Regulation and control method for welding and positioning process of steel reinforcement framework of electric pole

Abstract

The invention discloses an electric pole steel reinforcement framework welding positioning process regulation and control method, and relates to the technical field of electric pole steel reinforcement framework intelligent welding positioning. The nodes are synchronously locked in the dynamic adjustment process; dynamically optimizing a welding process parameter strategy; collaborative closed-loop regulation and control of positioning welding collaborative parameters are completed; and evaluating the geometric dimension after welding. High-precision positioning and self-adaptive control in the whole welding process of the electric pole steel reinforcement framework are achieved, it is ensured that the framework structure meets the concrete pouring requirement, welding consistency, size stability and the automation level are greatly improved, the defects that a traditional method depends on manpower, clamps are simple, and dynamic feedback is lacked are overcome, and the production efficiency is improved. And a reliable technical support is provided for intelligent manufacturing of the electric pole.

Description

Technical Field

[0001] This invention relates to the field of intelligent welding and positioning technology for the reinforcing steel cage of utility poles, and in particular to a method for controlling the welding and positioning process of the reinforcing steel cage of utility poles. Background Technology

[0002] With the continuous advancement of infrastructure construction, power transmission line projects in the fields of electricity and communications have placed higher demands on the quality and production efficiency of concrete poles. As a key component supporting overhead lines, the welding precision and structural stability of the internal steel reinforcement cage of the pole directly affect its load-bearing capacity and service life. In the context of the increasing popularity of intelligent manufacturing and green construction concepts, the automated welding process of steel reinforcement cages is rapidly developing towards high precision, high efficiency, and low energy consumption, which also places more stringent requirements on positioning control technology in the welding process.

[0003] However, existing welding processes for the steel reinforcement cages of utility poles generally suffer from problems such as insufficient positioning accuracy, difficulty in controlling welding deformation, and poor weld consistency. Traditional methods often rely on manual adjustment or simple mechanical clamps for steel reinforcement positioning. During the welding process, the steel reinforcement is easily affected by thermal stress, leading to displacement and deviations in the geometric dimensions of the cage, which in turn affects the quality of subsequent concrete pouring. At the same time, the lack of a coordinated control mechanism for welding parameters and positioning status makes it difficult to achieve dynamic feedback and adaptive adjustment throughout the entire process, thus restricting further improvements in welding quality and production efficiency. Summary of the Invention

[0004] In view of the problems existing in the existing methods for controlling the welding and positioning process of the steel reinforcement cage of utility poles, this invention is proposed.

[0005] Therefore, the problems to be solved by the present invention are: insufficient positioning accuracy, difficulty in controlling geometric deviations caused by thermal deformation, poor weld consistency, and lack of a coordinated control mechanism for welding parameters and positioning status in the existing pole reinforcement cage welding process, which affect the structural quality and production efficiency of the pole.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, embodiments of the present invention provide a method for controlling the welding positioning process of a pole reinforcement cage, which includes: using a visual recognition system to perform three-dimensional spatial positioning of the initial assembly state of the reinforcement cage, obtaining the position coordinates of the reinforcement cage components, and comparing them with a preset standard data template to generate positioning deviation data; Based on positioning deviation data, the multi-degree-of-freedom servo clamping mechanism is driven to dynamically adjust the steel reinforcement cage, complete the alignment before welding, and lock the nodes synchronously during the dynamic adjustment process to avoid displacement caused by thermal deformation. During the welding execution phase, multi-dimensional parameters of the welding process are collected, and the welding process parameter strategy is dynamically optimized in combination with the characteristics of the steel bar itself. Establish a thermo-mechanical coupling feedback mechanism during the welding process, monitor the welding area, and feed the monitoring results back to the execution and control unit; After the welding execution phase is completed, a retest and verification process is triggered to evaluate the geometry of the welded components.

[0007] As a preferred embodiment of the welding positioning process control method for the pole reinforcement cage described in this invention, the initial assembly state of the reinforcement cage is positioned in three-dimensional space by means of a visual recognition system. This includes deploying a multimodal visual perception unit of the visual recognition system to form a 360° no-dead-angle observation layout around the reinforcement cage, and acquiring multiple frames of deformed stripe images by projecting coded phase stripes onto the surface of the reinforcement cage component through the visual recognition system. The point cloud data of the steel skeleton component is reconstructed using a 3D reconstruction fusion algorithm. The geometric entities of the steel skeleton component are separated by a point cloud data segmentation method based on curvature features, and the 3D spatial position and pose of each steel skeleton component in the global coordinate system are calculated.

[0008] As a preferred embodiment of the pole reinforcement cage welding positioning process control method of the present invention, the method of obtaining the position coordinates of the reinforcement cage component includes: after completing the three-dimensional spatial position and attitude calculation of the reinforcement cage component, extracting the closed ring contour of the reinforcement cage component based on the point cloud skeletonization algorithm, performing topological association marking on each intersection node of the closed ring contour, mapping the feature nodes of each reinforcement cage component to a unified global coordinate system, and generating a position coordinate dataset. The comparison with the preset standard data template includes calling the preset standard data template corresponding to the current pole model from the process database. The preset standard data template stores the theoretical reference data of all steel reinforcement skeleton components under ideal conditions in the same format. By establishing a component-level matching index, the location coordinate dataset is mapped one-to-one with the corresponding component items in the preset standard data template, and the weighted least squares method is used to perform rigid body registration of the steel reinforcement cage. The generation of positioning deviation data includes calculating the residual vector between the actual position coordinates of each steel reinforcement skeleton component and the theoretical coordinates in the preset standard data template in the local coordinate system after rigid body registration is completed. Combining the component weight attributes of the steel reinforcement skeleton components, a differential sensitivity coefficient is assigned to the residual vector. The generated positioning deviation data is then fused and output to the servo control module in the form of a positioning deviation data package.

