Tube welding machine and method

Abstract

This invention discloses a pipe welding machine and welding method, belonging to the field of nuclear power technology. The pipe welding method includes the following steps: before welding, taking pictures of different positions of the weld seam to obtain continuous image frames of different positions of the circumferential weld seam; setting up a trajectory prediction network, taking the continuous image frames as input, and outputting a welding trajectory; and welding the weld seam according to the obtained welding trajectory. The pipe welding machine and welding method disclosed in this invention can obtain accurate pipe welding trajectories, thereby guiding the welding head to achieve precise welding.

Description

Technical Field

[0001] This invention relates to a pipe welding machine and welding method, belonging to the field of nuclear power technology. Background Technology

[0002] Steam generators have a large number of heat exchange tubes, with some nuclear power projects having more than 10,000 tubes. The tube diameter is relatively small, and the welding process between the tubes and the tube sheet holes mostly relies on manual welding. However, manual welding is inefficient and the welding yield fluctuates greatly.

[0003] Due to the small pipe diameter and high welding precision requirements, most existing automatic pipe welding machines cannot directly reach the welding position. Although some pipe welding machines can reach the welding position, the welding yield is low, and phenomena such as incomplete penetration are prone to occur, resulting in a high rework rate after welding.

[0004] Therefore, it is necessary to conduct in-depth research on existing steam generator tube sheet and heat exchanger tube welding devices and methods to solve the above problems. Summary of the Invention

[0005] To overcome the above problems, in-depth research was conducted, and a pipe welding machine was designed, including a welding power source, a system controller, a wire feeding mechanism, and a movable welding head. The welding head has a tungsten electrode holder with a tungsten electrode on it. The welding head is equipped with a light source and a camera that can emit a linear light band. The plane of the linear light emitted by the light source is parallel to the axis of the heat exchange tube. The camera is used to capture the light band illuminating the heat exchange tube, and the weld position and weld depth are determined by the bending of the light band.

[0006] In a preferred embodiment, the light emitted by the light source is not collinear with any radius in the circular cross-section of the heat exchange tube.

[0007] This invention also discloses a pipe welding method, comprising the following steps:

[0008] S1. Before welding, take pictures of different positions of the weld to obtain continuous image frames of different positions of the annular weld.

[0009] S2. Set up a trajectory prediction network, take continuous image frames as input, and output the welding trajectory;

[0010] S3. Weld the seam according to the obtained welding trajectory.

[0011] In a preferred embodiment, in S2, the trajectory prediction network includes an image branch subnetwork, a parameter branch subnetwork, a feature fusion subnetwork, and a process decision subnetwork, wherein,

[0012] The input to the image branch subnetwork is an image frame captured by the camera, and its output is the depth and width of the weld.

[0013] The parameter branch sub-network maps the relevant parameters of the input heat exchanger tube to be welded to the feature space;

[0014] The feature fusion subnetwork is used to concatenate the outputs of the image branch subnetwork and the parameter branch subnetwork to generate a comprehensive feature vector.

[0015] The process decision sub-network generates welding trajectories based on comprehensive feature vectors.

[0016] In a preferred embodiment, the image branch subnetwork includes a feature extraction unit and a geometric feature decoder arranged sequentially.

[0017] The feature extraction unit is used to extract features from the image sequence to obtain a feature map, and the geometric feature decoder is used to map the extracted features into weld depth and width information.

[0018] In a preferred embodiment, the feature extraction unit includes a multi-layer convolutional structure and a spatial attention mechanism. The multi-layer convolutional structure can extract image features from the image sequence, and the spatial attention mechanism can focus on the extracted image features to obtain a weighted feature map.

[0019] In a preferred embodiment, the geometric feature decoder includes a depth prediction subunit and a width prediction subunit.

[0020] The depth prediction subunit predicts information related to weld depth based on the extracted feature map.

[0021] The width prediction subunit predicts information related to weld width based on the extracted feature map.

