Tower foot base welding method

Abstract

The invention provides a tower foot base welding method which comprises the steps that two-dimensional plane graph data of an assembly are obtained, and a corresponding three-dimensional entity graph is generated through a geometric feature recognition algorithm; assembling the three-dimensional entity graph into a three-dimensional model according to the assembling position of the two-dimensional plane graph data, and forming a gap area between assemblies; the gap area in the three-dimensional model is automatically recognized to serve as a welding seam, and groove treatment is conducted on the welding seam; the plate thickness of the weld joint is detected, a groove form is selected in a self-adaptive mode according to the plate thickness, and meanwhile a robot welding path is generated; and welding parameters corresponding to the plate thickness and the groove form are matched from a welding parameter library, and a welding execution scheme is output.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, and in particular to a tower foot base welding method. BACKGROUND

[0002] As an important component of overhead line of power grid, the transmission line tower plays an indispensable role, and the precision and efficiency of its design and manufacturing are directly related to the safety and economic benefits of the power grid. With the continuous expansion of the power grid scale and the increase of the complexity of the tower structure, how to realize the efficient connection of the whole process from design to manufacturing has become an urgent demand for industry development.

[0003] However, in the current transmission line tower engineering, there are still many bottlenecks from design to actual manufacturing. Many existing methods often cannot effectively deal with the conversion problem between design data and actual manufacturing requirements when dealing with complex engineering projects, especially when facing diversified structural forms and construction environments, lack of flexibility and adaptability. This leads to frequent deviations between design intent and manufacturing execution, affecting the overall quality and progress of the project.

[0004] The deeper problem is that the component welding process in the transmission line tower engineering is a key quality control point, but the automation degree of the welding path planning, which is a core link, is insufficient. Welding path planning not only determines the welding sequence and position, but also needs to consider the structural characteristics, material thickness and special requirements of the welding area. If this link is not handled properly, it will lead to unstable welding quality, and even cause structural deformation and other problems. For example, in the manufacturing of transmission line towers, if the welding path is not optimized according to the component structure form and plate thickness difference, it is easy to cause component deformation due to increased heat input and uneven heat stress distribution, thereby weakening the overall structural stability, especially in complex engineering, which becomes a key technical bottleneck restricting quality, and becomes a technical barrier that needs to be overcome.

[0005] Therefore, how to realize the intelligentization and adaptive adjustment of welding path planning in the transmission line tower engineering to meet the needs of different structural forms and material characteristics has become a key problem to improve manufacturing precision and efficiency. SUMMARY

[0006] The present application provides a tower foot base welding method, mainly comprising: Obtaining two-dimensional planar graphic data of the component, and generating corresponding three-dimensional entity graphics through a geometric feature recognition algorithm; Assembling the three-dimensional entity graphics into a three-dimensional model according to the assembly position of the two-dimensional planar graphic data, and forming a gap region between the components, wherein the gap region includes a T joint of the plate and the plate; automatically identifying the gap region in the three-dimensional model as a weld seam, and performing a beveling process on the weld seam; detecting a plate thickness of the weld seam and adaptively selecting a bevel form according to the plate thickness, while generating a robot welding path; matching welding parameters corresponding to the plate thickness and the bevel form from a welding parameter library, and outputting a welding execution scheme.

[0007] Further, the acquisition component generates corresponding three-dimensional entity graphics through a geometric feature recognition algorithm, including: extracting part contours and assembly positioning information from engineering drawing formats as two-dimensional planar graphics data; applying a boundary extraction algorithm to process the part contours to obtain boundary feature data; generating corresponding three-dimensional entity graphics according to the boundary feature data and a topology reconstruction algorithm.

[0008] Further, the three-dimensional entity graphics are assembled into a three-dimensional model according to the assembly position of the two-dimensional planar graphics data, and a gap region is formed between the components, wherein the gap region includes a T-joint between plates, including: identifying constraint relationships in the assembly position of the two-dimensional planar graphics data, including a positioning point and a fitting surface; applying the constraint relationships to intelligently assemble the three-dimensional entity graphics to generate a three-dimensional model; detecting the gap distance between the components in the three-dimensional model, and marking the gap region including the T-joint between the plates.

[0009] Further, the automatic identification of the gap region in the three-dimensional model as a weld seam, and performing a beveling process on the weld seam, including: scanning the three-dimensional model through a gap distance threshold algorithm to automatically mark the gap region as a weld seam; classifying the joint types of the weld seam, and performing adaptive beveling processes for each type of joint type that needs beveling; calculating the geometric parameters of the weld seam, including the bevel angle and the bevel depth, while preliminarily determining the root face thickness range according to the weld joint type and the preset welding process specification; optimizing the beveling process by integrating deformation control rules, and adjusting the bevel depth in combination with the preliminarily determined root face thickness range, and outputting the processed weld seam data.

[0010] Further, the detection of the plate thickness of the weld seam and the adaptive selection of the bevel form according to the plate thickness, while generating a robot welding path, including: extracting a plate thickness value from the weld data, and selecting a single groove form if the plate thickness value is less than or equal to a preset plate thickness threshold, otherwise selecting a double groove form; optimizing the preliminarily determined root face thickness range based on the plate thickness value, and extracting a weld center line as a basic trajectory according to the groove form and the optimized root face thickness range; fine-tuning a basic trajectory offset in combination with a preset root gap parameter to ensure that a welding gun trajectory is aligned with a center of a root gap; generating the robot welding path by superimposing a standard welding sequence, and performing collision detection.

