A cap beam intelligent welding cloud control system

Abstract

The present invention relates to the field of industrial intelligent control technology, and in particular to a cap beam intelligent welding cloud control system. The pre-welding proofreading module includes a cap beam reinforcement frame unit and a suspicious reinforcement judgment unit. The former verifies the number of reinforcements based on the comparison between the cap beam frame diagram and the actual image data to ensure the correct positioning of the reinforcements before welding; the latter verifies the spacing between reinforcements, determines reinforcements with unreasonable layout and generates alarm information. Based on a convolutional neural network, the position of the reinforcements to be welded is determined from the image data and a two-dimensional matrix is ​​generated to ensure that the welding sequence and position are reasonable and safe. According to the verification results, the welding robot is adjusted in real time to perform precise welding and continuously update the welding data. The welding head pressure, welding head depth and sound parameters are monitored in real time, and a stop command is issued when an abnormality is detected to avoid dislocation or intrusion of debris. The present invention significantly improves welding efficiency and quality, reduces human errors and risks, reduces production costs and construction cycles, and realizes the intelligence and automation of the cap beam welding process.

Description

Technical Field

[0001] The present invention relates to the field of industrial intelligent control technology, and in particular to a cap beam intelligent welding cloud control system. Background Art

[0002] The cap beam is a critical component of bridges and other large structures that bears the loads of the superstructure. It is typically located atop piers or abutments, and its primary function is to transfer the bridge deck or bridge trusses to the substructure. The quality of the cap beam directly impacts the stability and safety of the entire bridge. The cap beam skeleton is a fundamental component of the cap beam. It consists of steel bars welded together to form a framework capable of bearing and transmitting loads. This framework requires precise positioning and connection during construction to ensure structural integrity and uniform load distribution. Conventional welding of the cap beam skeleton relies primarily on manual labor. The specific steps involve workers manually measuring and arranging the steel bars according to design drawings, followed by welding using welding equipment. While this method has met construction requirements to a certain extent, it has several significant drawbacks: the manual welding process is complex, time-consuming, and labor-intensive, extending the construction period. Manual measurement and placement of steel bars are subject to human error, making it difficult to accurately position and weld the steel bars. This can easily lead to problems such as misaligned steel bars and uneven spacing, impacting the overall quality of the cap beam.

[0003] In the prior art, steel bars are also fixed by welding robots and welded by welding robots that can move along the XY axis. Most existing welding robots rely on preset welding paths and programs and lack adaptive capabilities. As a result, welding work will still be carried out when steel bar deviations occur or the steel bars are not ready. The robots are highly dependent on repeated manual inspections and supervision, and are unable to stop automatically according to real-time conditions, resulting in a high degree of dependence on personnel and reduced production efficiency. Summary of the Invention

[0004] In order to solve the above problems, the present invention provides a cap beam intelligent welding cloud control system.

[0005] To achieve the above object, the technical solution adopted by the present invention is:

[0006] A cap beam intelligent welding cloud control system includes: an image receiving module, a pre-welding proofreading module, a welding robot cloud control module, and an abnormality monitoring module. The pre-welding proofreading module includes a cap beam reinforcement frame unit and a suspicious reinforcement judgment unit. The image receiving module, the pre-welding proofreading module, and the abnormality monitoring module are respectively connected to the welding robot cloud control module, and the image receiving module is connected to the pre-welding proofreading module.

[0007] The image receiving module is used to receive image data of the steel bar fixing platform;

[0008] The cap beam reinforcement frame unit is used to receive the cap beam frame diagram, compare the cap beam frame diagram with the image data of the reinforcement fixing platform, and verify the number of reinforcement bars;

[0009] The suspicious steel bar determination unit is used to verify the steel bar spacing according to the image data of the steel bar fixing platform, and determine the suspicious steel bars based on the steel bar spacing;

[0010] The welding robot cloud control module determines the steel bar to be welded position from the image data of the steel bar fixing platform based on the convolutional neural network, generates a two-dimensional matrix according to the steel bar to be welded position, updates the parameters of the two-dimensional matrix based on the jump welding logic, controls the welding robot to move and weld based on the verification result of the pre-welding proofreading module and the updated two-dimensional matrix, receives the displacement and welding data of the welding robot in real time and updates the two-dimensional matrix;

[0011] The abnormality monitoring module is used to receive the working parameters of the welding robot during welding, judge the welding abnormality according to the working parameters and send a stop command to the welding robot cloud control module. The working parameters include welding head pressure, welding head depth and sound parameters.

