelligence method, in particular to a long bridge tower fair-faced concrete quality real-time intelligent monitoring method and system. BACKGROUND

[0002] The bridge tower is the load-bearing unit of the bridge and the basis of the safety of the whole bridge. The bridge tower is often made of fair-faced concrete pouring, and the process is to pour it on the concrete structure formwork at one time, and after the formwork is removed, no additional decorative layer is needed, and the surface of the concrete is directly used as the exposed surface of the bridge. The construction technology of fair-faced concrete has high requirements, and it is beneficial to ensure the construction quality of fair-faced concrete to implement comprehensive operation during construction. Therefore, real-time intelligent monitoring of the quality of the bridge tower fair-faced concrete is an important cornerstone for improving the quality of the formed bridge tower, ensuring the safety of the bridge and personnel.

[0003] However, a prominent problem at present is that the construction quality of fair-faced concrete at the present stage is mainly determined by manual visual inspection combined with construction experience. This extensive manual method obviously has many disadvantages, not only is it unstable and low in accuracy, but also increases the labor intensity of construction personnel and technical personnel, and the risk coefficient in the special operation scene of bridge tower pouring, and therefore does not conform to the development trend of standardization, intelligentization and refinement of the quality of the bridge tower fair-faced concrete under the background of intelligent construction.

[0004] The patent document "CN213749147U" provides a drilling sampling device for quality detection of bridge and bridge, which cuts the concrete column of the bridge deck from the bridge body for detection, overcoming the problem of instability of traditional drilling device and high replacement cost after damage. The patent document "CN213705781U" provides a bridge pier concrete quality unmanned aerial vehicle geological radar detection device, which uses an unmanned aerial vehicle to carry a camera and a geological radar antenna to detect the quality of the bridge pier concrete, and ground personnel distinguish the geological radar pattern to provide scientific and accurate judgment for the quality of the bridge pier concrete, replacing the existing manual detection, improving the efficiency and accuracy of the bridge pier concrete quality detection, improving the safety factor of the bridge detection, and reducing the cost of the bridge detection. The patent document "CN113326550A" discloses a real-time detection method for the vibration quality of a concrete precast bridge, which constructs a vibration time course random power spectrum model and a model parameter reasonable interval range of a large number of qualified concrete precast bridge model parameters to guide the pouring and vibration of the to-be-tested concrete precast bridge. The concrete quality detection devices provided in the patent documents "CN213749147U" and "CN213705781U" are all post-detection devices after the concrete is formed, and cannot detect the quality of the concrete in real time, and lack corresponding control measures. In the patent document "CN113326550A", a vibration process random power spectrum model needs to be constructed, and the construction personnel and technical personnel need to have corresponding theoretical knowledge to use the technology skillfully, which may limit the application of the invention. Therefore, a new real-time intelligent monitoring method for bridge tower fair-faced concrete quality needs to be invented. SUMMARY

[0005] The present application provides a long bridge tower fair-faced concrete quality real-time intelligent monitoring method and system, which integrates machine vision, machine learning and other artificial intelligence technologies with the bridge tower construction process to standardize, intelligentize and refine the construction of bridge tower fair-faced concrete, and promotes the upgrading and iteration of building equipment and technology.

[0006] In the first aspect, a long bridge tower fair-faced concrete quality real-time intelligent monitoring method is provided, which comprises: collecting the resistivity ρ(t), temperature T(t) and pressure P(t) of the fair-faced concrete, and the image of the fair-faced concrete in real time during the pouring process of the long bridge tower; calculating the maximum diameter d max (t), minimum diameter d min (t), number N sum (t) and pixel area S pixel (t) of the bubbles in the fair-faced concrete according to the image of the fair-faced concrete, and predicting the resistivity ρ set , temperature Tset and pressure P set ; Compare the predicted resistivity ρ of the fair-faced concrete after molding during the current control cycle. set Temperature T set and pressure P set With expected resistivity Desired temperature and expected pressure and the maximum diameter d of air bubbles in the aforementioned fair-faced concrete. max (t), minimum bubble diameter d min (t), number of bubbles N sum (t) and bubble pixel area S pixel (t) and the desired maximum bubble diameter Minimum diameter of desired bubble Expected number of bubbles and expected bubble pixel area The deviation between the two is controlled; based on the deviation, the intelligent vibrator is used to compact the fair-faced concrete until the deviation reaches the set requirement.

