Intelligent laser welding workstation for high-voltage cable

Abstract

The invention discloses an intelligent laser welding workstation for a high-voltage cable. The intelligent laser welding workstation comprises an operation table, a mechanical arm, a welding head, a camera assembly, a power assembly, a clamping assembly and a control piece. A control part is installed on the outer side of the operation table and comprises an image acquisition module, an image processing module, a track planning module, a motion control module, a quality detection module, a parameter adjustment module and a synchronous control module. Visual guidance and real-time monitoring in the welding process are achieved through the camera assembly, a welding center point is accurately positioned through an image processing algorithm, the track planning module generates a smooth welding path, and the parameter adjusting module dynamically adjusts welding parameters according to a molten pool image. The synchronous control module receives a production line signal to realize full-process automation of automatic clamping, welding and loosening; efficient and stable laser welding seamlessly integrated with the cable manufacturing process is achieved, and the automation degree, welding precision and product reliability of a production line are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent welding technology, specifically to an intelligent laser welding workstation for high-voltage cables. Background Technology

[0002] With the rapid development of the power cable manufacturing industry, especially the continuous expansion of high-voltage and ultra-high-voltage cable production scale, higher requirements have been placed on the welding quality and production efficiency of cable conductor connections. During cable production, metal conductor units need to achieve efficient and reliable connection to ensure the cable's conductivity and mechanical strength. Currently, cable conductor connections are mainly achieved using resistance welding, argon arc welding, or manual laser welding.

[0003] However, existing technologies have the following shortcomings in practical applications:

[0004] Firstly, the level of automation is low, relying heavily on manual operation. Traditional cable welding mostly uses manual positioning and welding methods. Operators need to adjust the position of the welding torch and welding parameters based on experience, which is not only labor-intensive and inefficient, but also the welding quality is highly dependent on the operator's skill level, making it difficult to guarantee product consistency and stability.

[0005] Secondly, the positioning accuracy is insufficient, making it difficult to meet the high standards required for high-voltage cables. High-voltage cable conductor butt joints have strict requirements for concentricity, penetration depth, and appearance. Existing equipment lacks a precise visual guidance and positioning mechanism, making it impossible to automatically identify and accurately position the welding points, which easily leads to quality problems such as misaligned welds and incomplete welds.

[0006] Third, there is a lack of coordinated control capabilities with the production line cycle time. Most existing welding equipment operates at independent workstations, requiring manual loading and unloading and manual triggering of welding actions. This makes it impossible to achieve seamless integration with upstream processes such as cable length cutting and conveying, resulting in limited overall production line efficiency and difficulty in meeting the needs of continuous and large-scale production.

[0007] Fourth, the welding process lacks a real-time monitoring and dynamic adjustment mechanism. During the welding process, the state of the molten pool changes due to factors such as material differences and surface cleanliness. Existing technologies cannot dynamically adjust the laser power and welding speed based on real-time acquired images of the molten pool, resulting in large fluctuations in welding quality and a high scrap rate.

[0008] Fifth, weld quality inspection is lagging behind. Current technologies mostly rely on manual sampling inspection after welding is completed, which cannot achieve online automatic inspection of each weld. Defects are not detected in a timely manner, increasing rework costs and material waste.

[0009] Therefore, developing a high-voltage cable intelligent laser welding workstation capable of high-precision automatic positioning, real-time control of the welding process, coordination with the production line cycle, and online quality inspection has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0010] This invention provides an intelligent laser welding workstation for high-voltage cables. It achieves visual guidance and real-time monitoring of the welding process through a camera component, accurately locates the welding center point through an image processing algorithm, generates a smooth welding path through a trajectory planning module, dynamically adjusts welding parameters based on the molten pool image through a parameter adjustment module, and automatically controls the entire process of clamping, welding, and releasing by receiving signals from the production line. This invention solves the problems mentioned in the background art.

[0011] The present invention provides the following technical solution: a high-voltage cable intelligent laser welding workstation, including an operating table (1), a robotic arm (2) is installed on the top of the operating table (1), a welding head (21) and a camera assembly (22) are installed on the robotic arm (2), a power assembly (3) is installed on both sides of the top of the operating table (1), and a clamping assembly (4) is installed on the power assembly (3).

[0012] The power assembly (3) includes a fixed frame (36) fixedly installed on the top of the operating table (1). A motor (31) is installed at the bottom of the fixed frame (36). A rotating shaft (32) is fixedly connected to the output end of the motor (31). Several rotating rods (35) are rotatably connected to the top of the fixed frame (36). Synchronous pulleys (34) are fixedly connected to the outer walls of the rotating shaft (32) and the rotating rods (35). A synchronous belt (33) is sleeved between the synchronous pulleys (34).

[0013] The clamping assembly (4) includes two slide plates (45) fixedly mounted on the timing belt (33). The top of the slide plate (45) is fixedly connected to a mounting bracket (42). The top of the mounting bracket (42) is fixedly connected to a gripper (41). The top of the mounting bracket (36) is fixedly connected to a guide rail (44). The bottom of the mounting bracket (42) is fixedly connected to a slider (43). The bottom of the slider (43) is slidably connected to the guide rail (44).

[0014] A control component (5) is installed on the outside of the operating table (1), and the control component (5) is electrically connected to the robotic arm (2), welding head (21), camera assembly (22) and motor (31).

[0015] The control element (5) includes:

[0016] Image acquisition module; used to control the camera assembly (22) to acquire images from the cable end;

[0017] Image processing module: used to calculate the coordinates of the welding center point;

[0018] Trajectory planning module: used to generate the motion trajectory of the robotic arm (2);

[0019] Motion control module: used to control the robotic arm (2) to accurately track the welding trajectory;

[0020] Quality inspection module: Used to assess weld quality after welding is completed;

[0021] Parameter adjustment module: Used to dynamically adjust welding parameters during the welding process;

[0022] Synchronization control module: Used to coordinate with the production line cycle time.