[0009] As a preferred embodiment of the pole reinforcement cage welding positioning process control method of the present invention, the dynamic adjustment of the reinforcement cage includes receiving the positioning deviation data packet output by the visual recognition system and then analyzing the pose weight composite parameters of the feature nodes of each reinforcement cage component. Based on a preset motion mapping rule library, the residual vector of each feature node is decomposed into multi-axis linkage commands for the corresponding clamping unit, including translation compensation in the X / Y / Z directions and rotation angle around the axis. The main control unit sends control signals to each servo execution module in the multi-degree-of-freedom servo clamping mechanism in a time-sequence manner according to a priority scheduling strategy, so as to apply dynamic adjustment to the steel skeleton component with sub-millimeter precision, converge the actual position to the range allowed by the theoretical reference data, and complete the alignment before welding. The synchronous locking node during dynamic adjustment includes monitoring the clamping status feedback parameters of each clamping point while the multi-degree-of-freedom servo clamping mechanism performs reverse dynamic adjustment. When the position error of a certain feature node converges to within a preset threshold, the node locking execution unit is triggered to perform rigid locking on the feature node. The locking action is dynamically allocated by the embedded PLC according to the component weight attribute. The steel reinforcement cage includes main bars and stirrups; force locking is used at the intersection of the main bars and flexible limiting locking is used at the connection points of the stirrups; all locking operations are completed before the end of the dynamic adjustment process, forming a spatial constraint network to suppress the relative displacement of the steel reinforcement cage due to thermal expansion or stress release during the subsequent welding heat input stage.

[0010] As a preferred embodiment of the pole reinforcement cage welding positioning process control method of the present invention, the acquisition of multi-dimensional parameters of the welding process includes, during the welding execution stage, acquiring multi-dimensional parameters of the welding process through a sensor array integrated into the welding torch body, forming a multi-dimensional parameter stream, and continuously inputting it to the process control engine at a millisecond-level sampling frequency; The dynamic optimization strategy for welding process parameters includes using a process control engine to retrieve the steel body characteristics corresponding to the steel skeleton component to be welded from the process database, using the steel body characteristics and the input welding process multi-dimensional parameters as input variables, and loading them into a pre-trained lightweight neural network architecture. This lightweight neural network architecture is trained based on a historical high-quality weld point dataset, mapping out excellent welding process parameter strategies under different working conditions, and using the main control unit to dynamically adjust the welding machine output parameters and the robot trajectory interpolation instructions during the welding process.

[0011] As a preferred embodiment of the pole reinforcement cage welding positioning process control method of the present invention, the monitoring of the welding area includes, during the welding process, the synchronous operation of a thermo-coupling feedback mechanism. The thermo-coupling feedback mechanism integrates virtual and real monitoring data. First, a high frame rate infrared thermal imager is used to continuously collect the temperature field evolution sequence of the welding area to obtain the transient heat distribution of the weld point and the surface of the adjacent reinforcement cage components. Second, the thermo-response signal is captured, and the thermo-evolution state inside the thermo-response signal is deduced online based on the simplified finite element architecture to generate a comprehensive state vector. The step of feeding the monitoring results back to the execution and control unit includes encapsulating the comprehensive state vector into a feedback data packet and transmitting it to the execution and control unit with low latency via an industrial real-time communication bus. The collaborative closed-loop control of the positioning welding collaborative parameters includes integrating the hierarchical control instructions of the execution control unit with the current welding process parameter strategy, dynamically correcting the positioning welding collaborative parameters, and sending the dynamically corrected positioning welding collaborative parameters to the multi-degree-of-freedom servo clamping mechanism.

[0012] As a preferred embodiment of the pole reinforcement cage welding positioning process control method of the present invention, the trigger retest verification process includes activating the retest trigger module after the main control unit receives the welding process completion status signal; The retest trigger module is used to automatically generate retest task instructions based on the positioning welding coordination parameters, send a start command to the multimodal vision perception unit, and simultaneously release the clamping constraints on non-critical feature nodes in the node locking execution unit, so that the steel reinforcement skeleton is in a low-stress free state; at the same time, the retest trigger module sends a reset command to the servo control module to return the multi-degree-of-freedom servo clamping mechanism to the preset safe posture. The evaluation of the post-weld geometry includes: a multimodal visual perception unit performing a secondary 3D scan of the post-weld rebar skeleton using a visual scanning strategy consistent with the initial assembly and positioning stage to obtain complete point cloud data; reconstructing the point cloud data of the post-weld rebar skeleton components using a 3D reconstruction fusion algorithm; separating the geometric entities of each rebar skeleton component using a point cloud data segmentation method based on curvature features; extracting the coordinates of feature nodes of each post-weld rebar skeleton component using a point cloud skeletonization algorithm; matching the component-level data with theoretical benchmark data in a preset standard data template corresponding to the same pole model; introducing a welding residual deformation compensation factor in the evaluation process; dynamically weighting the tolerance thresholds of different feature nodes in combination with component weight attributes; calculating the geometric continuity index of different feature nodes; and generating the geometric dimension evaluation result.

[0013] Secondly, embodiments of the present invention provide a welding positioning process control system for a utility pole steel reinforcement cage, comprising: a visual positioning module, which uses a visual recognition system to perform three-dimensional spatial positioning of the initial assembly state of the steel reinforcement cage, obtains the position coordinates of the steel reinforcement cage components, and compares them with a preset standard data template to generate positioning deviation data; The servo control module, based on positioning deviation data, drives the multi-degree-of-freedom servo clamping mechanism to dynamically adjust the steel reinforcement cage, complete the alignment before welding, and lock the nodes synchronously during the dynamic adjustment process to avoid displacement caused by thermal deformation. The process optimization module collects multi-dimensional parameters of the welding process during the welding execution phase and dynamically optimizes the welding process parameter strategy in combination with the characteristics of the steel bar itself. The thermal feedback module establishes a thermal coupling feedback mechanism during the welding process, monitors the welding area, and feeds the monitoring results back to the execution and control unit to complete the coordinated closed-loop control of the positioning welding parameters. The retesting and evaluation module triggers the retesting and verification process after the welding execution phase is completed, and evaluates the geometric dimensions after welding.

[0014] Thirdly, embodiments of the present invention provide a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement any step of the above-described method for controlling the welding positioning process of the reinforcing bar skeleton of utility poles.

[0015] Fourthly, embodiments of the present invention provide a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the above-described method for controlling the welding positioning process of the pole reinforcement cage.