[0012] Furthermore, the verification of the number of steel bars based on the comparison of the cap beam frame diagram with the image data of the steel bar fixing platform includes the following steps:

[0013] Preprocess the cap beam frame diagram and real-time image data, including denoising, grayscale conversion and image enhancement;

[0014] Use image recognition algorithms to extract features from pre-processed images, identify separated steel bars and count the number of steel bars;

[0015] The number of steel bars of the identified steel bar fixing platform is compared with the number in the cap beam frame drawing to generate a deviation value for the number of steel bars.

[0016] Furthermore, the method of verifying the spacing between steel bars according to the image data of the steel bar fixing platform, determining suspicious steel bars based on the spacing between steel bars and generating an alarm includes the following steps:

[0017] Classify and mark the steel bars based on the identified and separated steel bars and steel bar location data;

[0018] Calculate the distance between adjacent bars based on the marking results;

[0019] Compare the distance between adjacent steel bars with the preset standard spacing to identify suspicious steel bars whose spacing does not meet the design requirements.

[0020] Furthermore, the convolutional neural network is trained by the following steps:

[0021] The steel bars identified and separated in the image data of the steel bar fixing platform are calibrated based on the same color value to obtain the image to be welded;

[0022] Annotate the image to be welded, including marking the weld position with another color value to obtain a target image;

[0023] The convolutional neural network is trained with the image to be welded as input and the target image as output.

[0024] Furthermore, the formula of the convolutional neural network is as follows:

[0025] ;

[0026] in, is the output vector, which represents the predicted value of the steel bar to be welded; is the input vector, specifically the image to be welded; and is the row index and column index of the convolution kernel; is the number of channels of the input image; is the weight matrix of the convolutional layer; is the weight parameter of the convolution kernel; is the bias vector of the convolutional layer; is the activation function of the convolutional layer; is the activation function of the fully connected layer; C is the bias vector of the fully connected layer; is the height of the convolution kernel; is the width of the convolution kernel.

[0027] Furthermore, generating a two-dimensional matrix according to the positions of the steel bars to be welded and updating the parameters of the two-dimensional matrix based on the skip welding logic includes the following steps:

[0028] Generate a two-dimensional matrix through the target image output by the convolutional neural network, where each element in the two-dimensional matrix represents the coordinates and status information of a point to be welded;

[0029] The jump welding logic is applied to update the parameters of the two-dimensional matrix, including updating the parameters of the consecutive points to be welded from the parameters to be welded to the parameters of the welded points through a preset step size.

[0030] Furthermore, the control of the welding robot to perform displacement and welding based on the verification result of the pre-welding verification module and the updated two-dimensional matrix includes the following steps:

[0031] Obtain the verification results of the pre-welding proofreading module;

[0032] If the deviation value of the number of steel bars is 0 and no suspicious steel bars are found, the welding robot is controlled to move and weld according to the updated two-dimensional matrix;

[0033] If the deviation value of the number of steel bars is not 0 or suspicious steel bars are found, the control stops the welding robot from moving and welding.

[0034] Furthermore, controlling the welding robot to perform displacement and welding includes the following steps:

[0035] According to the parameters to be welded in the two-dimensional matrix, the welding robot is controlled to move the X-axis and Y-axis to the position of the parameters to be welded and weld through the welding head;

[0036] After each welding point is completed, the location is obtained and the two-dimensional matrix is ​​updated, and the parameters to be welded are updated to the welded parameters.

[0037] Furthermore, judging welding abnormality according to working parameters includes the following steps:

[0038] Obtain the welding head depth, welding head pressure and sound parameters and the corresponding timestamp of a single welding when there is no abnormality in the history;

[0039] Based on the same timestamp, the value ranges of welding head pressure and sound parameters at different welding head depths are obtained;

[0040] If the welding head pressure or sound parameters do not belong to the value range at the current welding head depth during real-time welding, it is judged as welding abnormality;

[0041] If the welding head pressure or sound parameters do not belong to the value range at the current welding head depth during real-time welding work, it is judged that there is no abnormality.

[0042] The present invention has the beneficial effect of utilizing image processing technology to enable the system to obtain real-time information on the layout and quantity of rebar, reducing the need for manual intervention. The pre-weld verification module includes a cap beam rebar frame unit and a suspicious rebar identification unit. The cap beam rebar frame unit compares the cap beam frame diagram with actual image data to verify the number of rebars, ensuring correct rebar positioning before welding. The suspicious rebar identification unit verifies the spacing between rebars to determine whether there are suspicious rebars with unreasonable layouts and generates an alert prompting the operator to inspect and adjust the layout. The pre-weld verification module improves the accuracy of rebar layout through image recognition algorithms and feature extraction techniques. The welding robot cloud control module uses a convolutional neural network to determine the rebar locations to be welded from image data and generates a corresponding two-dimensional matrix. By training on a large amount of welding image data, the convolutional neural network accurately identifies the rebar locations to be welded and generates images of the rebars to be welded. The parameters of the two-dimensional matrix are updated based on skip welding logic to ensure the rationality and safety of the welding sequence and location. Based on the verification results of the pre-weld verification module, the welding robot cloud control module adjusts the welding robot in real time for precise welding and continuously updates welding data to ensure that each weld meets design requirements. The abnormality monitoring module monitors working parameters in real time during the welding process, including welding head pressure, welding head depth, and sound parameters. Once an abnormality is detected, the system will immediately issue a stoppage command to prevent welding defects. By comparing the real-time working parameters with the parameter range of historical normal operation, the module accurately determines whether there is an abnormality in the welding, thereby avoiding misalignment or debris in the welding, and ensuring the stability and reliability of quality. The present invention significantly improves welding efficiency and quality through real-time multi-parameter monitoring, reduces errors and risks of human operation, reduces production costs and construction cycles, and realizes the intelligence and automation of the cap beam welding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 It is a structural diagram of an intelligent welding cloud control system for a cap beam in the present invention.