[0007] Secondly, a real-time intelligent monitoring system for the quality of fair-faced concrete in the towers of long bridges is provided, including:

[0008] The physical property data acquisition subsystem collects the resistivity ρ(t), temperature T(t), and pressure P(t) of fair-faced concrete in real time during the pouring of the bridge towers of long bridges.

[0009] The visual information acquisition and processing subsystem acquires real-time images of fair-faced concrete during the pouring of bridge towers for long bridges, and calculates the maximum diameter d of air bubbles in the fair-faced concrete based on these images. max (t), minimum bubble diameter d min (t), Number of bubbles N sum (t) and bubble pixel area S pixel (t), and based on the resistivity ρ(t), temperature T(t), and pressure P(t) of the fair-faced concrete, predict the resistivity ρ after the fair-faced concrete is formed. set Temperature T set and pressure P set In the current control cycle, compare the predicted resistivity ρ of the fair-faced concrete after molding. set Temperature T set and pressure P set With expected resistivity Desired temperature and expected pressure and the maximum diameter d of air bubbles in the aforementioned fair-faced concrete. max (t), minimum bubble diameter d min (t), Number of bubbles Nsum (t) and bubble pixel area S pixel (t) and expected bubble maximum diameter expected bubble minimum diameter expected bubble number and expected bubble pixel area deviation between; and

[0010] Intelligent vibrator, which tamps the fair-faced concrete according to the deviation until the deviation reaches the set requirement.

[0011] In the above-mentioned first aspect and / or second aspect, an indicator light is arranged around the pouring formwork, and the indicator light is lit when the deviation deviates from the set requirement.

[0012] In the above-mentioned first aspect and / or second aspect, the bubble detection method for the fair-faced concrete comprises: performing gray-scale processing on the fair-faced concrete image to obtain a gray-scale image; performing inflation and then erosion on the gray-scale image to obtain a morphological operation result image; subtracting the gray-scale image from the morphological operation result image and performing an inversion operation to obtain a gray-scale image after water stains are removed; using a Gaussian filter to eliminate noise interference on the gray-scale image after water stains are removed and performing binaryzation processing; performing inflation operation on the image after binaryzation processing, expanding the image background and compressing the bubble part; subtracting the image after inflation from the image before inflation to locate the position of the bubble; applying a contour finding method to mark all bubbles in the image and analyzing and calculating the number and pixel area of the bubbles; and outputting the fair-faced concrete image with bubble position marking, and the number and pixel area of the bubbles.

[0013] The long bridge tower fair-faced concrete quality real-time intelligent monitoring method and system provided by the application establish a long bridge tower fair-faced concrete quality real-time intelligent monitoring closed-loop feedback control model, compare two groups of parameters related to the quality and appearance of the fair-faced concrete during the concrete pouring process, send a vibrating or shutdown instruction to the vibrator, realize real-time closed-loop control of the quality and appearance of the fair-faced concrete of the bridge tower, and standardize, intelligentize and refine the construction of the fair-faced concrete of the bridge tower, thereby effectively improving the quality of the formed bridge tower and ensuring the safety of the bridge and personnel. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the technical solutions of the embodiments of the application, the drawings of the embodiments will be briefly introduced below.

[0015] Figure 1 The long bridge tower fair-faced concrete quality real-time intelligent monitoring system provided by an embodiment of the application is shown in the figure.

[0016] Figure 2A long bridge tower fair-faced concrete quality real-time intelligent monitoring method schematic diagram is provided for an embodiment of the present application.

[0017] Figure 3 A concrete test block pre-pouring image information schematic diagram collected by a visual information acquisition and processing subsystem is provided for an embodiment of the present application.