[0023] The present invention has the following beneficial effects:

[0024] 1. The camera assembly, in conjunction with the control components, enables visual guidance and real-time monitoring of the welding process, greatly improving the automation and precision of welding. Meanwhile, the clamping assembly adopts a structure design with two sliding plates moving relative to each other via a synchronous belt. Utilizing the opposite movement direction of the two sides of the synchronous belt, the synchronous centering and clamping of the two jaws can be achieved with only one motor. This not only simplifies the transmission structure and reduces costs, but also ensures the synchronicity and centering accuracy of clamping, making it suitable for welding various specifications of high-voltage cables.

[0025] 2. This invention uses a visual recognition module to preprocess and fit the center of the cable end image, and combines coordinate transformation to achieve precise positioning of the welding point; combined with trajectory smoothing and servo control, it ensures continuous and stable operation of the welding head, and improves the accuracy of welding trajectory tracking.

[0026] 3. This invention acquires images of the molten pool in real time, dynamically adjusts laser power and welding speed, and constructs a closed-loop control mechanism for welding parameters to suppress welding defects and improve the stability and consistency of welding quality.

[0027] 4. This invention achieves online detection and defect alarm of welding quality through automatic analysis and feature comparison of weld seam images, replacing traditional manual sampling inspection, improving detection efficiency, and providing data support for product quality traceability;

[0028] 5. This invention receives production line start / stop signals through a synchronous control module, automatically triggering the entire process of clamping, welding, inspection, and releasing, achieving seamless connection with upstream processes and improving the overall automation level and production efficiency of the production line; Attached Figure Description

[0029] Figure 1 This is a schematic diagram of the overall structure of the present invention;

[0030] Figure 2 This is a side view of the present invention;

[0031] Figure 3 This is a schematic diagram of the structure of the clamping component and the power component of the present invention;

[0032] Figure 4 This is a schematic diagram of the structure of the robotic arm of the present invention;

[0033] Figure 5 This is a diagram showing the structure and data flow of the control module of the present invention;

[0034] Figure 6 This is a flowchart of the closed-loop control of the parameter adjustment module of the present invention.

[0035] In the diagram: 1. Control panel; 2. Robotic arm; 21. Welding head; 22. Camera assembly; 3. Power assembly; 31. Motor; 32. Rotary shaft; 33. Synchronous belt; 34. Synchronous pulley; 35. Rotating rod; 36. Fixture; 4. Clamping assembly; 41. Gripper; 42. Mounting bracket; 43. Slider; 44. Guide rail; 45. Slide plate; 5. Control components. Detailed Implementation

[0036] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. In addition, the forms of the various structures described in the following embodiments are merely illustrative. The intelligent laser welding workstation for high-voltage cables involved in the present invention is not limited to the structures described in the following embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0037] Please see Figures 1-4 The high-voltage cable intelligent laser welding workstation shown includes an operating table 1, a robotic arm 2 mounted on the top of the operating table 1, a welding head 21 and a camera assembly 22 mounted on the robotic arm 2, and power components 3 mounted on both sides of the top of the operating table 1. Clamping components 4 are mounted on the power components 3. The high-voltage cable to be welded can be stably clamped by the clamping components 4 on both sides, providing precise positioning for subsequent welding.

[0038] The power assembly 3 includes a fixed frame 36 fixedly installed on the top of the operating table 1. The fixed frame 36 is generally rectangular and is used to support other components. A motor 31 is installed at the bottom of the fixed frame 36. A rotating shaft 32 is fixedly connected to the output end of the motor 31. Several rotating rods 35 are rotatably connected to the top of the fixed frame 36. In this embodiment, there are three rotating rods 35. Synchronous pulleys 34 are fixedly connected to the outer walls of the rotating shaft 32 and the rotating rods 35. A synchronous belt 33 is sleeved between the synchronous pulleys 34. When the motor 31 is started, the motor 31 drives the rotating shaft 32 to rotate. Through the transmission of the synchronous pulleys 34 and the synchronous belt 33, the rotating rods 35 and the synchronous pulleys 34 on them rotate synchronously, so that the synchronous belt 33 circulates on the fixed frame 36.

[0039] The clamping assembly 4 includes two sliding plates 45 fixedly mounted on the synchronous belt 33. The synchronous belt 33 has a ring structure, and the two sliding plates 45 are respectively fixedly connected to the opposite sides of the synchronous belt 33. Specifically, when the synchronous belt 33 is running, because it is ring-shaped, the sliding plate 45 fixed on one side of it and the sliding plate 45 fixed on the opposite side of it always move in opposite directions. For example, when the motor 31 drives the synchronous belt 33 to run clockwise, the front sliding plate 45 moves to the right, and the rear sliding plate 45 moves to the left, thereby bringing the two sliding plates 45 closer to each other; when the motor 31 drives the synchronous belt 33 to run counterclockwise, the front sliding plate 45 moves to the left, and the rear sliding plate 45 moves to the right, and the two sliding plates 45 move away from each other.

[0040] A mounting bracket 42 is fixedly connected to the top of the slide plate 45. A gripper 41 is fixedly connected to the top of the mounting bracket 42. The inner side of the gripper 41 may be provided with anti-slip texture or flexible pad to avoid damaging the cable surface. A guide rail 44 is fixedly connected to the top of the fixing bracket 36. The guide rail 44 is arranged along the length direction of the fixing bracket 36. A slider 43 is fixedly connected to the bottom of the mounting bracket 42. The bottom of the slider 43 is slidably connected to the guide rail 44. When the slide plate 45 moves, the mounting bracket 42 drives the slider 43 to slide along the guide rail 44, thereby ensuring the stability and straightness of the movement of the gripper 41.