[0044] Figure 2 This is a flow chart of the training steps of the convolutional neural network used to determine the position of steel bars to be welded in the present invention. DETAILED DESCRIPTION

[0045] See also Figure 1-2 As shown, the present invention relates to a cap beam intelligent welding cloud control system, comprising: an image receiving module, a pre-welding proofreading module, a welding robot cloud control module and an abnormality monitoring module, wherein the pre-welding proofreading module comprises a cap beam reinforcement frame unit and a suspicious reinforcement judgment unit, the image receiving module, the pre-welding proofreading module and the abnormality monitoring module are respectively connected to the welding robot cloud control module, and the image receiving module is connected to the pre-welding proofreading module;

[0046] The image receiving module is used to receive image data of the steel bar fixing platform;

[0047] The cap beam reinforcement frame unit is used to receive the cap beam frame diagram, compare the cap beam frame diagram with the image data of the reinforcement fixing platform, and verify the number of reinforcement bars;

[0048] The suspicious steel bar determination unit is used to verify the steel bar spacing according to the image data of the steel bar fixing platform, and determine the suspicious steel bars based on the steel bar spacing;

[0049] The welding robot cloud control module determines the steel bar to be welded position from the image data of the steel bar fixing platform based on the convolutional neural network, generates a two-dimensional matrix according to the steel bar to be welded position, updates the parameters of the two-dimensional matrix based on the jump welding logic, controls the welding robot to move and weld based on the verification result of the pre-welding proofreading module and the updated two-dimensional matrix, receives the displacement and welding data of the welding robot in real time and updates the two-dimensional matrix;

[0050] The abnormality monitoring module is used to receive the working parameters of the welding robot during welding, judge the welding abnormality according to the working parameters and send a stop command to the welding robot cloud control module. The working parameters include welding head pressure, welding head depth and sound parameters.

[0051] It should be noted that the image receiving module is responsible for receiving image data from the rebar fixing platform. The rebar fixing platform is a work surface used to place and fix cap beam rebar, ensuring its stable position and posture during welding. The image receiving module uses a high-resolution industrial camera and advanced image processing technology to capture and process real-time image data of the rebar layout. This image data includes key information such as the spatial position, number, and spacing of the rebars. The image receiving module also provides pre-processing functions such as denoising, grayscale conversion, and image enhancement to ensure image data clarity and accuracy, providing a reliable foundation for subsequent image analysis. The cap beam rebar frame unit compares the cap beam frame diagram with the real-time image data from the rebar fixing platform. Using an image recognition algorithm, it extracts rebar features from the image, calculates the actual number of rebars, compares it with the rebar number in the design drawing, and generates a rebar quantity deviation report. This step ensures correct rebar placement before welding, avoiding welding quality issues caused by incorrect rebar numbers. The suspicious rebar identification unit verifies rebar spacing based on the image data from the rebar fixing platform. By calculating the distance between adjacent rebars and comparing it to the preset standard spacing, it identifies rebar that is suspected of not meeting design spacing requirements. For any suspected rebar identified, the unit locates its specific position within the image and generates an alert prompting the operator to inspect and adjust it. This step, through precise image analysis, ensures that the spacing between rebars meets design requirements, improving the overall quality of the cap beam. The welding robot cloud control module, the core component of the system, uses a convolutional neural network to determine the rebar locations to be welded from image data from the rebar fixing platform. Trained on a large amount of welding image data, the convolutional neural network accurately identifies the rebar locations to be welded and generates a two-dimensional matrix representing the weld points. Each element in this matrix represents the coordinates and status of a weld point. The system applies skip welding logic to update the parameters of the two-dimensional matrix, optimizing the welding path and sequence to avoid material deformation caused by heat concentration. Based on the updated two-dimensional matrix, the welding robot precisely controls its movement and welding operations, and receives and processes the displacement and welding data generated during the welding process in real time, ensuring the accuracy and quality of each weld point. Specifically, the welding robot system's movement path is designed based on an XY coordinate system, with two parallel tracks for Y-axis movement and a horizontal axis for X-axis movement, covering the entire area above the rebar fixing platform. The details are as follows: For Y-axis movement, the welding robot is mounted on two parallel tracks arranged longitudinally along the rebar fixing platform, enabling the robot to move forward and backward along the Y-axis. The tracks utilize high-precision linear guides and servo motors to ensure stable and accurate Y-axis movement. For X-axis movement, a horizontal axis is installed between the two Y-axis tracks, housing the welding gun and a vertical telescopic mechanism. This axis is driven by a drive mechanism for left and right movement along the X-axis. X-axis movement also utilizes linear guides and a servo motor to ensure accurate X-axis movement.The horizontal axis design enables the welding robot to cover the entire area above the rebar fixing platform, ensuring accurate positioning and welding of each weld point. The welding robot is operated by an integrated control system, which receives data from the image reception module and pre-weld calibration module and adjusts welding operations based on the preset welding path and real-time data. Using a convolutional neural network and jump welding logic, the control system generates and updates a two-dimensional matrix to guide the mechanical telescopic device for precise welding operations. The control system also receives and processes displacement and welding data generated during the welding process in real time, dynamically adjusting welding parameters to improve welding accuracy and consistency. The anomaly monitoring module monitors the welding robot's operating parameters, including welding head pressure, welding head depth, and acoustic parameters, in real time during the welding process. This module determines whether the welding is normal by comparing them with preset standard parameters. If an anomaly is detected, such as welding head pressure or welding temperature outside the normal range, the system immediately issues a stop command to prevent welding defects. The anomaly monitoring module also records abnormal data to provide a basis for subsequent analysis and improvement.