[0022] In a preferred embodiment, the geometric feature decoder further includes an edge location prediction subunit, which predicts a probability heatmap of the weld edge based on the extracted feature map.

[0023] In a preferred embodiment, the process decision subnetwork includes a two-stage LSTM network and an output layer.

[0024] The first-level LSTM network takes the comprehensive feature vector as input and outputs a short-term spatiotemporal sequence to capture the continuity features of the weld.

[0025] The second-stage LSTM network takes short-term spatiotemporal sequences as input and outputs long-term dependency sequences.

[0026] In a preferred embodiment, the output layer includes a shared feature processing unit and multiple parallel task output branch units.

[0027] The shared feature processing unit is a fully connected layer that takes long-term dependency sequences as input and outputs enhanced feature vectors. The multiple parallel task output branch units include a trajectory offset output branch unit, a welding speed branch unit, and a welding current branch unit.

[0028] The trajectory offset output branch unit outputs the welding head movement trajectory, the welding speed branch unit outputs the corresponding welding speed, and the welding current branch unit outputs the corresponding welding current.

[0029] The beneficial effects of this invention include:

[0030] (1) It can obtain accurate pipe welding trajectory, thereby guiding the welding head to achieve precise welding;

[0031] (2) Strong welding stability and strong uniformity. Attached Figure Description

[0032] Figure 1 A schematic diagram of a portion of the pipe welding machine equipment according to a preferred embodiment of the present invention is shown.

[0033] Figure 2 A schematic diagram showing the relative position of the linear beam and the weld seam of a pipe welding machine according to a preferred embodiment of the present invention is provided.

[0034] Figure 3 A schematic diagram of the linear beam position in a pipe welding machine according to a preferred embodiment of the present invention is shown.

[0035] Figure 4 A schematic diagram of the twisted structure of the linear light strip after the light source illuminates the weld seam according to a preferred embodiment of the present invention is shown.

[0036] Figure 5 A schematic diagram of a welding method according to a preferred embodiment of the present invention is shown.

[0037] Explanation of icon numbers

[0038] 1-Light source;

[0039] 11-Linear light;

[0040] 2-Camera;

[0041] 3-Weld;

[0042] 4-Tungsten electrode. Detailed Implementation

[0043] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Through these descriptions, the features and advantages of the present invention will become clearer and more apparent.

[0044] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments. Although various aspects of embodiments are shown in the accompanying drawings, the drawings are not necessarily drawn to scale unless specifically indicated otherwise.

[0045] The welding position of the steam generator tube sheet and heat exchange tubes is annular.

[0046] The pipe welding machine provided by the present invention is similar to conventional welding machine in that it has a welding power source, a system controller, a wire feeding mechanism, and a movable welding head.

[0047] The welding power source is used to provide the electrical energy required for welding.

[0048] The system controller is used to control the movement of various parts of the welding machine;

[0049] The wire feeding mechanism is used to uniformly deliver the welding wire to the welding area;

[0050] The welding head has a tungsten electrode clip, and a tungsten electrode is disposed on the tungsten electrode clip.

[0051] Unlike traditional welding equipment, such as Figure 1 As shown, the welding head is equipped with a light source 1 and a camera 2 capable of emitting linear light bands.

[0052] The plane containing the linear light 11 emitted by the light source is parallel to the axis of the heat exchange tube, such that the linear light 11 is perpendicular to the projection of the weld 3 on the plane containing the linear light. Figure 2 As shown;

[0053] The camera is used to capture a linear light band illuminating the heat exchange tube, and the location and depth of the weld are determined by the bending of the light band.

[0054] According to the present invention, in the plane where the weld is located, the light emitted by the light source is not collinear with any radius in the circular cross-section of the heat exchange tube, such as... Figure 3 As shown, this arrangement ensures that the light rays are obliquely incident on the weld along the heat exchanger tube cross-section, thereby guaranteeing a curvature change in the light band at the weld location. According to the present invention, when the linear light emitted by the light source is collinear with the radius of the circular cross-section of the heat exchanger tube, the curvature change of the light band at the weld location is small, and the accuracy of obtaining weld depth information decreases significantly.