[0041] A control component 5 is installed on the outside of the operating table 1. The control component 5 can be a PLC controller or an industrial computer, which is electrically connected to the robotic arm 2, welding head 21, camera assembly 22 and motor 31 respectively. The control component 5 is used to coordinate the work of each component to realize the automated welding process.

[0042] The control component 5 includes an image acquisition module, an image processing module, a trajectory planning module, a motion control module, a quality detection module, a parameter adjustment module, and a synchronization control module. The configuration of the control component modules and the data flow are as follows: Figure 5 As shown.

[0043] In this embodiment, the camera assembly 22 is used to acquire position images of the cable to be welded and real-time images during the welding process, and transmits the image signals to the control unit 5. The control unit 5 integrates an image processing module, which can analyze the images and identify the welding point position, weld shape, and welding quality of the cable. For example, before welding, the camera assembly 22 takes an image of the cable joint, and the control unit 5 calculates the precise welding coordinates through an image recognition algorithm, and then controls the robotic arm 2 to move the welding head 21 to the coordinates for welding. During the welding process, the camera assembly 22 can monitor the state of the molten pool and the weld formation in the welding area in real time. The control unit 5 dynamically adjusts the welding parameters according to the feedback information to ensure welding quality. After welding is completed, the camera assembly 22 can also take pictures of the weld appearance, and the control unit 5 can judge whether there are defects by image comparison, realizing online detection of welding quality.

[0044] Working Principle: First, the high-voltage cable to be welded is placed between the two clamping components 4. The controller 5 sends a command to start the motor 31. The motor 31 rotates forward, driving the rotating shaft 32 to rotate. Through the synchronous belt 33, it drives the two sliding plates 45 fixed on opposite sides to move closer together, thereby causing the grippers 41 on both sides to move smoothly towards the center along the guide rail 44, clamping the cable. After clamping, the controller 5 activates the camera component 22, which captures images of the cable welding area and sends the image data to the controller 5. The controller 5 obtains the precise welding position through image processing, and then controls the robotic arm 2 to move the welding head 21 to that position and start the welding head 21 to perform laser welding. During the welding process, the camera component 22 continuously captures images of the welding area. The controller 5 adjusts the welding parameters according to the real-time images to ensure welding quality. After welding is completed, the controller 5 controls the motor 31 to reverse, causing the two grippers 41 to move away from each other, releasing the cable, and the welded cable can be removed.

[0045] Furthermore, in the above technical solution, the image processing module is used to calculate the coordinates of the welding center point. The calculation process is as follows: the image acquisition module controls the camera component 22 to acquire images of the cable end; the image processing module performs grayscale conversion, filtering, and binarization preprocessing on the cable end image to extract the cable end face contour; based on the extracted end face contour, the end face circle is fitted by Hough transform to obtain the center coordinates (u, v); combined with the pre-calibrated camera intrinsic parameter matrix K and the transformation matrix between the robot arm 2 base coordinate system and the camera coordinate system... The pixel coordinates (u,v) are converted into three-dimensional spatial coordinates in the robot arm's base coordinate system. These coordinates are the coordinates of the welding center point.

[0046] It should be further noted that, in the preferred embodiment of this invention, the image acquisition module is the core hardware for realizing cable end-face image acquisition, and consists of an industrial camera, an industrial lens, a ring LED light source, and a camera mounting bracket. The selection and implementation details of each component are as follows:

[0047] Industrial camera: A 5-megapixel CMOS industrial camera with a global shutter is selected, model Basler acA2500-14gc, with a pixel size of 3.45μm×3.45μm, a resolution of 2592×1944, a maximum frame rate of 30fps, a GigE interface, and PoE power supply support. The camera selection criteria are: the global shutter can avoid motion blur caused by metal reflections on the cable end face, the 5-megapixel resolution meets the requirements for capturing the details of the end face contour, and the GigE interface is suitable for industrial field cabling scenarios.

[0048] Industrial lens: The matching lens is an 8mm fixed-focus manual aperture industrial lens, model reference Computar M0814-MP2, with an aperture adjustment range of F1.4-F16, a working distance of 100mm-200mm, and lens distortion <0.5%. The selection criteria are: fixed-focus lenses have better imaging stability than zoom lenses, an 8mm focal length can achieve complete imaging of the cable end face within a specified working distance, and low distortion ensures the accuracy of center positioning.

[0049] Ring LED light source: A white ring diffuse reflection LED light source is selected, model reference Keyence CA-DRW11, with an inner diameter of 40mm, an outer diameter of 60mm, an illumination angle of 45°, a brightness adjustment range of 0–10000 lux, and a constant current drive circuit design with ripple <1%; the selection criteria are: the diffuse reflection light source can eliminate the interference of metal reflection on the cable end face, the 45° illumination angle takes into account the brightness of the end face and the contour contrast, and the constant current drive avoids image acquisition errors caused by flicker.

[0050] Camera mounting bracket: The bracket is 3D printed or machined and is made of 6061 aluminum alloy. The bracket is fixed to the end effector of the robotic arm with bolts, with a fixing tolerance of ±0.05mm. After installation, the relative positions of the industrial camera, the ring LED light source and the laser welding head are fixed to ensure the spatial consistency of visual positioning and welding action.