[0011] Further, the welding parameters corresponding to the plate thickness and the groove form are matched from a welding parameter library, and a welding execution scheme is output, including: inquiring a preset welding parameter library to determine a matching condition according to the plate thickness, the groove form, the optimized root face thickness range, and a preset root gap parameter; obtaining the matched welding parameters, including a welding current, a welding voltage, a welding speed, a welding gun swing range including a swing width and an angle, a root face thickness specific value, and a root gap specific value; integrating the welding parameters and the robot welding path to form the welding execution scheme; adding a real-time monitoring mechanism to the welding execution scheme.

[0012] Further, the part contour is processed by using a boundary extraction algorithm to obtain boundary feature data, including: performing image processing on the part contour to identify an edge point set; constructing a topological relationship from the edge point set, including point adjacency, contour closure, and component groove edge identification correlation, to obtain preliminary boundary feature data, wherein the preliminary boundary feature data includes contour types, boundary line segment attributes, and coordinate accuracy verification results; integrating the preliminary boundary feature data and assembly positioning information, eliminating data conflicts through geometric consistency verification, to obtain boundary feature data.

[0013] Further, the gap distance threshold algorithm is used to scan the three-dimensional model to automatically mark the gap region as a weld, including: scanning all component inter-region areas of the three-dimensional model according to a preset gap distance threshold; determining whether the scanned region meets the gap distance threshold, and marking as the weld if it meets the gap distance threshold; classifying the marked weld types.

[0014] The technical scheme provided by the embodiments of the present application can include the following beneficial effects: The application discloses a tower foot base welding method, and aims at the whole-process automation problem from two-dimensional design to three-dimensional modeling and then to welding execution in a power transmission line tower engineering, and innovatively solves the integrated problem of tower part data conversion, tower foot model construction and welding path planning. In particular, in a complex scene such as a power transmission line tower and a substation gantry, the application accurately converts two-dimensional plane data of the tower into a three-dimensional entity model through intelligent processing, automatically identifies a tower foot welding seam area, optimizes a groove design and a welding path, and combines a plate thickness self-adaptive selection groove form and a deformation prevention strategy to ensure tower foot welding quality and efficiency. The application also dynamically adjusts welding parameters through a tower parameter library matching and a real-time monitoring mechanism to guarantee operation stability. Finally, the application realizes seamless connection from design to manufacturing, significantly improves the automation level of the power transmission line tower engineering, reduces manual intervention cost, improves the precision and reliability of complex welding components, and shows excellent technical application value. BRIEF DESCRIPTION OF DRAWINGS

[0015] Figure 1 A flowchart of the tower foot base welding method of the application. DETAILED DESCRIPTION

[0016] In order to make the purpose, technical scheme and advantages of the application clearer, the application will be described in detail below with reference to the drawings and specific embodiments.

[0017] As Figure 1 , the tower foot base welding method of the embodiment can specifically include: Step S1, two-dimensional plane graphic data of a steel structure assembly is acquired, and boundary detection is performed through a Canny edge extraction algorithm to extract an assembly contour; then, based on the contour data, three-dimensional geometric parameters are generated by combining depth information calculation and geometric feature template matching; finally, an entity graph is constructed by using a three-dimensional modeling algorithm. Data reading, structured processing and feature extraction are included to ensure modeling accuracy.

[0018] Step S11, part contours and assembly positioning information are extracted from engineering drawing formats as two-dimensional plane graphic data. In a possible implementation manner, engineering drawings are usually stored in standard formats, such as common vector graphic files or special design software formats. The system first parses these files, identifies part contour lines and label information contained therein, such as straight lines, circular arcs, hole diameters, corner points and other basic geometric elements. At the same time, assembly positioning information is extracted, including the relative positions between parts, connection point coordinates and possible alignment marks. These information provides basic data for subsequent geometric reconstruction. For example, when processing tower foot base assembly welding, the system extracts contour data such as the length, width and plate thickness of each piece from the drawing, as well as the positioning point coordinates of the T joint of the shoe plate and the base plate and the stiffener, to ensure the integrity and accuracy of the data.

[0019] Step S12, the Canny edge extraction algorithm is applied to process the part contour to obtain boundary feature data. The input is a gray image of the part contour, the algorithm detects the edge through gradient calculation and double threshold processing, and the output is structured boundary feature data including boundary point coordinates and topological relationship. Specifically, the boundary extraction process is to convert scattered lines and points into structured boundary information. The system performs image processing and geometric analysis on the contour, identifies closed boundary regions, and smooths the boundary lines to eliminate noise or irregular points, laying the foundation for three-dimensional entity generation.

[0020] Step S121, image processing is performed on the part contour to identify edge point sets. In one embodiment, the system first preprocesses the drawing data, filters out irrelevant annotations or auxiliary lines, and only retains line information related to the part contour. Then, through edge detection technology, key points on the contour line are identified to form edge point sets. For example, when processing the cross-sectional contour of the tower foot base assembly, the system identifies the edge points of the two sides and arranges these points in order to form a complete point set list. These edge point sets are the basis for subsequent construction of topological relationships.

[0021] Step S122, topological relationships are constructed from the edge point sets to obtain boundary feature data. Specifically, the system constructs the connection relationship between points based on the distribution and connection order of the edge point sets to form closed or open boundary lines. For example, when processing the tower foot connecting plate, the system identifies the boundary lines of the outer contour and the boundary lines of the bolt holes, and records the topological relationships of the two respectively to ensure that the inner and outer boundaries are not confused. The finally generated boundary feature data contains complete geometric information and structural information, providing a reliable basis for subsequent three-dimensional modeling.