[0055] While directly photographing the weld seam with a camera can identify the weld seam, its accuracy is low, and the information obtained, such as the depth and width of the weld seam, will have a large deviation, resulting in poor automatic welding effect.

[0056] In this invention, a linear light strip is used to illuminate the weld. When the linear light strip illuminates the weld, it twists, and this twist not only characterizes the planar position of the weld but also reflects its depth. Figure 4 As shown, the linear light band is more sensitive to irregularities at the weld edge, thus more accurately reflecting the morphology of the position to be welded, providing a basis for fine welding.

[0057] In a preferred embodiment, the tungsten electrode clamp is retractable. Before welding, when the light source and camera are working, the tungsten electrode clamp is retracted to prevent the tungsten electrode from blocking the light. During welding, the tungsten electrode clamp extends to weld the weld seam.

[0058] In a preferred embodiment, both the light source and the camera are located on the plane of the weld.

[0059] In a preferred embodiment, both the light source and the camera are fixedly mounted on the welding head, so that their positions are relatively fixed, ensuring image consistency when photographing different positions of the weld.

[0060] According to the present invention, the welding head can rotate about the heat exchange tube axis as the center line, thereby photographing and welding the annular weld.

[0061] In this invention, the specific relative positions between the light source and the camera are not limited, and those skilled in the art can freely set them according to actual needs.

[0062] In this invention, the specific structure of the welding head is not limited. Those skilled in the art can freely set it according to actual needs, as long as it can fix the light source, camera and tungsten electrode clamp, and rotate with the heat exchange tube axis as the center line.

[0063] Obviously, similar to traditional welding heads, the welding head is also equipped with necessary units such as protective gas nozzles. In this invention, the specific structure of these units is not limited, and those skilled in the art can freely set them according to actual needs.

[0064] This invention discloses a pipe welding method, such as... Figure 5 As shown, it includes the following steps:

[0065] S1. Before welding, take pictures of different positions of the weld to obtain continuous image frames of different positions of the annular weld.

[0066] S2. Set up a trajectory prediction network, take continuous image frames as input, and output the welding trajectory;

[0067] S3. Weld the seam according to the obtained welding trajectory.

[0068] In S1, the welding head is rotated to take pictures of different positions of the circumferential weld in sequence.

[0069] In S2, the welding trajectory includes the welding head trajectory curve and the welding current and welding speed at different positions on the curve.

[0070] The trajectory prediction network includes an image branch subnetwork, a parameter branch subnetwork, a feature fusion subnetwork, and a process decision subnetwork, wherein...

[0071] The input to the image branch subnetwork is an image frame captured by the camera, and its output is the depth and width of the weld. Preferably, it also includes a heat map of the weld edge.

[0072] Preferably, the image frame is processed into grayscale before input.

[0073] In a preferred embodiment, the image branch subnetwork includes a feature extraction unit and a geometric feature decoder arranged sequentially.

[0074] The feature extraction unit is used to extract features from the image frame to obtain a feature map, and the geometric feature decoder is used to map the extracted features into weld depth and width information.

[0075] In a preferred embodiment, the feature extraction unit includes a multi-layer convolutional structure and a spatial attention mechanism. The multi-layer convolutional structure can extract image features from image frames, and the spatial attention mechanism can focus on the extracted image features to obtain a weighted feature map.

[0076] According to the present invention, by setting a spatial attention mechanism in the feature extraction unit, the grating distortion features of the weld area are focused on and the background noise is suppressed, thereby adaptively enhancing the response weight of the weld edge area and improving the feature quality.

[0077] The multi-layer convolutional structure and spatial attention mechanism are commonly used structures in neural networks. In this invention, their specific parameters are not limited, and those skilled in the art can set them freely based on experience.