[0016] The beneficial effects of this invention are as follows: By constructing a closed-loop control system integrating visual positioning, servo regulation, process optimization, thermal feedback, and retesting and evaluation, this invention achieves high-precision positioning and adaptive control throughout the entire welding process of the rebar skeleton for utility poles. On the one hand, it utilizes multimodal visual perception and 3D reconstruction technology to accurately acquire the initial assembly state and generates differentiated positioning deviation data by combining component weight attributes, significantly improving alignment accuracy. On the other hand, it integrates the characteristics of the rebar itself with real-time multidimensional parameters during the welding process, dynamically optimizing the welding process strategy, and achieving coordinated closed-loop control of positioning and welding parameters through a thermal coupling feedback mechanism, effectively suppressing displacement caused by thermal deformation. Simultaneously, a retesting and verification process is automatically triggered after welding to quantitatively evaluate the geometric dimensions, ensuring that the skeleton structure meets the requirements for concrete pouring. Overall, this invention significantly improves welding consistency, dimensional stability, and automation level, solving the shortcomings of traditional methods such as reliance on manual labor, simple fixtures, and lack of dynamic feedback, providing reliable technical support for intelligent manufacturing of utility poles. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort. Wherein: Figure 1 This is a flowchart of a method for controlling the welding positioning process of a utility pole's reinforcing steel frame, provided as an embodiment of the present invention.

[0018] Figure 2 This is a schematic diagram of a method for controlling the welding positioning process of a utility pole reinforcement cage, provided in an embodiment of the present invention. Detailed Implementation

[0019] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0020] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0021] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0022] This invention is described in detail with reference to the schematic diagrams. When detailing the embodiments of this invention, for ease of explanation, the cross-sectional views illustrating the device structure may be partially enlarged, not adhering to the usual scale. Furthermore, the schematic diagrams are merely examples and should not be construed as limiting the scope of protection of this invention. In actual fabrication, the three-dimensional spatial dimensions of length, width, and depth should be included.

[0023] Furthermore, in the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are used solely for the convenience of describing the invention and for 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. Therefore, they should not be construed as limitations on the invention. In addition, the terms "first," "second," or "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0024] Unless otherwise explicitly specified and limited, the terms "installation," "connection," and "joining" in this invention should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; similarly, they can refer to mechanical connections, electrical connections, or direct connections, or indirect connections through an intermediate medium, or internal connections 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.

[0025] Example

[0026] Reference Figure 1 and Figure 2 This is the first embodiment of the present invention, which provides a method for controlling the welding positioning process of the reinforcing steel cage of a utility pole, including: S1: The initial assembly state of the steel reinforcement cage is located in three-dimensional space through a visual recognition system to obtain the position coordinates of the steel reinforcement cage components and compare them with the preset standard data template to generate positioning deviation data.

[0027] Among them, the three-dimensional spatial positioning of the initial assembly state of the steel reinforcement cage through the visual recognition system includes deploying a multimodal visual perception unit of the visual recognition system to form a 360° no-dead-angle observation layout around the steel reinforcement cage, and collecting multiple frames of deformation stripe images by projecting coded phase stripes onto the surface of the steel reinforcement cage component through the visual recognition system. The point cloud data of the steel skeleton component is reconstructed using a 3D reconstruction fusion algorithm. The geometric entities of the steel skeleton component are separated by a point cloud data segmentation method based on curvature features, and the 3D spatial position and pose of each steel skeleton component in the global coordinate system are calculated.

[0028] Obtaining the position coordinates of the steel skeleton components includes extracting the closed ring contour of the steel skeleton components based on the point cloud skeletonization algorithm after completing the three-dimensional spatial position and pose calculation of the steel skeleton components, performing topological association marking on each intersection node of the closed ring contour, mapping the feature nodes of each steel skeleton component to a unified global coordinate system, and generating a position coordinate dataset. For point cloud skeletonization algorithms, the local neighborhood centroid method is used to approximately extract the central skeleton line, and its calculation formula is as follows:

[0029] in, This represents the j-th skeleton point. This represents a reference point sampled along the circumferential direction in the original point cloud. Indicates The local neighborhood point set centered on the radial section, Indicates to The summation operation is performed on all points q within the local neighborhood point set.

[0030] The comparison with the preset standard data template includes retrieving the preset standard data template corresponding to the current pole model from the process database. The preset standard data template stores the theoretical reference data of all steel reinforcement skeleton components under ideal conditions in the same format. By establishing a component-level matching index, the location coordinate dataset is mapped one-to-one with the corresponding component items in the preset standard data template. The weighted least squares method is used to perform rigid body registration of the steel reinforcement cage, eliminating the error introduced by the overall displacement and ensuring that the comparison benchmark is consistent. The generation of positioning deviation data includes calculating the residual vector between the actual position coordinates of each steel skeleton component and the theoretical coordinates in the preset standard data template in the local coordinate system after rigid body registration is completed. Combining the component weight attributes of the steel skeleton components, a differential sensitivity coefficient is assigned to the residual vector. The generated positioning deviation data is then merged and output to the servo control module in the form of a positioning deviation data package.

[0031] Furthermore, a 3D reconstruction fusion algorithm is used to process multi-view images, and high-density point cloud data is reconstructed through phase-depth mapping. The core computation can be expressed as:

[0032] in, Represents the three-dimensional coordinates of the i-th point. Represents the projection function. This represents the x-coordinate of the i-th point in pixels on the image plane. This represents the ordinate of the pixel at point i on the image plane. This represents the phase value corresponding to the i-th point.

[0033] The calculation formula for extracting the closed ring contour of the steel skeleton component based on the point cloud skeletonization algorithm is as follows:

[0034] in, This represents a skeleton point within the j-th cross section or local neighborhood. This represents the k-th original point cloud point within the local neighborhood. This indicates the total number of point cloud points contained within this neighborhood.

[0035] S2: Based on positioning deviation data, the multi-degree-of-freedom servo clamping mechanism is driven to dynamically adjust the steel reinforcement cage, complete the alignment before welding, and lock the nodes synchronously during the dynamic adjustment process to avoid displacement caused by thermal deformation.