[0022] Step S123, integrate boundary feature data and assembly information to prepare for three-dimensional entity graph generation. In one possible implementation, the system integrates the extracted boundary feature data with the previously obtained assembly positioning information to form a comprehensive data set containing geometric and positional relationships. For example, when processing the tower foot main material assembly of the tower insert, the system combines the cross-sectional boundary data of the main material with its position coordinates in the inclined material connecting shoe plate overall structure to ensure that the three-dimensional entity graph generated subsequently can accurately reflect the design intent. This process also includes preliminary verification of the data, such as checking whether the boundary is closed and whether the position information is consistent, to avoid errors in subsequent modeling.

[0023] Step S13, generate a three-dimensional entity graph according to the boundary feature data and the Delaunay triangulation algorithm, and integrate the material attribute information. Specifically, the system takes the boundary point cloud data as input, and the Delaunay triangulation algorithm removes noise through point cloud preprocessing, constructs a triangular mesh of two-dimensional point set, ensures no overlap and meets geometric constraints, and outputs a three-dimensional triangular mesh model. The boundary optimization adjusts the grid quality using the minimum angle criterion. Material attribute information such as steel grade, density, strength, etc. is extracted from design drawings or databases for mechanical analysis and welding parameter matching. The final three-dimensional entity graph has both geometric shape and physical property description.

[0024] Step S14, verify the geometric consistency of the three-dimensional entity graph, and output the parameterized model file to support subsequent assembly. In one embodiment, the system performs geometric consistency check on the generated three-dimensional entity graph to ensure its matching degree with the original two-dimensional data. For example, check whether the boundary size of the three-dimensional entity is consistent with the drawing, whether there are geometric distortion or overlap problems, etc. If found inconsistent, the system will automatically adjust or prompt manual intervention. After verification, the system exports the three-dimensional entity graph as a parameterized model file, which supports subsequent assembly operations and can be shared between different software platforms. For example, when processing the tower foot base assembly of the tower, the system generates independent parameterized model files for each component's shoe plate and bottom plate and stiffener plate, and records their unique identification for quick calling during assembly.

[0025] Step S2, assemble the three-dimensional entity graph into a three-dimensional model according to the assembly position of the two-dimensional planar graph data, and form gap areas between components. Specifically, the system will automatically complete the alignment and connection of the three-dimensional entity graph according to the previously extracted assembly positioning information, and identify the gap areas between components. These gap areas are usually the target positions for subsequent welding operations, so special attention should be paid to their size and distribution.

[0026] Step S21, identify the constraint relationship in the assembly position, including the alignment point and the contact surface. In one possible implementation, the system analyzes the assembly information in the two-dimensional planar graph data and extracts the constraint relationship between parts. For example, when processing the T joint of the tower foot plate and the shoe plate of the tower, the system identifies the positioning position of the bottom plate and the shoe plate, and the contact surface of the two, and records the specific parameters of these constraint relationships, such as the coordinates of the alignment point, the normal vector of the contact surface, etc. These constraint relationships provide guidance for subsequent intelligent assembly, ensuring that parts can be accurately combined according to the design intent.

[0027] Step S22, the three-dimensional entity graphics are intelligently assembled by applying the constraint relationship to generate an overall three-dimensional model. Specifically, the system will adjust the position and rotate each three-dimensional entity graphics according to the identified fitting points and fitting surfaces, so as to meet the assembly requirements. For example, when assembling the tower foot base assembly of the iron tower, the system will assemble the large boot plate, the split boot plate and the stiffening plate according to the positioning combination position of the base according to the fitting points of the tower foot base plate and the boot plate, and ensure that the base of each tower and the boot plate fitting surface completely coincide. After assembly, the system generates an overall three-dimensional model, which not only contains the geometric information of each component, but also embodies the spatial relationship between them, providing a basis for subsequent weld seam identification.

[0028] Step S23, detect the gap distance between components in the three-dimensional model and mark the gap area. In one embodiment, the system scans the assembled three-dimensional model comprehensively through the nearest point distance algorithm. The specific steps are as follows: first, convert the three-dimensional model of the component into point cloud data or triangular mesh model; second, use KD tree structure to accelerate the nearest point search and calculate the minimum distance between point pairs of two components; finally, mark the area with a gap distance between 2mm and 5mm as a potential welding area. This range is suitable for most steel structure welding scenes according to common welding process specifications and steel structure welding specification standards, ensuring welding strength and process feasibility. These marked areas are the objects of subsequent weld seam identification and processing, which need to be further analyzed for their geometric characteristics.

[0029] Step S24, adjust the gap distance to meet the welding threshold requirements and generate a bill of materials after assembly. Specifically, the system will fine-tune the gap distance of the marked gap area according to the requirements of the welding process. For example, if the gap distance is too small in some places, it is not conducive to welding operation, the system will automatically adjust the position of the related components to make the gap reach the appropriate range, usually between 3mm and 6mm. After adjustment, the system will also generate a bill of materials after assembly, listing all components in the model, including their quantity, specifications, etc. For example, when processing the iron tower structure, the bill of materials will record the number, size and material grade of each tower foot plate, boot plate and stiffening plate in detail, facilitating subsequent manufacturing and welding preparation.

[0030] Step S3, automatically identify the gap area in the three-dimensional model as a weld seam and perform groove processing on the weld seam. Specifically, the system will use a pre-set algorithm to scan the gap area in the model and define it as a weld seam, and design a groove form according to the specific form of the weld seam. Groove processing is an important step before welding, which can effectively improve the welding quality and connection strength.