[0078] Preferably, the geometric feature decoder includes a depth prediction subunit and a width prediction subunit.

[0079] The depth prediction subunit predicts information related to weld depth based on extracted feature maps. Preferably, it has a series of dilated convolutional layers, a global max pooling layer, and a regression layer connected in sequence.

[0080] Among them, multiple convolutional layers can expand the receptive field to capture large-scale light band distortion, global max pooling can transform distortion features into feature vectors, and regression layers transform feature vectors into depth scalars.

[0081] The width prediction subunit predicts information related to the weld width based on the extracted feature map. Preferably, it has a series of dilated convolutional layers, a global average pooling layer, and a regression layer connected in sequence.

[0082] Among them, multi-layer dilated convolutional layers can expand the receptive field, global average pooling layers can comprehensively evaluate the overall distortion region to obtain feature vectors, and regression layers convert feature vectors into width scalars.

[0083] Preferably, the geometric feature decoder further includes an edge position prediction subunit, which predicts a probability heatmap of the weld edge based on the extracted feature map.

[0084] Preferably, the edge location prediction subunit has a transposed convolutional layer, a coordinate enhancement layer, a cascaded convolutional layer, and a heatmap generation layer.

[0085] The transposed convolutional layer recovers spatial details through learnable upsampling.

[0086] The coordinate enhancement layer is used to generate two matrices with the same size as the feature map output by the transposed convolutional layer: a horizontal coordinate matrix and a vertical coordinate matrix. The values ​​in each column of the horizontal coordinate matrix change linearly from -1 to 1, and the values ​​in each row of the vertical coordinate matrix change linearly from -1 to 1. The horizontal coordinate matrix and the vertical coordinate matrix are concatenated to the feature map output by the transposed convolutional layer and then output.

[0087] The use of coordinate enhancement layers can solve the translation invariance defect of CNN network structure, enabling the network to perceive absolute position and improve edge localization accuracy.

[0088] Cascaded convolutional layers are used to fuse the channel features output by the coordinate enhancement layer, thereby reducing the number of channels and parameters.

[0089] The heatmap generation layer performs 1×1 convolution compression and bilinear upsampling on the output of the cascaded convolutional layer to generate a single-channel heatmap. Each pixel value in the heatmap represents the probability that the pixel is an edge (the pixel value is 0 or 1, where 0 indicates a non-edge and 1 indicates an edge), thereby achieving sub-pixel weld edge localization and providing a reference for the subsequent weld head movement trajectory.

[0090] The parameter branch sub-network maps the relevant parameters of the input heat exchanger tube to be welded to the feature space, and adopts a parameter embedding layer structure.

[0091] Preferably, the relevant parameters of the heat exchange tube to be welded include the inner diameter of the tube and the material thickness.

[0092] The feature fusion subnetwork is used to concatenate the outputs of the image branch subnetwork and the parameter branch subnetwork to generate a comprehensive feature vector.

[0093] The process decision sub-network generates welding trajectories based on comprehensive feature vectors.

[0094] Preferably, the process decision sub-network includes a two-stage LSTM network and an output layer.

[0095] The first-level LSTM network takes the comprehensive feature vector as input and outputs a short-term spatiotemporal sequence to capture the continuity features of the weld.

[0096] The second-stage LSTM network takes short-term spatiotemporal sequences as input and outputs long-term dependency sequences.

[0097] The output layer includes a shared feature processing unit and multiple parallel task output branch units.

[0098] The shared feature processing unit is a fully connected layer that takes long-term dependency sequences as input and outputs enhanced feature vectors.

[0099] The multiple parallel task output branch units include a trajectory offset output branch unit, a welding speed branch unit, and a welding current branch unit.

[0100] Preferably, the trajectory offset output branch unit, welding speed branch unit, and welding current branch unit all adopt a fully connected layer and a ReLU activation function, taking the enhanced feature vector as input, and output the welding trajectory curve coordinates, the corresponding welding speed, and the welding current, respectively.