[0036] The dynamic adjustment of the steel reinforcement cage includes receiving the positioning deviation data packet output by the visual recognition system and then analyzing the pose weight composite parameters of the feature nodes of each steel reinforcement cage component. Based on a preset motion mapping rule library, the residual vector of each feature node is decomposed into multi-axis linkage commands for the corresponding clamping unit, including translation compensation in the X / Y / Z directions and rotation angle around the axis. The main control unit sends control signals to each servo execution module in the multi-degree-of-freedom servo clamping mechanism in a time-sequence manner according to a priority scheduling strategy, so as to apply dynamic adjustment to the steel skeleton component with sub-millimeter precision, converge the actual position to the range allowed by the theoretical reference data, and complete the alignment before welding. During the dynamic adjustment process, the synchronous locking node includes monitoring the clamping status feedback parameters of each clamping point while the multi-degree-of-freedom servo clamping mechanism performs reverse dynamic adjustment. When the position error of a certain feature node converges to within a preset threshold, the node locking execution unit is triggered to perform rigid locking on the feature node. The locking action is dynamically allocated by the embedded PLC according to the component weight attribute. The steel reinforcement cage includes main bars and stirrups; force locking is used at the intersection of the main bars and flexible limiting locking is used at the connection points of the stirrups; all locking operations are completed before the end of the dynamic adjustment process, forming a spatial constraint network to suppress the relative displacement of the steel reinforcement cage due to thermal expansion or stress release during the subsequent welding heat input stage.

[0037] Furthermore, during the dynamic adjustment of the reinforcing steel cage, a visual recognition system is first used to detect the positional deviation of the reinforcing steel cage components and output the corresponding positioning deviation data. This data includes the pose information of each feature node (i.e., the key point that needs precise alignment) and its weighted composite parameters, which reflect the importance of the node and the degree of adjustment required. Next, according to a preset motion mapping rule library, the residual vector of each feature node is decomposed into specific control commands, including translation compensation in the X, Y, and Z directions and rotation angles around the axis, to guide the multi-degree-of-freedom servo clamping mechanism on how to make precise adjustments.

[0038] The main control unit sends control signals to the servo execution module in an orderly manner according to a priority scheduling strategy, ensuring that adjustments in each degree of freedom can be achieved with high precision, so that the actual position of the steel reinforcement cage component matches the theoretical design position. During this process, to prevent displacement caused by thermal expansion or stress release due to heat input during welding, a specific locking mechanism is triggered when the position error of a certain feature node is adjusted to within a preset threshold range. Specifically, a force-locking method is used for main reinforcement intersections, while a flexible limit-locking method is used for stirrup connection points. In this way, before the entire dynamic adjustment process ends, all key nodes are rigidly or flexibly locked, forming a stable spatial constraint network, thereby effectively suppressing the displacement risk caused by factors such as temperature changes during welding, and ensuring the stability and accuracy of the steel reinforcement cage structure. This series of operations not only improves construction quality but also enhances the safety and reliability of the structure.

[0039] S3: During the welding execution phase, collect multi-dimensional parameters of the welding process and dynamically optimize the welding process parameter strategy in combination with the characteristics of the steel bar itself.

[0040] Among them, the acquisition of multi-dimensional parameters of the welding process includes the acquisition of multi-dimensional parameters of the welding process through a sensor array integrated into the welding torch body during the welding execution stage, forming a multi-dimensional parameter stream, and continuously inputting it to the process control engine at a millisecond-level sampling frequency; The dynamic optimization strategy for welding process parameters includes using a process control engine to retrieve the steel body characteristics corresponding to the steel skeleton component to be welded from the process database. The steel body characteristics and the input multi-dimensional parameters of the welding process are used as input variables and loaded into a pre-trained lightweight neural network architecture. This lightweight neural network architecture is trained based on a historical high-quality weld point dataset and maps out the best welding process parameter strategies under different working conditions. The main control unit dynamically adjusts the output parameters of the welding machine and the robot trajectory interpolation instructions during the welding process to complete the online adaptive optimization of the welding process parameter strategy.

[0041] The lightweight neural network architecture described in this invention specifically adopts a hybrid structure of one-dimensional convolutional neural network (1D-CNN) and gated recurrent unit (GRU), which takes into account both the local feature extraction and temporal dynamic modeling capabilities of welding parameters. The network consists of 5 layers: an input layer, two 1D-CNN layers (with a kernel size of 3 and channels of 32 and 64 respectively), a bidirectional GRU layer (with 128 hidden units), and a fully connected output layer. The activation function is ReLU in the convolutional and fully connected layers, and a combination of tanh and sigmoid is used inside the GRU. The input dimension is 12-dimensional, including real-time acquired welding current, voltage, wire feed speed, arc length, shielding gas flow rate, welding torch travel speed, temperature gradient of adjacent points, main rib diameter, material grade code, surface oxidation level, spatial tilt angle, and parameter change rate within a historical 50ms window; the output dimension is 6-dimensional, corresponding to the optimized pulse base current, peak current, duty cycle, welding speed, robot trajectory offset, and energy density compensation coefficient. The model was trained based on 127,000 sets of high-quality weld point data accumulated by the company over the past three years. The data came from the automated welding production line of the steel reinforcement cage of Φ150–Φ230 series electric poles. Each set of samples was verified to be qualified by X-ray flaw detection and tensile strength test, and was divided into training set, validation set and test set in a ratio of 8:1:1. The model's average absolute error in predicting process parameters on the test set was less than 3.2%, which meets the requirements for online control accuracy.

[0042] Furthermore, during the welding execution phase, the system uses a multi-source sensor array integrated into the welding torch body to collect welding process parameters in real time, covering multiple dimensions such as electricity, heat, and motion. These parameters include instantaneous waveforms of arc current and voltage, wire feed speed, shielding gas flow rate, and thermal field distribution in the weld area. This data is synchronously integrated into a continuous multi-dimensional parameter stream at a millisecond sampling frequency and transmitted to the process control engine in real time. Upon receiving this parameter stream, the process control engine immediately retrieves the steel reinforcement characteristics corresponding to the steel reinforcement skeleton component to be welded from the process database, including key attributes such as material grade, diameter, surface oxidation state, and spatial arrangement density. Subsequently, the steel reinforcement characteristics and real-time multi-dimensional parameters are used as input and loaded into a pre-trained lightweight neural network architecture. This neural network architecture, trained on a large amount of historical high-quality weld data, can accurately identify the current welding condition and map the optimal welding process parameter strategy, including the pulse current base and peak values, welding travel speed, arc length setting, and energy density distribution. Based on this strategy, the main control unit dynamically adjusts the output parameters of the welding machine and the interpolation instructions of the robot welding trajectory to achieve online adaptive optimization of the welding process, ensuring that each weld point can achieve consistent, stable, and high-quality welding results under different steel reinforcement structures and thermal environments.