[0031] Step S31, scan the three-dimensional model by a gap distance threshold algorithm based on geometric distance calculation, and automatically mark the gap region as a weld. In one possible implementation, the system sets the gap distance threshold range to 2-8 mm, calculates and compares the distances of the regions between components in the three-dimensional model. If the gap distance of a region falls within the range, the system marks it as a weld. At the same time, the weld joint type is classified according to the connection form of the components, for example, the connection between the inserted angle steel and the shoe plate is marked as a "butt joint", the connection between the tower foot plate and the shoe plate is marked as a "corner joint", and the connection between the shoe plate and the shoe plate is marked as a "corner joint". These marks provide targets for subsequent groove design.

[0032] Step S311, set the gap distance threshold, and scan all regions between components in the three-dimensional model. Specifically, the system sets the gap distance threshold range to 2-5 mm according to the general requirements of the welding process and industry standards such as the Steel Structure Welding Specification, and uses geometric analysis tools to scan the three-dimensional model comprehensively, records the gap distance value of each region, compares it with the threshold range, and ensures that all potential welding regions are identified.

[0033] Step S312, determine whether the scanned region meets the gap distance threshold, and if it does, mark it as a weld. In one embodiment, the system will judge the gap distance value obtained by scanning one by one, and if the gap distance of a certain region is within the preset threshold range, for example, 3-6 mm, it will be marked as a weld region. For example, when processing the node connection of the tower foot, the system will find that the gap distance of some nodes is 4.5 mm, which meets the threshold condition, so it will be marked as a weld and its position and geometric information will be recorded. These marked regions will become the focus of subsequent groove processing.

[0034] Step S313, classify the types of marked welds and associate them with the groove processing flow. Specifically, the system will classify the types of welds according to their position and geometric characteristics, such as butt welds, corner welds or lap welds, and associate the classification results with the groove processing flow. For example, when processing the connection between the tower leg and the node plate, the system will identify the weld as a T-joint weld and associate it with the groove design process dedicated to T-joint welds. This classification and association process helps to adopt different processing methods for different types of welds in the subsequent process.

[0035] Step S32, classify the joint type of the weld and perform the groove design for the typical joints of the tower. In one possible implementation, the system further analyzes the joint type of the marked weld, such as butt joint, T joint or cross joint, and designs a specific groove form for the T joint, butt joint and other joints that frequently occur in the tower. For example, when the tower base is connected by the inserted main material angle steel and the shoe plate, the butt joint of the angle steel and the shoe plate is connected, and a V-shaped groove is designed for it to increase the welding contact area and improve the connection strength. The groove design is personalized according to the joint type and welding requirements.

[0036] Step S33, calculate the geometric parameters of the weld, including the angle and the depth, and preliminarily determine the range of the root face thickness according to the joint type and the preset process specification. Specifically, the system calculates the geometric parameters required for groove processing according to the geometric shape of the weld and the joint type. For example, when processing the T joint of the tower foot, the system calculates the angle of the groove to be 45 degrees, the depth to be 6 millimeters, and preliminarily determines the range of the root face thickness to be 1.5-2.5 millimeters, and records these parameters for subsequent processing. The calculation of these geometric parameters is based on the standard requirements of the welding process, to ensure that the groove form can meet the actual welding requirements.

[0037] Step S34, integrate the deformation control rules to optimize the groove processing, and output the processed weld data. The deformation control rules are based on the material thermal expansion coefficient, the welding process parameters and the structure stress analysis. First, the material thermal expansion coefficient is obtained by querying the standard database, and if it is greater than 12 times 10 to the minus 6 power per degree Celsius, it is determined to be a high deformation risk. Second, the heat input is estimated by the welding current, voltage and welding speed, and if it exceeds 2 kilojoules per millimeter, the groove parameters need to be adjusted. Finally, the structure stress analysis is based on finite element simulation, and the stress distribution is used to determine the groove adjustment value. For high-risk materials, the groove depth is reduced by 10%, and the angle is adjusted to 45 degrees to reduce the heat affected zone. After optimization, the system outputs the weld data, including the groove depth, angle and position information, to support the subsequent welding path planning.

[0038] Step S4, detect the plate thickness of the weld and self-adaptively select the groove form according to the plate thickness, and generate the robot welding path. Specifically, the plate thickness is a key parameter that affects the welding quality and directly determines the selection of the groove form. The generation of the welding path needs to consider the weld position, groove form and the accessibility of the robot operation. Through this step, the system can ensure that the welding process meets the process requirements and realizes the efficiency of automatic operation.

[0039] Step S41, extracting plate thickness values from the weld data. In one possible implementation, the system extracts the plate thickness information of the components associated with the weld from the previously generated weld data. These plate thickness values are usually stored in the parameterized file of the three-dimensional model, containing the thickness data of the components on both sides of each weld area. For example, when processing the welds of the shoe plate and stiffening plate in the tower foot base, the system reads the thickness of the shoe plate and the stiffening plate, finds that the thickness of the shoe plate is 16 mm and the thickness of the stiffening plate is 10 mm, and records these data for subsequent analysis. This process ensures the accuracy and completeness of the plate thickness information, providing a reliable basis for the selection of the groove form.