[0018] Figure 4 A fair-faced concrete surface bubble detection and analysis digital image processing method flow chart is provided for an embodiment of the present application.

[0019] Figure 5 A bubble pixel area schematic diagram calculated by using a fair-faced concrete surface bubble detection and analysis digital image processing method is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0020] As shown in Figure 1 A long bridge tower fair-faced concrete quality real-time intelligent monitoring system, comprising: a physical property data acquisition subsystem 1, a visual information acquisition and processing subsystem 2, an intelligent vibrator control subsystem 3, and a cloud service subsystem 4.

[0021] The physical property data acquisition subsystem can be realized based on PLC technology, and is used to collect three mechanism data of resistivity, temperature and pressure in the fair-faced concrete pouring process of the bridge tower, and combine the quality and appearance discrimination conclusions of the visual information acquisition and processing subsystem to light up the indicator lights around the pouring template.

[0022] The visual information acquisition and processing subsystem completes the camera movement, electrical control and data communication functions related to the bridge tower fair-faced concrete image acquisition, and provides the functions of fair-faced concrete surface bubble detection and analysis digital image processing and fair-faced concrete physical property prediction deep learning software.

[0023] The intelligent vibrator control subsystem receives the quality and appearance discrimination conclusions of the visual information acquisition and processing subsystem, and according to the discrimination conclusions, controls the high-frequency vibration of the vibrator through the frequency converter to improve the quality and appearance indexes of the bridge tower fair-faced concrete until the specification requirements are met.

[0024] The cloud service subsystem provides a functional interface, so that the quality data and production whole process information of the bridge tower fair-faced concrete project are saved in the cloud, and the mainstream cloud service (such as Ali Cloud) is called to assist the training and replacement of the artificial intelligence model in the visual information acquisition and control subsystem.

[0025] Figure 2 A long bridge tower fair-faced concrete quality real-time intelligent monitoring method is shown. The method will be described in detail below. Figure 2 A long bridge tower fair-faced concrete quality real-time intelligent monitoring method is shown. The method will be described in detail below.

[0026] Step 1: Establish a real-time intelligent monitoring closed-loop feedback control model for the quality of the long bridge tower fair-faced concrete. The closed-loop feedback control model includes a controller, an actuator, a sensor, and a controlled object. The controller is a collection of all software and hardware functions related to the processing of quality and appearance judgment conclusions, calculation, and generation in the visual information acquisition and processing subsystem and the intelligent vibrator control subsystem, and sending control instructions to the vibrator. The actuator is an intelligent vibrator. The sensor is an image recognition and detection unit. The controlled object is the bridge tower fair-faced concrete. The input signal of the closed-loop feedback controller is the expected quality and appearance parameters of the fair-faced concrete, and the output signal is the actual quality and appearance parameters of the fair-faced concrete. The expected quality parameters of the fair-faced concrete include: expected resistivity expected temperature and expected pressure The actual quality parameters are all time-dependent variables, including: resistivity p(t), temperature T(t), and pressure P(t). The expected appearance parameters include: expected maximum bubble diameter expected minimum bubble diameter expected bubble number and expected bubble pixel area The actual appearance parameters include: maximum bubble diameter d max (t), minimum bubble diameter d min (t), bubble number N sum (t), and bubble pixel area S pixel (t).

[0027] Step 2: Pour multiple concrete test blocks in proportion to the long bridge tower to be built, and record the resistivity p blk (t), temperature T blk (t), and pressure P blk (t), as well as the maximum bubble diameter minimum bubble diameter bubble number and bubble pixel area data during the vibration process and after molding, and determine the value range of each parameter that meets the quality requirements of the fair-faced concrete based on the specific data after molding.

[0028] Step 3: During the pouring process of the long bridge tower, real-time collection of resistivity p(t), temperature T(t), and pressure P(t), as well as fair-faced concrete image information, calling the fair-faced concrete surface bubble detection and analysis digital image processing module of the visual information acquisition and processing subsystem, and the fair-faced concrete physical property prediction deep learning model, real-time calculation of the maximum bubble diameter d max (t), minimum bubble diameter d min (t), bubble number N sum(t) and bubble pixel area S pixel (t) data, and predict the resistivity ρ of fair-faced concrete after molding. set Temperature T set and pressure P set .