[0051] Furthermore, the specific steps of image preprocessing include:

[0052] 1. Using the weighted calculation method for RGB to grayscale conversion, the formula is as follows:

[0053] Gray=0.299·R+0.587·G+0.114·B

[0054] The RGB color images captured by the industrial camera are converted pixel by pixel to output an 8-bit depth grayscale image, reducing the computational load of subsequent algorithms.

[0055] 2. A 3×3 rectangular window median filtering algorithm is adopted, which replaces the gray value of the center pixel with the median gray value of the 9 pixels in the window, and the filtering iteration count is 1.

[0056] III. Image binarization is achieved using Otsu's maximum inter-class variance method. This algorithm is an unsupervised adaptive thresholding method, and its execution logic is as follows:

[0057] Iterate through all pixels of the grayscale image and count the number of pixels with different grayscale values;

[0058] Calculate the inter-class variance between the foreground (cable end face) and background under different grayscale thresholds, and select the threshold corresponding to the maximum variance as the binarization threshold;

[0059] Set the value to 255 when the pixel value is greater than the threshold (foreground), and set it to 0 when the pixel value is less than or equal to the threshold (background).

[0060] Output a binarized image to fully extract the cable end face contour without the need for manual thresholding.

[0061] Furthermore, the specific steps for center fitting and coordinate transformation include:

[0062] 1. The standard Hough circle transform algorithm is used to detect the circular profile of the cable end face; the detection parameters are set as follows:

[0063] Minimum circle radius: 5mm (corresponding to image pixel value, converted according to camera working distance);

[0064] Maximum circle radius: 25mm (corresponding to image pixel value, converted according to camera working distance);

[0065] Accumulator threshold: 80 (dynamically adjusted based on image grayscale contrast to adapt to different cable end materials).

[0066] After the algorithm is executed, it outputs the center pixel coordinates (u,v) of the end face, with a precision of sub-pixel (0.01 pixel).

[0067] II. Hand-eye calibration and coordinate transformation

[0068] 1. Camera Intrinsic Parameter Calibration: A 12×9 checkerboard calibration board (square size 20mm×20mm) is used. 10-15 sets of images of the calibration board from different angles are acquired using an industrial camera. The camera intrinsic parameter matrix K is calculated using the Zhang Zhengyou calibration method, as shown in the following formula:

[0069]

[0070] in , The focal length of the camera (in pixels). , Principal point coordinates (in pixels);

[0071] Hand-eye calibration: A nine-point hand-eye calibration method was adopted. Using the end effector of the robotic arm as the reference, nine sets of camera-calibration board images were acquired in different poses, and the homogeneous transformation matrix from the camera coordinate system to the robotic arm base coordinate system was calculated. The repeatability after calibration is ±0.05mm.

[0072] Coordinate transformation: Convert pixel coordinates (u,v) into three-dimensional spatial coordinates in the robot arm's base coordinate system. Execute the formula: ;

[0073] Output the three-dimensional coordinates (X,Y,Z) of the welding center point, with a system positioning error of ≤±0.1mm.

[0074] Furthermore, in the above technical solution, the trajectory planning module is used to generate the motion trajectory of the robotic arm 2. The specific process is as follows: Based on the welding center point coordinates output by the image processing module, combined with the preset welding start point offset, welding path length and welding overlap width, the key path points of the welding trajectory are generated in three-dimensional space; the key path points are smoothed by polynomial interpolation or B-spline curve algorithm to generate the continuous motion trajectory of the end effector (i.e., welding head (21)) of the robotic arm 2, which contains position and attitude information.

[0075] It should be further explained that in the preferred technical solution of this invention, the trajectory planning module generates a smooth welding path based on the welding center point and outputs a continuous pose sequence. The specific process is as follows:

[0076] 1. Keypoint Generation

[0077] Using the welding center point P0(X0,Y0,Z0) as the reference:

[0078] Offset 5mm forward along the cable's central axis (positive Z-axis direction) to determine the welding start point P. s (X s ,Y s Z s );

[0079] Extend 30mm backward along the cable's central axis (negative Z-axis direction) to determine the welding termination point P. e (X e ,Y e Z e );

[0080] Set the overlap in the width direction of the welding path to 2mm to ensure that the weld covers the entire processed area of ​​the cable end face;

[0081] Each key point contains position information (X,Y,Z) and attitude information. The attitude is represented by ZYX Euler angles. The welding head is required to always be perpendicular to the cable end face, and the attitude deviation is ≤0.1°.

[0082] 2. Smoothed cubic uniform B-spline trajectory

[0083] Select 6-8 key points along the path, including the welding start point, welding end point, and intermediate transition points (the transition points are set according to the curvature of the cable end face, with a spacing of 5mm); use a cubic uniform B-spline curve for interpolation, and the interpolation formula is as follows:

[0084]

[0085] Where u∈[0,1], is the normalization parameter, and N i,3 (u) is a cubic B-spline basis function, P i+k These are the key points of the selected path;

[0086] After the algorithm is executed, it outputs a pose sequence of interpolation points every 10ms. Each interpolation point contains position and attitude information to ensure that the trajectory is continuous in position, velocity and acceleration.

[0087] The maximum permissible error of the trajectory is ≤ ±0.1mm, which meets the accuracy requirements of laser welding trajectory.

[0088] Furthermore, in the above technical solution, the motion control module is used to control the robotic arm 2 to accurately track the welding trajectory. The specific process is as follows: The motion control module receives the continuous motion trajectory generated by the trajectory planning module, and converts the pose sequence of the end effector into the angle sequence of each joint of the robotic arm 2 in real time through inverse kinematics solution of the robotic arm 2; the servo control algorithm drives the motors of each joint of the robotic arm 2 so that the welding head 21 moves strictly according to the predetermined trajectory, and triggers the welding head 21 to emit light when it reaches the welding start point to perform laser welding.