[0101] In this invention, a two-level LSTM network is used to handle the continuous changes in the circumferential weld, making the predicted trajectory smooth and continuous, and avoiding welding abnormalities caused by abrupt changes.

[0102] According to the present invention, the trajectory offset output branch unit outputs the welding head movement trajectory, the welding speed branch unit outputs the corresponding welding speed, and the welding current branch unit outputs the corresponding welding current.

[0103] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," "outer," "front," and "rear," etc., indicate the orientation or positional relationship based on the orientation or positional relationship in the working state of this invention, and are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this invention. Furthermore, the terms "first," "second," "third," and "fourth" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0104] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal communication between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0105] The present invention has been described above with reference to preferred embodiments; however, these embodiments are merely exemplary and illustrative. Various substitutions and modifications can be made to the present invention based on these embodiments, all of which fall within the scope of protection of the present invention.

Claims

1. A method for welding pipe bodies, characterized in that, The process is performed using a tube welding machine, which includes a welding power source, a system controller, a wire feeding mechanism, and a movable welding head. The welding head has a tungsten electrode holder with a tungsten electrode on it. The welding head is also equipped with a light source and a camera that emit a linear light band. The plane of the linear light emitted by the light source is parallel to the axis of the heat exchange tube. The camera is used to capture the light band illuminating the heat exchange tube, and the weld position and depth are determined by the bending of the light band. The tube welding method includes the following steps: S1. Before welding, take pictures of different positions of the weld to obtain continuous image frames of different positions of the annular weld. S2. Set up a trajectory prediction network, take continuous image frames as input, and output welding trajectory. The welding trajectory includes the welding head trajectory curve and the welding current and welding speed at different positions on the curve. S3. Weld the seam according to the obtained welding trajectory; In S2, the trajectory prediction network includes an image branch subnetwork, a parameter branch subnetwork, a feature fusion subnetwork, and a process decision subnetwork, wherein, The input to the image branch subnetwork is an image frame captured by the camera, and its output is the depth and width of the weld. The parameter branch sub-network maps the relevant parameters of the input heat exchanger tube to be welded to the feature space; The feature fusion subnetwork is used to concatenate the outputs of the image branch subnetwork and the parameter branch subnetwork to generate a comprehensive feature vector. The process decision sub-network generates welding trajectories based on comprehensive feature vectors; The image branch subnetwork includes a feature extraction unit and a geometric feature decoder arranged sequentially. The feature extraction unit is used to extract features from the image sequence to obtain a feature map, and the geometric feature decoder is used to map the extracted features into weld depth and width information. The geometric feature decoder includes a depth prediction subunit and a width prediction subunit. The depth prediction subunit predicts information related to weld depth based on the extracted feature map. The width prediction subunit predicts information related to weld width based on the extracted feature map. The geometric feature decoder also includes an edge position prediction subunit, which predicts a probability heat map of the weld edge based on the extracted feature map; The process decision subnetwork includes a two-stage LSTM network and an output layer. The first-level LSTM network takes the comprehensive feature vector as input and outputs a short-term spatiotemporal sequence to capture the continuity features of the weld. The second-stage LSTM network takes short-term spatiotemporal sequences as input and outputs long-term dependency sequences. The output layer includes a shared feature processing unit and multiple parallel task output branch units; The shared feature processing unit is a fully connected layer that takes long-term dependency sequences as input and outputs enhanced feature vectors. The multiple parallel task output branch units include a trajectory offset output branch unit, a welding speed branch unit, and a welding current branch unit. The trajectory offset output branch unit outputs the welding head movement trajectory, the welding speed branch unit outputs the corresponding welding speed, and the welding current branch unit outputs the corresponding welding current.

2. The pipe welding method according to claim 1, characterized in that, The feature extraction unit includes a multi-layer convolutional structure and a spatial attention mechanism. The multi-layer convolutional structure can extract image features from the image sequence, and the spatial attention mechanism can focus on the extracted image features to obtain a weighted feature map.

Images