[0040] Step S42, selecting a single-sided groove form if the plate thickness value is less than or equal to the preset threshold, otherwise selecting a double-sided groove form, and optimizing the preliminary determined root face thickness range based on the extracted plate thickness value. Specifically, the system will judge the extracted plate thickness value according to the preset plate thickness threshold, for example, 8 mm. If the plate thickness of the components on both sides of the weld is less than or equal to the threshold, a single-sided groove form is selected to reduce the welding workload; if the plate thickness of any side exceeds the threshold, a double-sided groove form is selected to ensure the welding strength. When the plate thickness is ≤10 mm, the lower limit of the preliminary root face thickness is adjusted downward by 0.2-0.3 mm; when the plate thickness is >10 mm and ≤20 mm, the upper limit of the preliminary root face thickness is adjusted upward by 0.3-0.5 mm; when the plate thickness is >20 mm, an additional 0.2 mm root face thickness is added. For example, when processing the connection of the secondary nodes of the tower foot, the system finds that the thickness of the steel plates on both sides of the weld is 6 mm and 7 mm respectively, both of which are lower than the threshold, so a single-sided groove form is selected for processing. This adaptive selection mechanism can flexibly adjust the groove design according to the actual situation.

[0041] Step S43, extracting the weld centerline as the base trajectory according to the groove form. In one embodiment, the system will analyze the geometric form of the weld according to the determined groove form, and extract the centerline of the weld area as the base trajectory of the robot welding. For example, when processing the welds of the tower foot base, if a single-sided groove form is selected, the system will calculate the centerline position of the groove bottom and define it as the reference line of the welding trajectory. This centerline trajectory not only serves as the basis for the welding path, but also provides a reference for subsequent path optimization, ensuring that the welding operation can accurately cover the weld area.

[0042] Step S44, superimpose the standard welding sequence to generate the robot welding path, and perform collision detection to ensure accessibility. The standard welding sequence refers to the sequence determined according to the welding process specification, which usually includes linear sequence from one end of the weld to the other end, or the sequence of layered multi-pass welding, such as welding the bottom layer first and then the surface layer, and welding from left to right for each layer to ensure welding quality and efficiency. The system converts the base trajectory into a complete path according to this sequence, while performing collision detection to analyze whether the robot arm movement will interfere with other components or equipment, ensuring the path executability.

[0043] After collision detection is completed, the approximate range of the welding gun swing is preliminarily determined according to the groove width and the optimized root face thickness, and the range is written into the weld data as a reference basis for subsequent welding parameter matching to avoid conflicts between the matched swing parameters and the path adaptability.

[0044] Step S45, adjust the robot welding path to prevent deformation and output the path planning data. The system introduces anti-deformation rules to optimize the welding path based on the principle of uniform heat distribution, reducing deformation caused by heat concentration. The specific rules are as follows: divide the weld into several small sections, each section is about 10% of the total length, this proportion is verified according to industry standards and experimental data to ensure the balance between heat distribution and welding efficiency. Weld in an alternating sequence, i.e. first weld the first section, then jump to the last section, gradually move towards the middle to ensure heat dispersion and reduce the risk of local stress concentration. The system simulates heat accumulation through finite element analysis to calculate the temperature field distribution during welding of each section, and preferentially selects the path with heat-affected zone temperature below 300 degrees Celsius. After adjustment, the final path planning data is output, including coordinate points and welding sequence, providing support for subsequent parameter matching and execution scheme.

[0045] Step S5, match the welding parameters corresponding to the plate thickness and groove form from the welding parameter library, and output the welding execution scheme. Specifically, the selection of welding parameters needs to consider the characteristics of plate thickness, groove form and welding path, while the output of execution scheme needs to integrate all information to form a complete operation guide. Through this step, the system can ensure that the parameter settings of the welding process are highly matched with the actual needs, thereby improving the welding quality and efficiency.

[0046] Step S51, query the welding parameter library to determine the matching conditions according to the plate thickness and groove form. The system accesses the pre-constructed welding parameter library, which is based on the enterprise's internal 10-year historical welding data, refers to the "Steel Structure Welding Specification" and is constructed through experimental verification. The data covers plate thickness of 5 to 30 millimeters, and the groove forms include single-sided V-type, double-sided V-type, etc. The construction process is to collect historical data, combine standard specifications, and verify the effectiveness of the parameters through welding tests, including weld mechanical property testing and non-destructive testing. The update mechanism is to adjust every half year combined with new project data and standard revision. The matching conditions are that the plate thickness error is within 1 millimeter, the groove form is completely consistent, and the parameter combination that has been verified by the process is preferred to ensure the pertinence and accuracy of parameter selection.

[0047] Step S52, obtain the matched welding parameters, including welding current, welding voltage, welding speed, welding gun swing with swing width and angle, root gap specific value, and root gap specific value. Specifically, the system will extract the specific parameter values corresponding to the matching conditions from the welding parameter library, including welding current, voltage, welding speed, and welding swing, etc. For example, when processing the node corner weld of the steel structure tower, the system obtains the parameter combination of welding current 200 amperes, voltage 24 volts, welding speed 0.3 meters per minute, and swing 2 millimeters from the parameter library according to the conditions of plate thickness 8 millimeters and single-sided groove form. These parameters will be directly applied to the subsequent welding operation to ensure the stability of the welding process and the quality of the weld.

[0048] Step S53, integrate the welding parameters and the robot welding path to form a welding execution scheme. In one possible implementation, the system integrates the obtained welding parameters with the previously generated robot welding path to form a complete welding execution scheme. This scheme not only contains the coordinate points and sequence of the welding path, but also includes the welding parameter settings corresponding to each path segment. For example, when processing the tower foot base connection weld of the iron tower, the system will distinguish the welding path according to different plate thickness, different material and structure characteristics, etc., and assign specific current, voltage and speed parameters to each form of weld path to ensure the matching degree of parameters and path in the welding process. The final welding execution scheme is a detailed operation guide file that can be directly used for programming of the robot welding equipment.