[0043] S4: Establish a thermo-mechanical coupling feedback mechanism during the welding process, monitor the welding area, and feed the monitoring results back to the execution and control unit to complete the collaborative closed-loop control of the positioning welding collaborative parameters.

[0044] Monitoring the welding area includes simultaneously running a thermo-coupling feedback mechanism during the welding process. This mechanism integrates virtual and real monitoring data. First, a high-frame-rate infrared thermal imager continuously collects the temperature field evolution sequence of the welding area to obtain the transient thermal distribution on the surface of the weld point and adjacent steel reinforcement skeleton components. Second, it captures the thermo-response signal and, based on a simplified finite element architecture, online deduces the thermo-evolution state within the thermo-response signal to generate a comprehensive state vector. Feeding the monitoring results back to the execution and control unit involves encapsulating the integrated state vector into a feedback data packet and transmitting it to the execution and control unit with low latency via the industrial real-time communication bus. The collaborative closed-loop control of positioning welding parameters involves integrating the hierarchical control commands of the execution control unit with the current welding process parameter strategy to dynamically correct the positioning welding parameters. The dynamically corrected positioning welding parameters are then sent to the multi-degree-of-freedom servo clamping mechanism to complete the collaborative closed-loop control of the positioning welding parameters.

[0045] To achieve efficient online simulation of the thermo-mechanical coupling state during welding, this invention employs a proxy model based on Modal Order Reduction (MOR) as a simplified finite element architecture: based on a full-size three-dimensional thermo-mechanical coupling model, the first 20 dominant modes are retained, compressing the degrees of freedom from approximately 500,000 to less than 300; the element type is a four-node tetrahedral thermo-mechanical coupling element, the material parameters are set according to HRB400 steel reinforcement, the boundary conditions include Gaussian heat source loading, natural convection and radiation heat dissipation, and displacement constraints are applied to the locked nodes; the online solver adopts the explicit Euler integration method with a time step of 10ms, and the single-step calculation time on an embedded industrial PC is ≤3ms. The thermal response signals are clearly categorized into three types: the transient surface temperature field acquired by a high-frame-rate infrared thermal imager (≥200Hz), the micro-displacement of key points captured by a miniature laser displacement sensor, and the local strain measured by a high-temperature fiber optic strain gauge. After synchronous alignment, these three types of signals are input into a reduced-order model to generate an 8-dimensional comprehensive state vector (including predicted deformation, stress trend, temperature gradient, etc.), which is then transmitted in real time to the execution and control unit via an EtherCAT bus (1ms cycle) to support millisecond-level closed-loop control of positioning welding collaborative parameters.

[0046] Furthermore, during the welding process, the system simultaneously activates a thermo-mechanical coupling feedback mechanism to perform high-precision, multi-dimensional real-time monitoring of the welding area. This mechanism first continuously acquires the temperature field evolution sequence of the weld point and the surface of its adjacent reinforcing steel frame components using a high-frame-rate infrared thermal imager, obtaining transient thermal distribution data. Simultaneously, it uses miniature displacement sensors and strain gauges placed near key feature nodes to capture in real-time the local deformation and stress response signals caused by the thermal input, i.e., the thermo-mechanical response signals. The measured data from these physical sensors are fused with the thermo-mechanical evolution state derived online based on a simplified finite element architecture to form a virtual-real fusion monitoring data set, which further generates a comprehensive state vector containing information such as temperature gradient, predicted thermal expansion displacement, and residual stress distribution.

[0047] The integrated state vector is then encapsulated into a structured feedback data packet and transmitted to the execution control unit in a low-latency manner via an industrial-grade real-time communication bus (such as EtherCAT). After parsing the data packet, the execution control unit generates hierarchical control commands based on the intensity and location of the thermal disturbance. For example, it prioritizes fine-tuning compensation for significant thermal drift in the main reinforcement cross-node area, while adjusting the local energy input for slight deformation in the stirrup connection area. These hierarchical control commands are deeply integrated with the currently executed welding process parameter strategy to dynamically correct the positioning and welding coordination parameters, including the real-time clamping force distribution of the clamping mechanism, the local offset compensation amount of the welding torch trajectory, and the fine-tuning phase of the pulse timing. The corrected parameters are then sent to the multi-degree-of-freedom servo clamping mechanism, driving it to perform millisecond-level pose fine-tuning, thereby maintaining the spatial geometric stability of the reinforcing steel skeleton throughout the welding process and achieving high-precision collaborative closed-loop control of the positioning state and the welding process.

[0048] S5: After the welding execution phase is completed, the retest and verification process is triggered to evaluate the geometry after welding.

[0049] The triggering retest verification process includes activating the retest triggering module after the main control unit receives the welding process completion status signal; The retest trigger module is used to automatically generate retest task instructions based on the positioning welding coordination parameters, send a start command to the multimodal vision perception unit, and simultaneously release the clamping constraints on non-critical feature nodes in the node locking execution unit, so that the steel reinforcement skeleton is in a low-stress free state; at the same time, the retest trigger module sends a reset command to the servo control module, so that the multi-degree-of-freedom servo clamping mechanism retracts to the preset safe pose, providing conditions for unobstructed 3D scanning. The evaluation of post-weld geometry includes a multimodal visual perception unit performing a secondary 3D scan of the post-weld rebar skeleton using a visual scanning strategy consistent with the initial assembly and positioning stage to acquire complete point cloud data; reconstructing the point cloud data of the post-weld rebar skeleton components using a 3D reconstruction fusion algorithm; separating the geometric entities of each rebar skeleton component using a point cloud data segmentation method based on curvature features; extracting the coordinates of feature nodes of each post-weld rebar skeleton component using a point cloud skeletonization algorithm; matching the component-level coordinates with theoretical benchmark data in a preset standard data template corresponding to the same pole model; introducing a welding residual deformation compensation factor in the evaluation process; dynamically weighting the tolerance thresholds of different feature nodes in conjunction with component weight attributes; calculating the geometric continuity index of different feature nodes; and generating geometric dimension evaluation results.