[0052] Furthermore, the comparison between the cap beam frame diagram and the image data of the steel bar fixing platform to verify the steel bar position and the number of steel bars includes the following steps:

[0053] Preprocess the cap beam frame diagram and real-time image data, including denoising, grayscale conversion and image enhancement;

[0054] Use image recognition algorithms to extract features from pre-processed images, identify separated steel bars and count the number of steel bars;

[0055] The number of steel bars of the identified steel bar fixing platform is compared with the number in the cap beam frame drawing to generate a deviation value for the number of steel bars.

[0056] In some embodiments, the real-time image data of the cap beam frame diagram and the rebar fixing platform are first preprocessed. This preprocessing step includes denoising, grayscale conversion, and image enhancement. Denoising is performed to remove noise from the image and improve image quality. Common denoising methods include Gaussian filtering and median filtering. Grayscale conversion converts a color image into a grayscale image, simplifying subsequent image processing steps. Image enhancement adjusts the image's contrast and brightness to make the edges of the rebars clearer, thereby improving image recognizability. Next, an image recognition algorithm is used to extract features from the preprocessed image. The purpose of feature extraction is to identify and separate the rebars in the image. Common image recognition algorithms include edge detection algorithms (such as Canny edge detection), Hough transform, and template matching. These algorithms can accurately identify the edges and positions of rebars, generating a binary image containing rebar features. In the binary image, rebars are marked as white areas, while the background is black. The extracted rebar features are then further processed to count the number of rebars. This step identifies each individual rebar by performing connected domain analysis on the white areas in the binary image. The connected domain analysis algorithm detects and marks connected regions in the image and counts these regions. This method accurately calculates the number of rebars on the rebar fixing platform. Finally, the number of rebars in the identified rebar fixing platform is compared with the designed number in the cap beam frame drawing to generate a deviation value for the rebar quantity.

[0057] Furthermore, the method of verifying the spacing between steel bars according to the image data of the steel bar fixing platform, determining suspicious steel bars based on the spacing between steel bars and generating an alarm includes the following steps:

[0058] Classify and mark the steel bars based on the identified and separated steel bars and steel bar location data;

[0059] Calculate the distance between adjacent bars based on the marking results;

[0060] Compare the distance between adjacent steel bars with the preset standard spacing to identify suspicious steel bars whose spacing does not meet the design requirements.