[0049] Step S54, add real-time monitoring mechanism to the welding execution plan, and generate electronic process record. The system embeds a real-time monitoring mechanism in the welding execution plan, uses high-precision current and voltage sensors and a laser speedometer to monitor welding current, voltage and speed, and the data acquisition frequency is once every 10 seconds, which is based on ensuring the minimum period of capturing parameter fluctuations. If the current or voltage exceeds the preset threshold of ±5% or the speed deviation exceeds ±10%, the system automatically alarms and suspends welding, records abnormal data, and notifies the operator to check the equipment or adjust the parameters. At the same time, an electronic process record is generated, which records information such as weld position, groove form, welding parameters and path planning in detail, and is summarized into a file for easy traceability and optimization.

[0050] In one embodiment, for plate thickness detection and groove form selection in step S4, the system makes personalized adjustments according to the tower application scenario, such as a power transmission line tower. Different scenarios have differentiated needs for strength and efficiency, so the system uses an ultrasonic plate thickness sensor to detect plate thickness through ultrasonic reflection principles and adjusts it in combination with engineering strength and welding efficiency requirements. If the plate thickness is less than 5 millimeters, single-sided groove is preferred, with welding current controlled at 80 to 100 amperes and voltage at 18 to 22 volts; if the plate thickness is greater than 20 millimeters, double-sided groove is selected, with the welding pass increased to 3 to 5 times for layered filling of the weld to ensure strength. This quantitative adjustment method effectively adapts to different engineering requirements.

[0051] In the collision detection of step S44, multi-angle simulation analysis constructs a three-dimensional model of the tower node through three-dimensional simulation software, inputs robot arm parameters and obstacle position data, and simulates trajectories approaching the weld from different angles. The A-star algorithm takes the starting point and ending point as input, calculates the cost of each trajectory, and outputs the path with the lowest collision risk and the best welding quality by considering distance and obstacle avoidance. This process effectively improves the reliability of path planning, especially for complex scenarios.

[0052] In one embodiment, for deformation prevention adjustment in step S45, the system designs a welding sequence strategy according to the weld length, component shape and thermal deformation characteristics. The strategy development principle is to balance heat distribution and reduce stress concentration. The system first analyzes the weld length and material thermal conductivity coefficient, calculates the heat accumulation area, simulates the thermal stress distribution through finite element analysis software, uses ANSYS tools, sets material parameters such as thermal conductivity, specific heat capacity and expansion coefficient, and the meshing accuracy is 1 millimeter. The temperature field and stress field during welding are simulated to determine the welding starting point and path, and the area with fast heat diffusion is preferred for starting welding, and the welding time is controlled in segments to ensure that the temperature difference after each welding does not exceed 100 degrees Celsius, thereby reducing the risk of deformation.

[0053] In the welding parameter matching of step S5, the system optimizes parameter selection through historical data. Welding parameter data of past similar engineering projects, including plate thickness, groove form, welding current, voltage, and welding results, are extracted from the database as training data. A support vector machine algorithm is used for classification, with plate thickness, groove form, and welding environment temperature as input features, and parameter categories as output. A random forest algorithm is used for regression analysis, with classification results as input and parameter combination values as output, with a success rate of more than 90% of welding qualification rate as the evaluation standard. Combined with the current weld characteristics, the system automatically recommends the optimal parameters, improves applicability, reduces test costs, and ensures stable welding quality.

[0054] In one embodiment, for the welding parameters obtained in step S52, the system adjusts the parameters according to the welding equipment type. The adjustment is based on the working characteristics of the equipment and the weld quality requirements. For example, gas shielded welding equipment has concentrated heat input, welding voltage is increased by 5 to 10 volts, and swing parameters are increased by 10% to 15% to ensure weld formation. Submerged arc welding equipment requires high welding depth, welding speed is reduced by 20% to 30%, welding current is increased by 15% to 20%, to ensure the depth and strength of the weld. The adjustment process includes first analyzing the difference between the rated parameters of the equipment and the actual working conditions, then determining the adjustment range according to the weld quality standards, and finally automatically matching the parameters through the system algorithm. This way improves the adaptability of the parameters to the equipment and ensures smooth operation.

[0055] In one possible implementation, for the welding execution scheme formed in step S53, the system prioritizes according to the engineering project progress requirements. First, through the stress analysis of the overall structure of the tower, the welding path and parameters of the key load-bearing parts are prioritized to ensure welding quality. The priority of load-bearing parts is defined as the area with stress exceeding 80% of the design value. Second, based on the urgency of the construction period, parts that have a significant impact on subsequent processes, such as welds that affect more than 3 days of construction period, are prioritized. Finally, the welding scheme for secondary parts is arranged. This priority integration method based on stress importance and construction period requirements can effectively improve the management efficiency of complex engineering projects.

[0056] In one embodiment, for the real-time monitoring mechanism in step S54, the system expands the monitoring range to cover various environmental factors during welding, including environmental temperature, wind speed, and humidity. For example, in outdoor tower welding operations, the system records the impact of temperature below 5 degrees Celsius or wind speed exceeding 8 meters per second on welding parameters in real time, and prompts to reduce welding speed or increase current value when necessary to avoid welding defects. Monitoring is achieved through sensor integration to ensure accurate data and significantly improve welding adaptability and reliability in complex environments.