[0050] Furthermore, once the welding process is completed, the main control unit immediately activates the retest trigger module and initiates the post-weld retest verification process upon receiving the "welding complete" status signal from the welding machine. This module first automatically generates a structured retest task instruction based on the positioning welding coordination parameters used in this welding process (including adjustment trajectory, clamping force distribution, and thermal feedback response). Then, it sends a start command to the multimodal vision perception unit, preparing to execute a secondary 3D scan. Simultaneously, to ensure measurement accuracy is not affected by clamping stress, the system synchronously releases the clamping constraints applied to non-critical feature nodes by the node locking execution unit, retaining only the foundation support of the main reinforcement intersection nodes, thus placing the entire steel reinforcement skeleton in a low-stress, near-free state. In addition, the retest trigger module sends a reset command to the servo control module, controlling the multi-degree-of-freedom servo clamping mechanism to smoothly retract to the preset safe posture, completely eliminating mechanical obstruction and creating ideal conditions for subsequent blind-spot-free 3D scanning.

[0051] Based on this, the multimodal visual perception unit strictly follows the same visual scanning strategy as the initial assembly and positioning stage, including the same surround observation layout and coded phase fringe projection method—performing a high-precision secondary scan of the post-weld steel reinforcement skeleton to obtain complete point cloud data containing details such as weld point bulges and heat-affected zone deformation. The system then calls a 3D reconstruction fusion algorithm to reconstruct the post-weld point cloud model and accurately separates the geometric entities of each steel reinforcement skeleton component, such as main bars and stirrups, using a point cloud data segmentation method based on curvature features. Next, a point cloud skeletonization algorithm is used to extract the coordinates of key feature nodes of each component and perform precise component-level matching with the theoretical benchmark data in the preset standard data template corresponding to the same pole model. During the evaluation process, the system introduces a welding residual deformation compensation factor to correct the influence of heat-induced permanent deformation on dimensional judgment, and dynamically weights the tolerance thresholds of different feature nodes (such as main bar intersections and stirrup connection points) in combination with component weight attributes. Finally, the system comprehensively calculates the dimensional deviation, deformation direction, and geometric continuity index of each node to generate a structured geometric dimensional evaluation result, which is used to determine whether the welding quality meets the standards and provides data basis for possible repair welding or process iteration.

[0052] During point cloud data segmentation, the system estimates the curvature of each point by analyzing the geometric distribution of its local neighborhood: main reinforcement bars are straight lines or large-radius cylinders with low surface curvature; stirrups are small-radius closed loops with significantly higher curvature. Based on calibration using a large number of measured samples, a curvature threshold of 0.08 is set. Points below this value are classified as main reinforcement bars, and points above or equal to this value are classified as stirrups. This threshold has been verified to be effective for various reinforcement configurations (e.g., main reinforcement bar diameter 12–16 mm, stirrup spacing 80–120 mm) on poles ranging from Φ150 to Φ230, and can stably and accurately separate the geometric entities of the two types of components.

[0053] To verify the practical effects of this invention, an engineering implementation was conducted on a standard tapered pole production line of Φ190×12m at a pole manufacturing enterprise. The method of this invention was used to control the welding of a steel reinforcement cage containing 12 main reinforcing bars (14mm in diameter) and spiral stirrups (100mm spacing). Test data showed that the visual positioning module's 3D recognition error for feature nodes was ≤±0.2mm; after dynamic adjustment, the positional deviation of key nodes before welding converged to within ±0.35mm (compared to an average deviation of ±2.1mm using the traditional manual clamping method); the thermal feedback mechanism, upon detecting a temperature rise rate of 180℃ / s in the main reinforcing bar area, drove the servo clamping mechanism to complete a 0.42mm reverse compensation within 80ms; post-weld retesting showed that the cage roundness deviation was ≤1.2mm and the main reinforcing bar spacing qualification rate reached 99.6%, significantly better than the control group's 3.8mm and 87.3%; simultaneously, the welding cycle time per piece was shortened by 12%, and the weld rework rate decreased from 5.7% to 0.4%. The above data fully demonstrate the engineering feasibility and technical advantages of this invention in terms of positioning accuracy, thermal deformation suppression, dimensional stability, and production efficiency.

[0054] In a preferred embodiment, a welding positioning process control system for utility pole reinforcing bar skeletons includes a visual positioning module, which uses a visual recognition system to perform three-dimensional spatial positioning of the initial assembly state of the reinforcing bar skeleton, obtains the position coordinates of the reinforcing bar skeleton components, and compares them with a preset standard data template to generate positioning deviation data; a servo control module, which drives a multi-degree-of-freedom servo clamping mechanism to dynamically adjust the reinforcing bar skeleton based on the positioning deviation data, completes the alignment before welding, and locks the nodes synchronously during the dynamic adjustment process to avoid displacement caused by thermal deformation; a process optimization module, which collects multi-dimensional parameters of the welding process during the welding execution stage, and dynamically optimizes the welding process parameter strategy in combination with the characteristics of the reinforcing bar itself; a thermal feedback module, which establishes a thermal coupling feedback mechanism during the welding process, monitors the welding area, and feeds the monitoring results back to the execution control unit to complete the collaborative closed-loop control of positioning welding parameters; and a retesting and evaluation module, which triggers a retesting and verification process after the welding execution stage is completed to evaluate the geometric dimensions after welding.