[0061] In some embodiments, the system acquires real-time image data from the rebar fixing platform and preprocesses this data. Preprocessing steps include denoising, grayscale conversion, and image enhancement to ensure image clarity and accuracy. These preprocessing techniques enhance the rebar features in the image, providing a reliable foundation for subsequent identification and analysis. The system then uses an image recognition algorithm to extract features from the preprocessed image, identifying and isolating the specific locations of the rebars. This image recognition algorithm can be trained and applied using a convolutional neural network (CNN). By learning from a large number of rebar images, it can accurately identify the locations of the rebars. The identified rebar location data serves as the basis for subsequent processing. Based on this identified rebar location data, the system classifies and labels the rebars. This classification and labeling process involves numbering each rebar and marking its specific location in the image. The purpose of classification and labeling is to clearly identify the spatial location of each rebar, facilitating subsequent distance calculation and analysis. After the rebar classification and labeling are completed, the system begins calculating the distances between adjacent rebars. This method iterates through all marked rebar locations and calculates the straight-line distance between each pair of adjacent rebars. The resulting distance data reflects the actual spacing between the rebars during construction. After completing the classification and marking of the rebars, the system begins calculating the distances between adjacent rebars. This is done by traversing all marked rebar locations and calculating the straight-line distance between each pair of adjacent rebars. The resulting distance data reflects the spacing between the rebars during actual construction. Once suspicious rebars are identified, the system highlights their locations in the image and generates an alert. This alert includes the specific location of the suspected rebar, the deviation between the actual spacing and the standard spacing, and recommended action. The system then sends the alert to the construction site operator, prompting them to inspect and adjust the suspected rebar.

[0062] Furthermore, the convolutional neural network is trained by the following steps:

[0063] The steel bars identified and separated in the image data of the steel bar fixing platform are calibrated based on the same color value to obtain the image to be welded;

[0064] Annotate the image to be welded, including marking the weld position with another color value to obtain a target image;

[0065] The convolutional neural network is trained with the image to be welded as input and the target image as output.

[0066] In some embodiments, the system first acquires image data of the rebar fixing platform. This image data contains the spatial position and arrangement of all rebars on the platform. To ensure data quality, the image data undergoes preprocessing, including noise removal, grayscale conversion, and image enhancement. These preprocessing steps improve image clarity and make the edges of the rebar more distinct. Next, the system performs feature extraction on the preprocessed image data to identify and separate the rebars in the image. This feature extraction process typically uses edge detection algorithms, such as Canny edge detection, to identify the outline and position of the rebars. The identified and separated rebars are marked in the image as edges with the same color value, a step called edge calibration. Through edge calibration, the system can clearly identify the specific position and shape of each rebar and generate an image to be welded. After generating the image to be welded, the system annotates the image. Annotation involves clearly marking the specific locations in the image where welding is to be performed. To achieve this, the system marks these weld locations with another identical color value. The annotated weld locations appear as specific color blocks in the image, referred to as target images. This step ensures clear identification of the weld locations, facilitating subsequent neural network training. After image annotation, the system begins training a convolutional neural network (CNN) using the image to be welded as input and the target image as output. A CNN is a deep learning algorithm that, by learning from large amounts of annotated data, can automatically extract high-level features from images and recognize and classify complex image patterns. During training, the system feeds the image to be welded into the CNN. The network gradually extracts feature information from the image through multiple convolutional, pooling, and fully connected layers. Each convolutional layer convolves the input image with several convolutional kernels to extract local features. The pooling layer downsamples the data, retaining important features and improving computational efficiency. The fully connected layer maps the extracted features to the output space, generating the final prediction: the target weld location. During training, the system continuously adjusts the convolution kernel weights and bias vectors using a backpropagation algorithm to minimize the error between the prediction and the target image. This error is typically measured using a mean squared error (MSE) or cross-entropy loss function. Through multiple iterations and training on a large amount of data, the CNN gradually learns how to accurately identify the target weld location in the image.

[0067] Furthermore, the formula of the convolutional neural network is as follows:

[0068] ;

[0069] in, is the output vector, which represents the predicted value of the steel bar to be welded; is the input vector, specifically the image to be welded; and is the row index and column index of the convolution kernel; is the number of channels of the input image; is the weight matrix of the convolutional layer; is the weight parameter of the convolution kernel; is the bias vector of the convolutional layer; is the activation function of the convolutional layer; is the activation function of the fully connected layer; C is the bias vector of the fully connected layer; is the height of the convolution kernel; is the width of the convolution kernel.

[0070] First, the system acquires image data of the rebar fixing platform. This image data undergoes preprocessing, including denoising, grayscale conversion, and image enhancement, to ensure image clarity and accuracy. These preprocessing steps make the rebar features in the image more distinct, providing a reliable foundation for subsequent identification and analysis. During training, the network optimizes parameters using a backpropagation algorithm. The loss function typically uses the mean squared error (MSE) or cross-entropy loss function to measure the error between the predicted value and the true label. Using gradient descent, the network continuously adjusts the weight matrix W and bias vector b to minimize the loss function. The specific training steps include: Forward propagation: The input image is processed through multiple layers of convolution, pooling, and fully connected layers to generate a predicted output. Loss calculation: The loss function is calculated based on the predicted output and the true label. Backpropagation: The gradient of the loss function with respect to each parameter is calculated and the parameters are updated based on these gradients.