[0089] It should be further noted that, in the preferred embodiment of this invention, the motion control module is the core hardware for realizing the motion control of the robotic arm, and consists of a six-axis industrial robotic arm, a motion controller, and a servo driver. The selection and implementation details of each component are as follows:

[0090] Six-axis industrial robotic arm: The ABB IRB 120 six-axis industrial robotic arm is selected, with a load capacity of 3kg, a repeatability of ±0.02mm, and a working radius of 580mm. The selection criteria are: the six-axis structure meets the requirements of multi-degree-of-freedom motion, the ±0.02mm repeatability is suitable for the accuracy requirements of welding trajectory, and the 580mm working radius covers the processing range of cable end faces.

[0091] Motion controller: The ABB IRC5 compact motion controller is selected, with a built-in DSP main control chip and FPGA coprocessor, and a control cycle of 1ms. The controller supports functions such as inverse kinematics solving, trajectory interpolation, and servo closed-loop control.

[0092] Servo drive: The ABB MotiFlex RC servo drive is selected to match the joint motor of the robotic arm. It supports dual closed-loop control of position loop + speed loop and has a response frequency ≥1kHz. The selection criteria are: high response frequency ensures real-time trajectory tracking and dual closed-loop control improves motion accuracy.

[0093] Furthermore, the inverse kinematics problem of the six-axis industrial robot is solved analytically. The solution logic is as follows:

[0094] Based on the end effector pose sequence (X,Y,Z,α,β,γ) output by the trajectory planning module, kinematic equations are established using the improved DH method to achieve high-precision inverse kinematics solution.

[0095] 1. Establishment of kinematic equations based on the improved DH method

[0096] 1.1 ABB IRB 120 Official Improved DH Parameter Table

[0097] The kinematic model was established using the improved DH (MDH) parameters officially released by ABB. The parameters are in units of length (m) and angle (rad); θ1~θ6 are the rotation angles of the six-axis joints (variables for kinematic solution).

[0098] Table 1. ABB IRB 120 Official Improved DH Parameter Table

[0099] Link i <![CDATA[α i-1 ]]> <![CDATA[a i-1 ]]> <![CDATA[d i ]]> <![CDATA[Joint variable θ i > 1 0 0 0.290 <![CDATA[θ1]]> 2 π / 2 0.070 0 <![CDATA[θ2]]> 3 0 0.300 0 <![CDATA[θ3]]> 4 π / 2 0.070 0.300 <![CDATA[θ4]]> 5 −π / 2 0 0 <![CDATA[θ5]]> 6 π / 2 0 0.080 <![CDATA[θ6]]>

[0100] 1.2 Improved DH single-link homogeneous transformation matrix

[0101] The improved DH method single-link coordinate system transformation formula is as follows:

[0102]

[0103] Expand into standard homogeneous matrix form:

[0104]

[0105] 1.3 Overall Kinematic Transformation Matrix of the Robotic Arm

[0106] The total homogeneous transformation matrix from the base coordinate system to the end effector is the product of the transformation matrices of each link:

[0107]

[0108] Overall matrix form:

[0109]

[0110] Where: [X,Y,Z] T Let [n,o,a] be the end position, and [n,o,a] be the end pose rotation matrix.

[0111] 1.4 Construction of Homogeneous Endpoint Pose Matrix

[0112] Based on the ZYX Euler angles (α,β,γ) output by the trajectory planning module, construct the target homogeneous matrix of the end effector. :

[0113]

[0114] In the formula: c = cos, s = sin.

[0115] 1.5 Core Kinematic Equations

[0116] By combining the total transformation matrix and the target pose matrix, the core kinematic equations of the robotic arm can be established:

[0117]

[0118] With corresponding elements on both sides of the equation being equal, we obtain 12 independent algebraic equations, which provide the only mathematical basis for solving inverse kinematics.

[0119] 2. Solving inverse kinematics

[0120] The motion controller uses an analytical method to solve the above kinematic equations. Based on the pose sequence (X,Y,Z,α,β,γ) output by trajectory planning, the six-axis joint angles (θ1,θ2,θ3,θ4,θ5,θ6) are calculated. In the case of multiple solutions, the optimal solution with the minimum joint motion and no singular configuration is selected.

[0121] Solving the kinematic equations yields the angles of each joint (θ1, θ2, θ3, θ4, θ5, θ6). In the case of multiple solutions, the optimal solution with the minimum joint movement distance and no singular configuration is selected.

[0122] 3. Servo closed-loop control

[0123] Each joint motor is controlled by a dual closed-loop PID controller consisting of a position loop and a speed loop, with the following parameter settings:

[0124] Position ring proportionality coefficient K p =100 (value range 80-120, dynamically adjusted according to load);

[0125] Speed ​​ring proportional coefficient Kp =40 (value range 30-50, dynamically adjusted according to load);

[0126] Velocity loop integral coefficient K i =3 (value range 2-5, dynamically adjusted according to load);

[0127] With a control cycle of 1ms, the servo driver provides real-time feedback on joint position and speed information, dynamically adjusting the output torque to ensure that the dynamic tracking error between the robotic arm's end effector trajectory and the theoretical trajectory is ≤±0.2mm.

[0128] 4. Welding trigger logic

[0129] The motion controller monitors the position of the robotic arm's end effector in real time, and when the welding head reaches the welding start point P... s When the position error is less than 0.1mm, a 5V TTL high-level signal is triggered;

[0130] The signal output response time is <1μs, and it is transmitted to the fiber laser controller through a shielded cable to trigger the fiber laser to emit light.

[0131] Trigger delay <5ms to ensure accurate soldering initiation position.