[0057] In a possible implementation, for the whole process of step S5, the system can also provide a switching function of multiple welding parameter libraries to adapt to different standards or specifications. For example, when processing the welding task of domestic power transmission tower, if the project follows the domestic standard, the system will call the parameter library based on the domestic welding specification to extract the corresponding current, voltage and speed parameters; if the project follows the international standard, the system will switch to the corresponding international parameter library to ensure the compliance of parameter selection. This parameter library switching function can effectively support cross-regional or cross-standard engineering projects and improve the versatility and flexibility of the system.

[0058] In an embodiment, for the combined application of steps S4 and S5, the system can further introduce a welding quality prediction function to evaluate the feasibility of the welding execution scheme in advance. For example, when processing the butt weld of the thick plate of the tower foot, the system will simulate the heat distribution and stress change in the welding process according to the generated welding path and matched welding parameters, predict the possible defects such as cracks or pores in the weld area, and prompt to adjust the path or parameters. This prediction function can effectively reduce the risk of welding failure and improve the reliability of the overall process, providing a strong guarantee for the smooth implementation of the engineering project.

[0059] In a possible implementation, for the path planning and deformation prevention adjustment in steps S44 and S45, the system can also design different path generation strategies according to the geometric complexity of the weld. For example, when processing the straight weld of the tower, the system will adopt a simple linear path planning combined with uniform welding speed to ensure the welding efficiency; while processing the curved weld of the tower foot, the system will generate a smooth curved path and adjust the welding speed and swing according to the curvature change to ensure the uniformity of the weld. This way of adjusting the path strategy according to the geometric characteristics can effectively adapt to the needs of different types of welds and improve the adaptability and precision of the welding path.

[0060] In an embodiment, for the parameter matching process in steps S51 and S52, the system can further introduce a multi-parameter combination optimization function to improve the comprehensive performance of parameter selection. For example, when processing the butt weld of the thick plate of the tower, the system will extract multiple parameter combinations that meet the conditions from the parameter library and perform comprehensive evaluation according to multiple indicators such as welding speed, weld strength and heat deformation control to select a parameter combination that balances between efficiency and quality. This multi-index optimized parameter selection method can effectively improve the overall performance of the welding process and meet the high standard requirements of complex engineering projects.

[0061] In a possible implementation, for the execution scheme integration and monitoring mechanism in steps S53 and S54, the system can also provide visual interface support to allow the operator to view and adjust the scheme content in real time. For example, when processing the connection weld of the tower foot base weldment of the iron tower, the system displays the three-dimensional view of the welding path and the parameter setting details in the visual interface, and allows the operator to manually adjust the welding sequence or parameter value of certain path segments according to the on-site situation. At the same time, the interface updates the monitoring data in real time during the welding process, such as current fluctuation and temperature change, to ensure that the operator can timely find and solve problems. This visual support function can significantly improve the practicability and operability of the welding execution scheme.

[0062] In an embodiment, for the overall output process of step S5, the system can generate welding execution scheme files in multiple formats to meet the needs of different devices and platforms. For example, the system generates a special format file suitable for robotic welding devices, such as a G code file, which contains path coordinates and parameter instructions; at the same time, a general CSV format file is generated, which records the weld position, parameter settings and monitoring requirements, to facilitate manual checking and archiving. This multi-format output method effectively supports scheme application in different working scenarios, improving the compatibility and convenience of the system.

[0063] In a possible implementation, for the robotic welding path generation in step S4, the system can also make path adaptation adjustments according to the specific model of the welding device. For example, when processing the tower foot weld of the iron tower, if the robotic device used has six-axis freedom, the system will fully utilize its flexibility to design a complex multi-angle welding path to cover all parts of the weld; if the device only has four-axis freedom, the system will simplify the path design to ensure that the path is executable within the device's capabilities. This way of adjusting the path according to the device model can effectively improve the matching degree of the welding path and the device, ensuring the smooth implementation of automated welding.

[0064] In an embodiment, for the electronic process record generation in step S5, the system can also add a quality traceability coding function to support subsequent engineering acceptance and maintenance work. For example, when processing the weld of the iron tower, the system generates a unique traceability code for each weld, and associates the code with the welding parameters, path data and monitoring records in the electronic process record. If a quality problem is found in a weld later, the relevant records can be quickly retrieved through the traceability code to analyze the problem. This traceability coding function can significantly improve the efficiency of welding quality management and provide reliable support for the long-term operation of engineering projects.

[0065] In a possible implementation, for the groove form selection and center line extraction in steps S42 and S43, the system can also be personalized according to the stress characteristics of the weld. For example, when processing the load-bearing weld of a tower, the system analyzes the stress of the weld area, and if it is found that the weld mainly bears tension, the double V groove form is preferentially selected, and the center line track is designed as a multi-pass welding path to increase the strength and durability of the weld. This way of adjusting the groove and path according to the stress characteristics can effectively improve the mechanical properties of the weld and meet the needs of high-strength engineering projects.

[0066] In an embodiment, for the welding parameter acquisition in step S52, the system can also correct the parameters according to the characteristics of the welding material. For example, when processing the low-alloy structural steel weld of a tower, the system appropriately reduces the welding current and speed parameters according to the high hardness and low toughness characteristics of the material to reduce the risk of welding heat cracks; and when processing the carbon structural steel weld, the standard parameter setting is used to ensure the welding efficiency. This way of correcting the parameters according to the material characteristics can effectively adapt to the welding needs of different steels and improve the stability of the weld quality.