[0071] Furthermore, generating a two-dimensional matrix according to the positions of the steel bars to be welded and updating the parameters of the two-dimensional matrix based on the skip welding logic includes the following steps:

[0072] Generate a two-dimensional matrix through the target image output by the convolutional neural network, where each element in the two-dimensional matrix represents the coordinates and status information of a point to be welded;

[0073] The jump welding logic is applied to update the parameters of the two-dimensional matrix, including updating the parameters of the consecutive points to be welded from the parameters to be welded to the parameters of the welded points through a preset step size.

[0074] In some embodiments, a two-dimensional matrix is ​​first generated using the target image output by a convolutional neural network. The convolutional neural network is trained to accurately identify and locate the points to be welded. During the recognition process, the network extracts features from the rebar in the image and generates a target image containing all the points to be welded. Each point to be welded is represented as a bright spot in the image, and the location of these bright spots indicates the location of the rebar to be welded. The generated target image is then processed and converted into a two-dimensional matrix. Each element in the two-dimensional matrix represents the status of a point to be welded, with different states represented by numerical values: 0 for ready to be welded, 1 for already welded, and 2 for currently welded. Initially, the status of all weld points is set to 0 (ready to be welded). Initial weld point selection: A point to be welded is selected from the two-dimensional matrix as the starting point, typically the first point to be welded in the matrix. Its status is updated to 2 (currently welded), and the welding operation begins. Welding operation execution: The welding robot moves to the selected weld point and performs the welding operation. After welding is completed, the status of the weld point is updated from 2 (currently welded) to 1 (already welded). A pre-set jump weld step size (e.g., 2 weld points) is used to determine the location of the next weld point to be welded. The step size can be adjusted according to the specific welding requirements. Repeat the above steps until the status of all welding points is updated to 1 (welded).

[0075] Furthermore, the control of the welding robot to perform displacement and welding based on the verification result of the pre-welding verification module and the updated two-dimensional matrix includes the following steps:

[0076] Obtain the verification results of the pre-welding proofreading module;

[0077] If the deviation value of the number of steel bars is 0 and no suspicious steel bars are found, the welding robot is controlled to move and weld according to the updated two-dimensional matrix;

[0078] If the deviation value of the number of steel bars is not 0 or suspicious steel bars are found, the control stops the welding robot from moving and welding.

[0079] In some embodiments, the system first obtains the verification results from the pre-weld verification module. The pre-weld verification module includes a cap beam reinforcement frame unit and a suspicious reinforcement identification unit, responsible for verifying the number and spacing of reinforcement bars. The cap beam reinforcement frame unit compares the cap beam frame diagram with the image data of the reinforcement fixing platform to determine whether the actual number of reinforcement bars is consistent with the design drawings. The suspicious reinforcement identification unit verifies the reinforcement spacing based on the image data of the reinforcement fixing platform, identifying and marking suspicious reinforcement bars that do not meet the design requirements. After obtaining the verification results, the system first checks the deviation value of the reinforcement number. A deviation value of 0 indicates that the actual number of reinforcement bars is exactly the same as the design drawings. In this case, the system also checks for suspicious reinforcement bars. If the verification results indicate that no suspicious reinforcement bars are found, the system controls the welding robot's movement and welding operations based on the updated two-dimensional matrix. For example, the state of an element in the two-dimensional matrix is ​​0, indicating that it is ready to be welded, and 1 indicates that it is already welded. The welding robot first moves to the first weld point in the two-dimensional matrix with a state of 0 and begins welding. After welding is completed, the system updates the state of that weld point to 1 (welded) and selects the next weld point based on the preset skip welding logic. The welding robot continues to move and weld according to the instructions of the two-dimensional matrix until all weld points are complete. However, if the verification results indicate a non-zero deviation in the number of rebars or if a suspicious rebar is detected, the system immediately stops the robot's movement and welding operations. In this case, the system generates an alarm and notifies the on-site operator of the deviation and the specific location of the suspicious rebar. The operator is required to check and adjust the actual construction situation to ensure that the number and spacing of rebars meet the design requirements. For example, during actual construction, if the cap beam reinforcement frame unit detects that the number of rebars is less than the design drawing requirements, the system will immediately pause welding and prompt the operator to replenish the missing rebar. Similarly, if the suspicious rebar identification unit identifies that the rebar spacing is too large or too small, the system will prompt the operator to readjust the spacing. Through these steps, the system effectively ensures the accuracy and quality of welding operations. Based on the verification results of the pre-weld verification module, the system can dynamically adjust the welding plan to avoid welding defects caused by issues with the number or spacing of rebars. This not only improves construction efficiency but also significantly enhances the overall stability and safety of the structure.