[0132] Furthermore, in the above technical solution, the parameter adjustment module is used to dynamically adjust welding parameters during the welding process, such as... Figure 6 As shown, the specific process is as follows: During the welding process, the camera component 22 acquires real-time images of the molten pool in the welding area; the parameter adjustment module extracts the width and length features of the molten pool based on the molten pool image, and compares the extracted molten pool features with the preset standard molten pool feature range; if the molten pool width exceeds the upper limit of the standard range, the laser power of the welding head 21 is reduced or the welding speed is increased by a preset step size; if the molten pool width is lower than the lower limit of the standard range, the laser power is increased or the welding speed is decreased by a preset step size.

[0133] It should be further noted that, in the preferred embodiment of the present invention, the closed-loop control of the parameter adjustment module is implemented as follows:

[0134] Molten pool image acquisition: During the welding process, the camera component acquires real-time images of the molten pool in the welding area at a frame rate of 100fps. The image resolution is 1280×1024, and the images are transmitted to the parameter adjustment module via the GigE interface.

[0135] Molten pool feature extraction: Image processing algorithms are used to extract the width and length features of the molten pool.

[0136] The molten pool image is preprocessed by grayscale conversion and median filtering;

[0137] The edge contour of the molten pool is extracted using the Canny edge detection algorithm;

[0138] Calculate the minimum bounding rectangle of the molten pool profile to obtain the molten pool width W and length L;

[0139] The average feature value of 5 consecutive frames is taken as the current feature value of the melt pool to eliminate the influence of single-frame noise.

[0140] Dynamic parameter adjustment strategy: The standard weld pool width range is set to [2.5mm, 3.5mm] (this can be preset and adjusted according to different cable specifications). The parameter adjustment module compares the current weld pool width with the standard range in real time.

[0141] If the weld pool width is greater than 3.5mm (exceeding the upper limit), reduce the laser power by 5% or increase the welding speed by 5% according to the preset step size;

[0142] If the weld pool width is less than 2.5mm (below the lower limit), increase the laser power by 5% or decrease the welding speed by 5% according to the preset step size;

[0143] If the molten pool width is within the standard range, keep the current parameters unchanged.

[0144] The adjustment step size can be dynamically set according to the welding material and cable specifications (e.g., power step size 1%-5%, speed step size 2%-10%), with an adjustment cycle of 50ms (adjusted once every 5 frames of images) to ensure that the molten pool state quickly converges to a stable range.

[0145] Furthermore, in the above technical solution, the quality inspection module is used to evaluate the weld quality after welding is completed. The specific process is as follows: After welding is completed, the camera component 22 acquires a complete image of the weld; the quality inspection module performs image segmentation on the weld image, extracts the weld area, and calculates the weld width uniformity and excess height; the calculated weld width uniformity and excess height are compared with a preset quality standard threshold. If any indicator exceeds the threshold range, the weld point is determined to be unqualified, and an alarm signal is generated.

[0146] It should be further explained that the online detection implementation method of the quality inspection module is as follows:

[0147] 1. Weld seam image acquisition: After welding is completed, the robotic arm drives the camera assembly to scan along the weld seam path at a constant speed of 5mm / s, continuously acquiring 10-15 weld seam images (image overlap rate 20%). An image stitching algorithm is then used to synthesize multiple images into a complete weld seam image.

[0148] 2. Weld Seam Region Extraction: A deep learning semantic segmentation model (U-Net architecture) is used to perform pixel-level segmentation of the weld seam images. The model training dataset contains 2000 labeled weld seam images. The model outputs a binarized mask image to accurately extract the weld seam region.

[0149] 3. Weld feature calculation:

[0150] Width uniformity: Take 50 cross-sections at equal intervals along the weld direction, calculate the weld width of each cross-section, and obtain the average width μ. w and standard deviation σ w If σ w If the width is less than 0.2mm, the width uniformity is considered acceptable; otherwise, it is unacceptable.

[0151] Residual height: The three-dimensional point cloud of the weld surface is obtained using structured light 3D reconstruction technology, and the height difference H between the highest point of the weld surface and the surface of the cable substrate is calculated. If 0.5mm≤H≤1.0mm, the residual height is considered qualified; otherwise, it is unqualified.

[0152] 4. Quality Judgment and Alarm: The calculated width uniformity and excess height are compared with preset quality standard thresholds. If any indicator exceeds the threshold range, the system records the defect location (saves the defect image and coordinates), triggers an audible and visual alarm (buzzer + red warning light), and displays the defect information on the HMI interface. For qualified welds, the inspection data is recorded, providing data support for product quality traceability.

[0153] Furthermore, in the above technical solution, the synchronization control module is used to coordinate with the production line cycle. The specific process is as follows: The synchronization control module receives the cable arrival start / stop signal sent by the upstream production line in real time; upon receiving the arrival signal, the synchronization control module first sends a forward rotation command to the motor 31 to drive the clamping component 4 to close and clamp the cable; after clamping, it sends a start command to the image acquisition module and subsequent modules to trigger the automatic welding process; after the welding process is completed, the synchronization control module sends a reverse rotation command to the motor 31 to drive the clamping component 4 to release, and sends a welding completion signal to the production line, allowing the cable to enter the next station.

[0154] It should be further noted that, in the preferred embodiment of the present invention, the collaborative workflow of the synchronization control module is as follows:

[0155] Signal interface configuration: A Siemens S7-1200 PLC is used as the core controller of the synchronous control module, communicating with the production line main control system via Profinet industrial Ethernet. Hard-wired I / O signals serve as a backup communication method, including:

[0156] Inputs: Cable arrival signal (24V DC), emergency stop signal;

[0157] Output: Clamping complete signal, welding complete signal, fault alarm signal.