[0067] In a possible implementation, for the real-time monitoring mechanism in step S54, the system can also add an automatic alarm function to improve the safety of the welding process. For example, when processing the tower foot node connection weld of a tower, the system sets the normal range of the welding current and voltage, and if the monitoring data exceeds the range, for example, the current suddenly rises above the preset value, the system automatically sends an alarm signal and suspends the welding operation for manual inspection. This automatic alarm function can timely detect abnormal conditions in the welding process and avoid welding defects caused by out-of-control parameters.

[0068] The above description is only the preferred embodiment of the present application and the explanation of the applied technical principles. Those skilled in the art should understand that the scope of the invention involved in the present application is not limited to the technical solutions formed by the specific combination of the above technical features, and should also cover other technical solutions formed by any combination of the above technical features or their equivalent features without departing from the concept of the present application. For example, the above features are replaced with the technical features disclosed in the present application (but not limited to) having similar functions to form technical solutions.

Claims

1. A method of welding a tower foot base, characterized by, include: The system acquires the two-dimensional planar graphic data of the component and generates the corresponding three-dimensional solid graphic through a geometric feature recognition algorithm. Based on the assembly positions of the two-dimensional planar graphic data, the three-dimensional solid graphics are assembled into a three-dimensional model, and gap areas are formed between the components, including T-joints between the plates. The gap area in the three-dimensional model is automatically identified as a weld, and the weld is beveled. The thickness of the weld is detected and the bevel type is adaptively selected based on the thickness, while a robot welding path is generated. Match welding parameters corresponding to plate thickness and bevel type from the welding parameter library, and output welding execution plan.

2. The method of welding a base pad of a tower foot as defined in claim 1, wherein, The process of acquiring two-dimensional planar graphic data of the component and generating corresponding three-dimensional solid graphics through a geometric feature recognition algorithm includes: Extract part outlines and assembly positioning information from engineering drawing formats as two-dimensional planar graphic data; The contour of the part is processed using a boundary extraction algorithm to obtain boundary feature data; The corresponding three-dimensional solid graphics are generated based on the boundary feature data and the topology reconstruction algorithm.

3. The method of welding a base pad of a tower foot as defined in claim 1, wherein, The assembly of three-dimensional solid graphics into a three-dimensional model based on the assembly positions of two-dimensional planar graphic data, and the formation of gap regions between components, wherein the gap regions include T-joints between plates, including: Identify the constraints in the assembly position based on two-dimensional planar graphic data, including alignment points and mating surfaces; The three-dimensional solid graphics are intelligently assembled using the aforementioned constraint relationships to generate a three-dimensional model; The gap distance between components in the three-dimensional model is detected, and the gap area including the T-joint of the plate to plate is marked.

4. The method of welding a base pad of a tower foot as defined in claim 1, wherein, The automatic identification of the gap region in the three-dimensional model as a weld and the beveling of the weld include: The three-dimensional model is scanned using a gap distance threshold algorithm, and the gap area is automatically marked as a weld. The weld joint types are classified, and appropriate beveling treatments are performed for each type of joint requiring beveling. Calculate the geometric parameters of the weld, including the bevel angle and bevel depth, and preliminarily determine the range of blunt edge thickness based on the weld joint type and the preset welding process specifications; The integrated deformation control rules optimize the bevel treatment, and the bevel depth is adjusted in combination with the initially determined blunt edge thickness range, and the processed weld data is output.

5. The method of welding a base pad of a tower foot as defined in claim 1, wherein, The process of detecting the plate thickness of the weld and adaptively selecting the bevel type based on the plate thickness, while simultaneously generating a robot welding path, includes: Extract the plate thickness value from the weld data. If the plate thickness value is less than or equal to the preset plate thickness threshold, select the single-sided bevel form; otherwise, select the double-sided bevel form. Based on the plate thickness value, the initially determined blunt edge thickness range is optimized, and the weld centerline is extracted as the basic trajectory according to the bevel form and the optimized blunt edge thickness range. Fine-tune the basic trajectory offset by combining the preset root gap parameters to ensure that the welding torch trajectory is aligned with the center of the root gap; The standard welding sequence is superimposed to generate the robot welding path, and collision detection is performed.

6. The method of welding a base pad of a tower foot as defined in claim 1, wherein, The process of matching welding parameters corresponding to plate thickness and bevel type from the welding parameter library and outputting a welding execution plan includes: Querying a preset welding parameter library, and determining a matching condition according to the plate thickness, the groove form, the optimized root face thickness range, and a preset root gap parameter; Obtaining the matched welding parameters, including welding current, welding voltage, welding speed, welding gun swing with swing width and angle, root face thickness specific value, and root gap specific value; Integrating the welding parameters and the robot welding path to form a welding execution scheme; Adding a real-time monitoring mechanism to the welding execution scheme.

7. The method of welding a base pad of a tower foot as defined in claim 2, wherein, The application boundary extraction algorithm processes the part contour to obtain boundary feature data, including: Image processing is performed on the part contour to identify an edge point set; Topology relationships including point adjacency, contour closure, and component groove edge identification correlation are constructed from the edge point set to obtain preliminary boundary feature data, wherein the preliminary boundary feature data contains contour types, boundary line segment attributes, and coordinate precision verification results; The preliminary boundary feature data and assembly positioning information are integrated, and data conflicts are eliminated through geometric consistency verification to obtain boundary feature data.

8. The method of welding a base pad of a tower foot as defined in claim 4, wherein, The gap distance threshold algorithm is used to scan the three-dimensional model to automatically mark the gap region as a weld, including: According to a preset gap distance threshold, scanning all component inter-region areas of the three-dimensional model; Determining whether the scanned region meets the gap distance threshold, and if so, marking it as the weld; Classifying the marked weld types.

Images