[0080] Furthermore, controlling the welding robot to perform displacement and welding includes the following steps:

[0081] According to the parameters to be welded in the two-dimensional matrix, the welding robot is controlled to move the X-axis and Y-axis to the position of the parameters to be welded and weld through the welding head;

[0082] After each welding point is completed, the location is obtained and the two-dimensional matrix is ​​updated, and the parameters to be welded are updated to the welded parameters.

[0083] Furthermore, judging welding abnormality according to working parameters includes the following steps:

[0084] Obtain the welding head depth, welding head pressure and sound parameters and the corresponding timestamp of a single welding when there is no abnormality in the history;

[0085] Based on the same timestamp, the value ranges of welding head pressure and sound parameters at different welding head depths are obtained;

[0086] If the welding head pressure or sound parameters do not belong to the value range at the current welding head depth during real-time welding, it is judged as welding abnormality;

[0087] If the welding head pressure or sound parameters do not belong to the value range at the current welding head depth during real-time welding work, it is judged that there is no abnormality.

[0088] In some embodiments, the system first obtains historical data on weld head depth, weld head pressure, and acoustic parameters during normal operation, along with the timestamps corresponding to these parameters for a single weld. This historical data is collected by long-term monitoring and recording of changes in various parameters during normal welding operations, ensuring that the data accurately reflects the parameter ranges under normal operating conditions. After obtaining the historical data without abnormalities, the system then obtains the ranges of weld head pressure and acoustic parameters at different weld head depths based on the same timestamps. This step uses data analysis methods to statistically analyze the historical data and determine the normal ranges of weld head pressure and acoustic parameters at each weld head depth. For example, for a weld head depth of 5mm, the system might determine that the normal ranges are 50-60N for weld head pressure and 70-80dB for acoustic parameters. Next, the system monitors the weld head depth, weld head pressure, and acoustic parameters during real-time welding. When the welding robot begins welding, the system records the current weld head depth in real time and compares and determines the weld head pressure and acoustic parameters based on the normal ranges corresponding to the current weld head depth. Real-time parameter acquisition: The system acquires the weld head depth, weld head pressure, and acoustic parameters during the current welding operation in real time. For example, at a given moment, the system records the current weld head depth as 8mm, weld head pressure as 65N, and acoustic parameters as 85dB. The system compares the real-time weld head pressure and acoustic parameters with preset normal ranges. If the current weld head depth is 5mm, the normal range for weld head pressure is 50-60N, and the normal range for acoustic parameters is 70-80dB. The system determines whether the current weld head pressure and acoustic parameters are within the normal range. If the weld head pressure or acoustic parameters are not within their ranges for the current weld head depth, the system determines that the welding is abnormal. For example, if the current weld head pressure is 65N, which is outside the normal range of 50-60N, the system determines that the welding is abnormal, generates an alarm, and records the time and parameters of the abnormality. The system continuously monitors the weld head depth, weld head pressure, and acoustic parameters during each welding cycle and updates the judgment results in real time. If the welding head pressure and sound parameters remain within the normal range throughout the welding process, the system will determine that there are no abnormalities and the welding operation is proceeding normally. In particular, by monitoring the welding head depth, welding head pressure, and sound parameters during the welding process, welding abnormalities caused by misaligned rebar or the presence of foreign matter can be promptly detected. For example, when rebar is misaligned, the welding head pressure may deviate significantly from the normal range, and the sound parameters will also fluctuate abnormally. The system can immediately identify this abnormality, suspend the welding operation, and issue an alarm to notify the operator to check and make adjustments. This method not only improves the accuracy of the welding process, but also effectively avoids welding defects caused by misaligned rebar or foreign matter. Through the system's real-time monitoring and automatic exception handling functions, the cap beam intelligent welding cloud control system can significantly improve welding quality, ensuring that every weld meets the design requirements, thereby protecting the overall stability and safety of bridges and other large structures.

[0089] The above embodiments are merely descriptions of preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary engineering technicians in this field should fall within the scope of protection determined by the claims of the present invention.