[0029] Step 4: Compare the predicted value ρ after the fair-faced concrete has been formed within the current control cycle. set T set P set Compared with expected value And the real-time calculated d max (t), d min (t), N sum (t), S pixel (t) and expected value The deviation between the two is determined, and it is judged whether the deviation is within the allowable range (determined by step 2, which may have different allowable ranges depending on the process scenario). If it is, proceed to step 6; otherwise, continue to the next step.

[0030] Step 5: Send a continuous vibration command to the vibrator through the intelligent vibrator control subsystem, and then proceed to step 3.

[0031] Step 6: Once the quality and appearance of the fair-faced concrete have met the set requirements, send a stop command to the vibrator to complete the process.

[0032] Figure 3 This is a schematic diagram showing images of concrete test blocks collected by the visual information acquisition and processing subsystem before pouring, during the application of this invention on a large bridge engineering project in China. In the image, the reinforcing steel frame of the test block has been tied, and the exterior is covered by a transparent template and timber. Through the transparent template, an industrial camera in the visual information acquisition and processing subsystem can capture the appearance of the fair-faced concrete.

[0033] Figure 4 A digital image processing method for detecting and analyzing air bubbles on the surface of fair-faced concrete in bridge towers is shown, including the following steps:

[0034] Step 1: Convert the original image of the bridge tower's exposed concrete to grayscale.

[0035] Step 2: Perform an expansion and then erosion operation on the grayscale image of the bridge tower's fair-faced concrete after grayscale processing to obtain the resulting image after morphological operation.

[0036] Step 3: Subtract the grayscale image of the bridge tower's exposed concrete from the resulting image after morphological operations, and then invert the image to obtain the grayscale image after removing water stains.

[0037] Step 4: Gaussian filter is used to remove noise interference for the gray image after water stain removal, and binarization processing is performed.

[0038] Step 5: an expansion operation is performed on the result image after binarization processing in step 4, the image background is expanded, and the bubble part is compressed.

[0039] Step 6: the image before expansion in step 5 is subtracted from the image after expansion, and the position of the bubble is accurately located.

[0040] Step 7: a contour finding method is applied to mark all the bubbles in the result image in step 6, and the number and pixel area of the bubbles are calculated and analyzed.

[0041] Step 8: the bridge tower fair-faced concrete image with bubble position marking, and the number and pixel area of the bubbles are output.

[0042] Figure 5 The leftmost column is the local picture of the appearance of the fair-faced concrete which is segmented from the original image, the second column is the detected bubble, the third and fourth columns are respectively marked with different colors for the detected bubble, and the rightmost column is the calculated pixel area of the bubble, which is respectively 120.5, 203 and 709 pixels 2 .

Claims

1. A method for real-time intelligent monitoring of the quality of fair-faced concrete in the towers of long bridges, characterized in that, include: During the pouring of the bridge towers of the long bridge, the resistivity ρ(t), temperature T(t), and pressure P(t) of the fair-faced concrete, as well as images of the fair-faced concrete, were collected in real time. Calculate the maximum diameter d of air bubbles in the fair-faced concrete based on the image of the fair-faced concrete. max (t), minimum bubble diameter d min (t), Number of bubbles N sum (t) and bubble pixel area S pixel (t), and based on the resistivity ρ(t), temperature T(t), and pressure P(t) of the fair-faced concrete, predict the resistivity ρ after the fair-faced concrete is formed. set Temperature T set and pressure P set ; The predicted resistivity ρ of the fair-faced concrete after molding is compared during the current control cycle. set Temperature T set and pressure P set With expected resistivity Desired temperature and expected pressure and the maximum diameter d of air bubbles in the aforementioned fair-faced concrete. max (t), minimum bubble diameter d min (t), Number of bubbles N sum (t) and bubble pixel area S pixel (t) and the desired maximum bubble diameter Minimum diameter of desired bubble Expected number of bubbles and expected bubble pixel area Deviation between; The intelligent vibrator is controlled according to the deviation to compact the fair-faced concrete until the deviation reaches the set requirement.