[0158] Fully automated collaborative process:

[0159] Cable arrival detection: After the upstream production line delivers the cable to the designated location, the proximity switch triggers a cable arrival signal (24V high level), and the synchronous control module receives the signal.

[0160] Automatic clamping: The synchronous control module sends a forward rotation command to the motor 31, driving the clamping assembly 4 to close and clamp the cable. Clamping is completed as feedback is received from the pressure sensor inside the gripper (with a set clamping force threshold of 50N), and the feedback signal triggers the next step.

[0161] Automatic welding: After clamping in place, the synchronous control module sends a start command to the image acquisition module, which in turn triggers the image processing, trajectory planning, and motion control modules to complete the automatic welding process.

[0162] Automatic release: After the welding process is completed, the synchronous control module sends a reverse command to the motor 31, driving the clamping assembly 4 to release the cable. The release is confirmed by a limit switch.

[0163] Signal feedback: The synchronous control module sends a welding completion signal (24V high level, pulse width 200ms) to the production line, allowing the cable to enter the next station.

[0164] Abnormal handling mechanism: If the clamping force exceeds the safety threshold (>80N) or the clamping time exceeds 5s during the clamping process, the system will immediately stop the operation and alarm; if a fault occurs during the welding process (such as laser abnormality or robotic arm out of tolerance), the synchronous control module will pause the process and send a fault signal to the production line.

[0165] Furthermore, in the above technical solution, the synchronous clamping process of the clamping component 4 is as follows: when the motor 31 rotates forward, it drives the rotating shaft 32 to rotate, which drives the synchronous belt 33 to perform cyclic motion through the synchronous wheel 34; since the synchronous belt 33 is a ring structure, the sliding plate 45 fixed on its upper side and the sliding plate 45 fixed on its lower side move in opposite directions, driving the two mounting brackets 42 and their clamping claws 41 to move closer to each other along the guide rail 44, thereby achieving centering and clamping the cable; when the motor 31 rotates in reverse, the two clamping claws 41 move away from each other, releasing the cable.

[0166] Furthermore, in the above technical solution, the inner side of the gripper 41 is provided with anti-slip texture or flexible pad to protect the cable surface and enhance clamping stability.

[0167] It should be further explained that the mechanical structure of the clamping assembly and the clamping force control method are as follows:

[0168] Synchronous clamping mechanism: The power assembly 3 includes a fixed frame 36 fixedly mounted on the top of the operating table 1. A motor 31 (model reference: Mitsubishi HG-KN73J-S100, power 750W) is mounted on the bottom of the fixed frame 36. The output end of the motor 31 is fixedly connected to the rotating shaft 32. Three rotating rods 35 are rotatably connected to the top of the fixed frame 36. Synchronous pulleys 34 are fixedly connected to the outer walls of the rotating shaft 32 and the rotating rods 35. A synchronous belt 33 (tooth pitch 5mm, width 15mm) is sleeved between the synchronous pulleys 34.

[0169] The clamping assembly 4 includes two slide plates 45 fixedly mounted on the timing belt 33. The top of the slide plates 45 is fixedly connected to the mounting bracket 42, and the top of the mounting bracket 42 is fixedly connected to the gripper 41. The top of the mounting bracket 36 is fixedly connected to the guide rail 44 (HIWIN EGH15CA, 500mm in length), and the bottom of the mounting bracket 42 is fixedly connected to the slider 43. The bottom of the slider 43 is slidably connected to the guide rail 44.

[0170] When motor 31 rotates forward, it drives shaft 32 to rotate, which in turn drives synchronous belt 33 to circulate via synchronous pulley 34. Since synchronous belt 33 is a ring structure, the sliding plate 45 fixed to its upper side and the sliding plate 45 fixed to its lower side move in opposite directions, causing the two mounting brackets 42 and their grippers 41 to move closer together along guide rail 44, thus centering and clamping the cable. When motor 31 rotates in reverse, the two grippers 41 move away from each other, releasing the cable.

[0171] Clamping force control: A flexible rubber pad (3mm thick, Shore A hardness 60A) is provided on the inner side of the clamping jaws 41 to protect the cable surface and enhance clamping stability. The clamping force is indirectly controlled by the output torque of the motor 31, with a set current threshold corresponding to a clamping force of 50N±5N. The controller monitors the motor current in real time and automatically cuts off the motor power when the current reaches the threshold to prevent excessive clamping and damage to the cable.

[0172] Application scope: The clamping stroke ranges from 0-100mm, accommodating cable diameters from 10mm to 50mm. When changing to different cable specifications, the clamping force parameters can be preset via the HMI interface, and the system will automatically adjust the current threshold without requiring replacement of mechanical parts.