Claims

1. A cap beam intelligent welding cloud control system, characterized in that: include: An image receiving module, a pre-welding proofreading module, a welding robot cloud control module, and an abnormality monitoring module, wherein the pre-welding proofreading module includes a cap beam reinforcement frame unit and a suspicious reinforcement judgment unit, the image receiving module, the pre-welding proofreading module, and the abnormality monitoring module are respectively connected to the welding robot cloud control module, and the image receiving module is connected to the pre-welding proofreading module; The image receiving module is used to receive image data of the steel bar fixing platform; The cap beam reinforcement frame unit is used to receive the cap beam frame diagram, compare the cap beam frame diagram with the image data of the reinforcement fixing platform, and verify the number of reinforcement bars; The suspicious steel bar determination unit is used to verify the steel bar spacing according to the image data of the steel bar fixing platform, and determine the suspicious steel bars based on the steel bar spacing; The welding robot cloud control module determines the steel bar to be welded position from the image data of the steel bar fixing platform based on the convolutional neural network, generates a two-dimensional matrix according to the steel bar to be welded position, updates the parameters of the two-dimensional matrix based on the jump welding logic, controls the welding robot to move and weld based on the verification result of the pre-welding proofreading module and the updated two-dimensional matrix, receives the displacement and welding data of the welding robot in real time and updates the two-dimensional matrix; The abnormality monitoring module is used to receive the working parameters of the welding robot during welding, determine the welding abnormality according to the working parameters, and issue a stop command to the welding robot cloud control module. The working parameters include welding head pressure, welding head depth and sound parameters; The convolutional neural network is trained by the following steps: The steel bars identified and separated in the image data of the steel bar fixing platform are calibrated based on the same color value to obtain the image to be welded; Annotate the image to be welded, including marking the weld position with another color value to obtain a target image; The convolutional neural network is trained with the image to be welded as input and the target image as output; The formula of the convolutional neural network is as follows: ， in, is the output vector, which represents the predicted value of the steel bar to be welded; is the input vector, specifically the image to be welded; and is the row index and column index of the convolution kernel; is the number of channels of the input image; is the weight matrix of the convolutional layer; is the weight parameter of the convolution kernel; is the bias vector of the convolutional layer; is the activation function of the convolutional layer; is the activation function of the fully connected layer; C is the bias vector of the fully connected layer; is the height of the convolution kernel; is the width of the convolution kernel.

2. The cap beam intelligent welding cloud control system according to claim 1 is characterized in that: The verification of the number of steel bars based on the comparison of the cap beam frame diagram and the image data of the steel bar fixing platform includes the following steps: Preprocess the cap beam frame diagram and real-time image data, including denoising, grayscale conversion and image enhancement; Use image recognition algorithms to extract features from pre-processed images, identify separated steel bars and count the number of steel bars; The number of steel bars of the identified steel bar fixing platform is compared with the number in the cap beam frame drawing to generate a deviation value for the number of steel bars.

3. The cap beam intelligent welding cloud control system according to claim 2 is characterized in that: Verifying the steel bar spacing based on the image data of the steel bar fixing platform and determining suspicious steel bars based on the steel bar spacing comprises the following steps: Classify and mark the steel bars based on the identified and separated steel bars and steel bar location data; Calculate the distance between adjacent bars based on the marking results; Compare the distance between adjacent steel bars with the preset standard spacing to identify suspicious steel bars whose spacing does not meet the design requirements.

4. The cap beam intelligent welding cloud control system according to claim 1 is characterized in that: Generating a two-dimensional matrix according to the steel bar to be welded positions and updating the parameters of the two-dimensional matrix based on the jump welding logic comprises the following steps: Generate a two-dimensional matrix through the target image output by the convolutional neural network, where each element in the two-dimensional matrix represents the coordinates and status information of a point to be welded; The jump welding logic is applied to update the parameters of the two-dimensional matrix, including updating the parameters of the consecutive points to be welded from the parameters to be welded to the parameters of the welded points through a preset step size.

5. The cap beam intelligent welding cloud control system according to claim 3 is characterized in that: The method of controlling the welding robot to perform displacement and welding based on the calibration result of the pre-welding calibration module and the updated two-dimensional matrix comprises the following steps: Obtain the verification results of the pre-welding proofreading module; If the deviation value of the number of steel bars is 0 and no suspicious steel bars are found, the welding robot is controlled to move and weld according to the updated two-dimensional matrix; If the deviation value of the number of steel bars is not 0 or suspicious steel bars are found, the control stops the welding robot from moving and welding.

6. The cap beam intelligent welding cloud control system according to claim 1 is characterized in that: Controlling the welding robot to perform displacement and welding comprises the following steps: According to the parameters to be welded in the two-dimensional matrix, the welding robot is controlled to move the X-axis and Y-axis to the position of the parameters to be welded and weld through the welding head; After each welding point is completed, the location is obtained and the two-dimensional matrix is ​​updated, and the parameters to be welded are updated to the welded parameters.

7. The cap beam intelligent welding cloud control system according to claim 1 is characterized in that: The method of judging welding abnormality according to working parameters comprises the following steps: Obtain the welding head depth, welding head pressure and sound parameters and the corresponding timestamp of a single welding when there is no abnormality in the history; Based on the same timestamp, the value ranges of welding head pressure and sound parameters at different welding head depths are obtained; If the welding head pressure or sound parameters do not belong to the value range at the current welding head depth during real-time welding, it is judged as welding abnormality; If the welding head pressure or sound parameters in the real-time welding work belong to their value range at the current welding head depth, it is judged that there is no abnormality.

Images