[0055] To enhance the clarity of system implementation and hardware support, the control system described in this invention includes the following specific hardware configuration in actual deployment: The visual positioning module consists of a ring-shaped multimodal visual perception unit composed of four sets of high-resolution industrial cameras (5 megapixels, 60 fps) and four blue light structured light projectors (wavelength 450 nm, fringe frequency 120 Hz), evenly distributed at 90° around the steel reinforcement skeleton to form a 360° observation layout without blind spots. The system uses the Zhang Zhengyou calibration method to complete the joint calibration of internal and external parameters, with a reprojection error ≤0.15 pixels, ensuring a 3D reconstruction accuracy better than ±0.2 mm. The core of the servo control module is a six-degree-of-freedom parallel servo clamping mechanism, containing 12 independent electric servo grippers (repeat positioning accuracy ±0.02 mm). The main reinforcement clamping unit uses a double-sided pneumatic-electric hybrid drive gripper with closed-loop force control (maximum clamping force 500 N); the stirrup clamping unit integrates a flexible silicone limit block and a piezoelectric fine adjuster, enabling 0–2 mm adaptive compensation. In the welding process of a typical Φ190×12 m tapered pole (12 main ribs HRB400 Φ14 mm, spiral stirrups Φ6 mm@100 mm spacing), the system executes the following key parameters: initial positioning deviation threshold ±1.5 mm, dynamic adjustment convergence target ±0.35 mm, welding current 280–320 A, pulse frequency 5 Hz, wire feed speed 6.2 m / min, shielding gas (Ar+CO2 80 / 20) flow rate 18 L / min, thermal feedback trigger temperature gradient threshold ≥150 ℃ / s, and post-weld roundness tolerance ≤1.5 mm. The above hardware structure and process parameters have been running stably for over 6 months on a national-level intelligent pole manufacturing demonstration production line, verifying the system's engineering feasibility and technical integrity.

[0056] The above-mentioned unit modules can be embedded in the processor of the computer device in hardware form or independent of it, or they can be stored in the memory of the computer device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0057] In one embodiment, a computer device is provided, which may be a terminal. The computer device includes a processor, memory, a communication interface, a display screen, and an input device connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage medium. The communication interface of the computer device is used for wired or wireless communication with external terminals. Wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen of the computer device may be an LCD screen or an e-ink display screen. The input device of the computer device may be a touch layer covering the display screen, or buttons, a trackball, or a touchpad located on the casing of the computer device, or an external keyboard, touchpad, or mouse, etc.

[0058] In summary, this invention achieves high-precision positioning and adaptive control throughout the entire welding process of the rebar skeleton for utility poles by constructing a closed-loop control system integrating visual positioning, servo regulation, process optimization, thermal feedback, and retesting and evaluation. On one hand, multimodal visual perception and 3D reconstruction technology are used to accurately acquire the initial assembly state, and differentiated positioning deviation data is generated by combining component weight attributes, significantly improving alignment accuracy. On the other hand, during the welding process, the characteristics of the rebar itself and real-time multidimensional parameters are integrated to dynamically optimize the welding process strategy. A thermo-coupling feedback mechanism is used to achieve coordinated closed-loop control of positioning and welding parameters, effectively suppressing displacement caused by thermal deformation. Simultaneously, a retesting and verification process is automatically triggered after welding to quantitatively evaluate the geometric dimensions, ensuring that the skeleton structure meets the requirements for concrete pouring. Overall, this invention significantly improves welding consistency, dimensional stability, and automation levels, solving the shortcomings of traditional methods such as reliance on manual labor, simple fixtures, and lack of dynamic feedback, providing reliable technical support for intelligent manufacturing of utility poles.

[0059] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for controlling the welding positioning process of the reinforcing steel cage for utility poles, characterized in that: The application relates to a steel bar skeleton positioning and welding system. The initial assembly state of the steel bar skeleton is positioned in three-dimensional space through a visual identification system, position coordinates of the steel bar skeleton component are acquired, and the position coordinates are compared with preset standard data templates to generate positioning deviation data; Based on the positioning deviation data, a multi-degree-of-freedom servo clamping mechanism is driven to dynamically adjust the steel bar skeleton, and alignment before welding is completed, and nodes are locked synchronously during the dynamic adjustment process to avoid displacement caused by thermal deformation; During the welding execution stage, multi-dimensional parameters of the welding process are collected, and welding process parameter strategies are dynamically optimized in combination with the characteristics of the steel bar body; A thermal force coupling feedback mechanism is established in the welding process, the welding area is monitored, and the monitoring results are fed back to an execution control unit; After the welding execution stage is completed, a retest verification process is triggered to evaluate the geometric size after welding.

2. The method of claim 1, wherein the method further comprises: determining a welding position of the steel reinforcement frame; and adjusting the welding position of the steel reinforcement frame based on the determined welding position. The initial assembly state of the steel bar skeleton is positioned in three-dimensional space through a visual identification system, position coordinates of the steel bar skeleton component are acquired, and the position coordinates are compared with preset standard data templates to generate positioning deviation data; Point cloud data of the steel bar skeleton component is reconstructed by using a three-dimensional reconstruction fusion algorithm, geometric entities of the steel bar skeleton component are separated by a point cloud data segmentation method based on curvature characteristics, and the three-dimensional spatial position and posture of each steel bar skeleton component in a global coordinate system are calculated.

3. The method of claim 2, wherein the method further comprises: determining a welding position of the steel reinforcement frame; and adjusting the welding position of the steel reinforcement frame based on the determined welding position. The position coordinates of the steel bar skeleton component are acquired by extracting a closed annular contour of the steel bar skeleton component based on a point cloud skeleton algorithm after the three-dimensional spatial position and posture of the steel bar skeleton component are calculated, topologically associating and marking each intersection node of the closed annular contour, mapping the feature nodes of each steel bar skeleton component to a unified global coordinate system, and generating a position coordinate data set; The preset standard data templates corresponding to the current electric pole model are called from a process database, and the preset standard data templates store the theoretical reference data of all steel bar skeleton components in an ideal state in the same format; The position coordinate data set and the corresponding component items in the preset standard data templates are one-to-one mapped by establishing a component-level matching index, and a weighted least square method is used for rigid registration of the steel bar skeleton; The actual position coordinates of each steel bar skeleton component and the residual error vectors of the theoretical coordinates in the preset standard data templates are calculated in the local coordinate system after the rigid registration is completed, the residual error vectors are given different sensitivity coefficients in combination with the component weight attributes of the steel bar skeleton component, the generated positioning deviation data are fused, and the positioning deviation data are output to a servo control module in the form of a positioning deviation data packet.