2. The method for real-time intelligent monitoring of the quality of fair-faced concrete in the towers of long bridges according to claim 1, characterized in that, Methods for detecting air bubbles in fair-faced concrete include: The image of the fair-faced concrete is converted to grayscale to obtain a grayscale image; The grayscale image is first dilated and then eroded to obtain the morphological operation result image. Subtract the grayscale image from the image resulting from the morphological operation, and then invert the image to obtain the grayscale image after removing the water stains. The grayscale image after removing water stains is subjected to Gaussian filtering to eliminate noise interference and then binarized. The image after binarization is dilated to expand the background and compress the bubble portion. The position of the bubble is determined by subtracting the expanded image from the image before expansion. The contour lookup method is applied to mark all bubbles in the image, and the number and pixel area of ​​the bubbles are calculated by parsing. Output an image of fair-faced concrete with bubble location markers, as well as the number and pixel area of ​​the bubbles.

3. The method for real-time intelligent monitoring of the quality of fair-faced concrete in the towers of long bridges according to claim 1, characterized in that, Indicator lights are arranged around the casting template, and the indicator lights will illuminate when the deviation deviates from the set requirements.

4. A real-time intelligent monitoring system for the quality of fair-faced concrete in the towers of long bridges, characterized in that, The physical property data acquisition subsystem collects the resistivity ρ(t), temperature T(t), and pressure P(t) of fair-faced concrete in real time during the pouring of the bridge towers of long bridges. The visual information acquisition and processing subsystem acquires real-time images of fair-faced concrete during the pouring of bridge towers for long bridges, and calculates the maximum diameter d of air bubbles in the fair-faced concrete based on these images. max (t), minimum bubble diameter d min (t), Number of bubbles N sum (t) and bubble pixel area S pixel (t), and based on the resistivity ρ(t), temperature T(t), and pressure P(t) of the fair-faced concrete, predict the resistivity ρ after the fair-faced concrete is formed. set Temperature T set and pressure P set In the current control cycle, compare the predicted resistivity ρ of the fair-faced concrete after molding. set Temperature T set and pressure P set With expected resistivity Desired temperature and expected pressure and the maximum diameter d of air bubbles in the aforementioned fair-faced concrete. max (t), minimum bubble diameter d min (t), Number of bubbles N sum (t) and bubble pixel area S pixel (t) and the desired maximum bubble diameter Minimum diameter of desired bubble Expected number of bubbles and expected bubble pixel area The deviation between them; and A smart vibrator compacts fair-faced concrete according to the deviation until the deviation reaches the set requirement.

5. The real-time intelligent monitoring system for the quality of fair-faced concrete in the towers of long bridges according to claim 4, characterized in that, Methods for detecting air bubbles in fair-faced concrete include: The image of the fair-faced concrete is converted to grayscale to obtain a grayscale image; The grayscale image is first dilated and then eroded to obtain the morphological operation result image. Subtract the grayscale image from the image resulting from the morphological operation, and then invert the image to obtain the grayscale image after removing the water stains. The grayscale image after removing water stains is subjected to Gaussian filtering to eliminate noise interference and then binarized. The image after binarization is dilated to expand the background and compress the bubble portion. The location of the bubble is determined by subtracting the dilated image from the binarized image before dilation. The contour lookup method is applied to mark all bubbles in the image, and the number and pixel area of ​​the bubbles are calculated by parsing. Output an image of fair-faced concrete with bubble location markers, as well as the number and pixel area of ​​the bubbles.

6. The real-time intelligent monitoring system for the quality of fair-faced concrete in the towers of long bridges according to claim 4, characterized in that, Indicator lights are arranged around the casting template, and the indicator lights will illuminate when the deviation deviates from the set requirements.

Images