[0173] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0174] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A high-voltage cable intelligent laser welding workstation, comprising an operating table (1), characterized in that: The top of the operating table (1) is equipped with a robotic arm (2), on which a welding head (21) and a camera assembly (22) are installed. Power assemblies (3) are installed on both sides of the top of the operating table (1), and clamping assemblies (4) are installed on the power assemblies (3). The power assembly (3) includes a fixed frame (36) fixedly installed on the top of the operating table (1). A motor (31) is installed at the bottom of the fixed frame (36). A rotating shaft (32) is fixedly connected to the output end of the motor (31). Several rotating rods (35) are rotatably connected to the top of the fixed frame (36). Synchronous pulleys (34) are fixedly connected to the outer walls of the rotating shaft (32) and the rotating rods (35). A synchronous belt (33) is sleeved between the synchronous pulleys (34). The clamping assembly (4) includes two slide plates (45) fixedly mounted on the timing belt (33). The top of the slide plate (45) is fixedly connected to a mounting bracket (42). The top of the mounting bracket (42) is fixedly connected to a gripper (41). The top of the mounting bracket (36) is fixedly connected to a guide rail (44). The bottom of the mounting bracket (42) is fixedly connected to a slider (43). The bottom of the slider (43) is slidably connected to the guide rail (44). A control component (5) is installed on the outside of the operating table (1), and the control component (5) is electrically connected to the robotic arm (2), welding head (21), camera assembly (22) and motor (31). The control element (5) includes: Image acquisition module; used to control the camera assembly (22) to acquire images from the cable end; Image processing module: used to calculate the coordinates of the welding center point; Trajectory planning module: used to generate the motion trajectory of the robotic arm (2); Motion control module: used to control the robotic arm (2) to accurately track the welding trajectory; Quality inspection module: Used to assess weld quality after welding is completed; Parameter adjustment module: Used to dynamically adjust welding parameters during the welding process; Synchronization control module: Used to coordinate with the production line cycle time.

2. The intelligent laser welding workstation for high-voltage cables according to claim 1, characterized in that: The calculation process of the image processing module for calculating the coordinates of the welding center point is as follows: The image processing module performs grayscale, filtering and binarization preprocessing on the cable end image to extract the cable end face contour; based on the extracted end face contour, the end face circle is fitted by Hough transform to obtain the center coordinates (u, v); combined with the pre-calibrated camera intrinsic parameter matrix K and the transformation matrix between the robot arm (2) base coordinate system and the camera coordinate system. The pixel coordinates (u,v) are converted into three-dimensional spatial coordinates in the robot arm's base coordinate system. These coordinates are the coordinates of the welding center point.

3. The intelligent laser welding workstation for high-voltage cables according to claim 1, characterized in that: The specific process of the trajectory planning module generating the motion trajectory of the robotic arm (2) is as follows: Based on the welding center point coordinates output by the image processing module, combined with the preset welding start point offset, welding path length and welding overlap width, the key path points of the welding trajectory are generated in three-dimensional space; the key path points are smoothed by polynomial interpolation or B-spline curve algorithm to generate the continuous motion trajectory of the end effector (i.e. welding head (21)) of the robotic arm (2), which contains position and attitude information.

4. The intelligent laser welding workstation for high-voltage cables according to claim 1, characterized in that: The specific process of the motion control module controlling the robotic arm (2) to accurately track the welding trajectory is as follows: The motion control module receives the continuous motion trajectory generated by the trajectory planning module, and converts the pose sequence of the end effector into the angle sequence of each joint of the robotic arm (2) in real time through the inverse kinematics solution of the robotic arm (2); the servo control algorithm drives the motors of each joint of the robotic arm (2) so that the welding head (21) moves strictly according to the predetermined trajectory, and triggers the welding head (21) to emit light when it reaches the welding start point to perform laser welding.

5. The intelligent laser welding workstation for high-voltage cables according to claim 1, characterized in that: The specific process of the parameter adjustment module dynamically adjusting the welding parameters during the welding process is as follows: During the welding process, the camera component (22) collects the molten pool image of the welding area in real time; the parameter adjustment module extracts the width and length features of the molten pool based on the molten pool image, and compares the extracted molten pool features with the preset standard molten pool feature range; if the molten pool width exceeds the upper limit of the standard range, the laser power of the welding head (21) is reduced or the welding speed is increased by a preset step size; if the molten pool width is lower than the lower limit of the standard range, the laser power is increased or the welding speed is reduced by a preset step size.

6. The intelligent laser welding workstation for high-voltage cables according to claim 1, characterized in that: The specific process of the quality inspection module in evaluating the weld quality after welding is as follows: After welding is completed, the camera component (22) acquires a complete image of the weld; the quality inspection module performs image segmentation on the weld image, extracts the weld area, and calculates the weld width uniformity and excess height; the calculated weld width uniformity and excess height are compared with the preset quality standard threshold. If any indicator exceeds the threshold range, the weld point is determined to be unqualified and an alarm signal is generated.

7. The intelligent laser welding workstation for high-voltage cables according to claim 1, characterized in that: The specific process by which the synchronous control module works in coordination with the production line cycle is as follows: The synchronous control module receives the cable arrival start / stop signal sent by the upstream production line in real time; when the arrival signal is received, the synchronous control module first sends a forward rotation command to the motor (31) to drive the clamping component (4) to close and clamp the cable; after clamping in place, it sends a start command to the image acquisition module and subsequent modules to trigger the automatic welding process; after the welding process is completed, the synchronous control module sends a reverse rotation command to the motor (31) to drive the clamping component (4) to release, and sends a welding completion signal to the production line to allow the cable to enter the next station.

8. The intelligent laser welding workstation for high-voltage cables according to claim 1, characterized in that: The synchronous clamping process of the clamping assembly (4) is as follows: When the motor (31) rotates forward, it drives the rotating shaft (32) to rotate, and drives the synchronous belt (33) to perform cyclic motion through the synchronous wheel (34); Since the synchronous belt (33) is a ring structure, the sliding plate (45) fixed on its upper side moves in the opposite direction to the sliding plate (45) fixed on its lower side, which drives the two mounting brackets (42) and their clamping claws (41) to move closer to each other along the guide rail (44) to achieve centering and clamping of the cable; When the motor (31) reverses, the two clamping claws (41) move away from each other and release the cable.

9. The intelligent laser welding workstation for high-voltage cables according to claim 1, characterized in that: The inner side of the gripper (41) is provided with anti-slip texture or flexible pad to protect the cable surface and enhance clamping stability.

Images