4. The method of claim 3, wherein the method further comprises: determining a welding position of the steel reinforcement frame; and adjusting the welding position of the steel reinforcement frame. The dynamic adjustment of the steel bar skeleton includes analyzing the pose weight composite parameters of the feature nodes of each steel bar skeleton component after receiving the positioning deviation data packet output by the visual identification system; Based on the preset motion mapping rule library, the residual vector of each feature node is decomposed into multi-axis linkage instructions of the corresponding clamping unit, including X / Y / Z direction translation compensation and rotation angle around the axis; the main control unit sends control signals to each servo execution module in the multi-degree-of-freedom servo clamping mechanism in time sequence according to the priority scheduling strategy, so as to apply dynamic adjustment to the steel reinforcement framework member with sub-millimeter level precision, converge the actual position to the theoretical reference data within the allowable range, and complete the alignment before welding; The synchronous locking node in the dynamic adjustment process includes monitoring the clamping state feedback parameters of each clamping point while the multi-degree-of-freedom servo clamping mechanism performs reverse dynamic adjustment; When the position error of a certain feature node converges to within the preset threshold, the node locking execution unit triggers rigid locking of the feature node, and the locking action is dynamically allocated by the embedded PLC according to the component weight attribute. The steel reinforcement framework member includes main reinforcement and stirrup; the cross node of the main reinforcement is force-locked, and the connection point of the stirrup is flexibly limited; all locking operations are completed before the end of the dynamic adjustment process, forming a spatial constraint network to inhibit the relative displacement of the steel reinforcement framework member caused by thermal expansion or stress release in the subsequent welding heat input stage.

5. The method of claim 4, wherein the method further comprises: determining a welding position of the steel reinforcement frame; and adjusting the welding position of the steel reinforcement frame based on the determined welding position. The multi-dimensional parameter acquisition in the welding process includes collecting multi-dimensional parameters in the welding process through a sensing array integrated in the welding gun body during the welding execution stage, forming a multi-dimensional parameter flow, and continuously inputting the multi-dimensional parameter flow to the process control engine at a millisecond level sampling frequency; The dynamic optimization of the welding process parameter strategy includes using the process control engine to retrieve the steel reinforcement body characteristics corresponding to the current steel reinforcement framework member to be welded from the process database, using the steel reinforcement body characteristics and the input multi-dimensional parameters in the welding process as input variables, and loading them into a pre-trained lightweight neural network architecture, which is trained based on historical high-quality weld data set, to map out excellent welding process parameter strategies under different working conditions, and dynamically adjust the welding machine output parameters and the robot trajectory interpolation instructions in the welding process using the main control unit.

6. The method of claim 5, wherein the method further comprises: determining a welding position of the steel reinforcement frame; and determining a welding direction of the steel reinforcement frame. The monitoring of the welding area includes synchronously running a thermal coupling feedback mechanism during the welding process, which fuses virtual and real monitoring data. First, a high-frame-rate infrared thermal imager is used to continuously collect the temperature field evolution sequence of the welding area to obtain the transient thermal distribution of the weld and the adjacent steel reinforcement framework member surface; Second, the thermal response signal is captured, and the thermal evolution state inside the thermal response signal is deduced online based on a simplified finite element architecture to generate a comprehensive state vector; The feedback of the monitoring results to the execution control unit includes using the comprehensive state vector to be packaged as a feedback data packet, which is transmitted to the execution control unit through an industrial real-time communication bus with low delay; The collaborative closed-loop regulation of the positioning welding collaborative parameters includes fusing the hierarchical regulation instructions of the execution control unit with the current welding process parameter strategy, dynamically correcting the positioning welding collaborative parameters, and sending the dynamically corrected positioning welding collaborative parameters to the multi-degree-of-freedom servo clamping mechanism.

7. The method of claim 6, wherein the method further comprises: determining a welding position of the steel reinforcement frame; and adjusting the welding position of the steel reinforcement frame based on the determined welding position. The trigger retest verification process includes activating the retest trigger module when the main control unit receives the welding process completion state signal. The re-measurement triggering module is used for automatically generating a re-measurement task instruction according to the positioning welding coordination parameter, sending a start command to the multi-modal visual perception unit, synchronously releasing the clamping constraint of the non-key feature node in the node locking execution unit, so that the reinforcement framework is in a low-stress free state; at the same time, the re-measurement triggering module sends a reset instruction to the servo control module, and the multi-degree-of-freedom servo clamping mechanism is reset to a preset safe pose; The evaluation of the geometric size after welding includes that the multi-modal visual perception unit performs secondary three-dimensional scanning on the welded reinforcement framework according to the visual scanning strategy consistent with the initial assembly positioning stage, and obtains complete point cloud data; The point cloud data of the welded reinforcement framework component is reconstructed by using a three-dimensional reconstruction fusion algorithm, the geometric entities of each reinforcement framework component are separated by using a point cloud data segmentation method based on curvature characteristics, the feature node coordinates of each welded reinforcement framework component are extracted by using a point cloud skeletonization algorithm, the theoretical reference data in the preset standard data template corresponding to the same pole model are matched at the component level, a welding residual deformation compensation factor is introduced in the evaluation process, the tolerance threshold of different feature nodes is dynamically weighted in combination with the component weight attribute, the geometric continuity index of different feature nodes is calculated, and a geometric size evaluation result is generated.

8. A pole reinforcement framework welding positioning process control system based on the pole reinforcement framework welding positioning process control method of any one of claims 1-7, characterized in that: It comprises, a visual positioning module that performs three-dimensional spatial positioning on the initial assembly state of the reinforcement framework by using a visual recognition system, obtains the position coordinates of the reinforcement framework component, and compares the position coordinates with a preset standard data template to generate positioning deviation data; a servo control module that drives a multi-degree-of-freedom servo clamping mechanism to dynamically adjust the reinforcement framework based on the positioning deviation data, and completes the alignment before welding, and synchronously locks the nodes during the dynamic adjustment process to avoid displacement caused by thermal deformation; a process optimization module that collects multi-dimensional parameters during the welding process, dynamically optimizes the welding process parameter strategy in combination with the characteristics of the reinforcement body during the welding execution stage; a thermal feedback module that establishes a thermal coupling feedback mechanism during the welding process, monitors the welding area, and feeds back the monitoring result to the execution control unit; a re-measurement evaluation module that triggers a re-measurement verification process to evaluate the geometric size after welding after the welding execution stage is completed. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is characterized in that: The processor executes the computer program to realize the steps of the electric pole reinforcement framework welding positioning process control method in any one of claims 1-7.

10. A computer readable storage medium having stored thereon a computer program, characterized in that: The computer program is executed by the processor to realize the steps of the electric pole reinforcement framework welding positioning process control method in any one of claims 1-7.

Images