Multi-angle submerged arc welding method and system for large steel column

Abstract

The invention discloses a multi-angle submerged-arc welding method and system for a large steel column, and relates to the technical field of submerged-arc welding, the method comprises the following steps: reading a submerged-arc welding scheme configured for a target large steel column, and establishing a welding posture transformation time sequence; on the basis of determining a plurality of attitude transformation nodes, extracting a plurality of pieces of attitude transformation information in a submerged arc welding scheme; a time sequence of all welding postures is decomposed, and monitoring attention configuration and welding scheme updating configuration under the corresponding postures are carried out; welding control is conducted through the welding scheme updating result, and abnormity monitoring is conducted through the monitoring attention configuration result synchronously; and controlling and monitoring the attitude change. The technical problems that in the prior art, during multi-angle submerged-arc welding of the large steel column, the control precision is insufficient, defects are difficult to monitor, and consequently the welding quality and stability are poor are solved, and the technical effects that accurate control and defect monitoring of multi-angle submerged-arc welding of the large steel column are achieved, and the welding quality and stability are improved are achieved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of submerged arc welding, in particular to a multi-angle submerged arc welding method and system for large steel columns. BACKGROUND

[0002] In the welding processing of large steel columns, submerged arc welding is widely used due to its high efficiency and stable quality. However, due to the large structure of large steel columns and the complex welding position, various angle welding operations such as flat welding, horizontal welding, vertical welding and overhead welding are required. The traditional submerged arc welding method has difficulty in accurately controlling the transformation of the welding posture, and does not adequately consider the influence of gravity on the molten pool under different postures, which can easily lead to welding defects. At the same time, there is a lack of targeted monitoring and compensation mechanism during the welding process, which makes it difficult to discover and compensate for welding abnormalities in real time, affecting the welding quality and stability.

[0003] The prior art has the technical problem of poor welding quality and stability due to insufficient control accuracy and difficulty in monitoring defects during multi-angle submerged arc welding of large steel columns. SUMMARY

[0004] The present application provides a multi-angle submerged arc welding method and system for large steel columns to solve the technical problem of poor welding quality and stability due to insufficient control accuracy and difficulty in monitoring defects during multi-angle submerged arc welding of large steel columns in the prior art.

[0005] In view of the above problems, the present application provides a multi-angle submerged arc welding method and system for large steel columns.

[0006] In a first aspect of the present application, a multi-angle submerged arc welding method for large steel columns is provided, the method comprising: reading a submerged arc welding scheme configured for a target large steel column, analyzing the welding posture, and establishing a welding posture transformation time sequence; determining a plurality of posture transformation nodes based on the welding posture transformation time sequence, and extracting a plurality of posture transformation information in the submerged arc welding scheme; decomposing each welding posture time sequence in the welding posture transformation time sequence, and performing monitoring attention configuration and welding scheme update configuration under the corresponding posture; in each welding posture time sequence, using the welding scheme update result for welding control, and simultaneously using the monitoring attention configuration result for abnormal monitoring; at the plurality of posture transformation nodes, controlling and monitoring the posture transformation according to the plurality of posture transformation information.

[0007] In a second aspect of the present application, a multi-angle submerged arc welding system for large steel columns is provided, the system comprising: The timing establishment module is used for reading the submerged arc welding scheme configured for the target large steel column, performing welding posture analysis, and establishing a welding posture transformation timing; the posture transformation information extraction module is used for determining a plurality of posture transformation nodes based on the welding posture transformation timing, and extracting a plurality of posture transformation information in the submerged arc welding scheme; the update configuration module is used for decomposing each welding posture timing in the welding posture transformation timing, and performing monitoring attention configuration and welding scheme update configuration under the corresponding posture; the abnormality monitoring module is used for performing welding control by using the welding scheme update result in each welding posture timing, and simultaneously performing abnormality monitoring by using the monitoring attention configuration result; and the posture transformation monitoring module is used for performing posture transformation control and monitoring according to the plurality of posture transformation information in the plurality of posture transformation nodes.

[0008] The one or more technical solutions provided in the present application have at least the following technical effects or advantages: The timing establishment module is used for reading the submerged arc welding scheme configured for the target large steel column, performing welding posture analysis, and establishing a welding posture transformation timing; the posture transformation information extraction module is used for determining a plurality of posture transformation nodes based on the welding posture transformation timing, and extracting a plurality of posture transformation information in the submerged arc welding scheme; the update configuration module is used for decomposing each welding posture timing in the welding posture transformation timing, and performing monitoring attention configuration and welding scheme update configuration under the corresponding posture; the abnormality monitoring module is used for performing welding control by using the welding scheme update result in each welding posture timing, and simultaneously performing abnormality monitoring by using the monitoring attention configuration result; and the posture transformation monitoring module is used for performing posture transformation control and monitoring according to the plurality of posture transformation information in the plurality of posture transformation nodes. BRIEF DESCRIPTION OF DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0010] Figure 1 A flowchart of the multi-angle submerged arc welding method for large steel columns provided by the embodiments of the present application is shown in the figure. Figure 2 A structural schematic diagram of the multi-angle submerged arc welding system for large steel columns provided by the embodiments of the present application is shown in the figure.

[0011] Label explanation: timing establishment module 10, posture transformation information extraction module 20, update configuration module 30, abnormality monitoring module 40, posture transformation monitoring module 50. DETAILED DESCRIPTION

[0012] This application provides a method and system for multi-angle submerged arc welding of large steel columns, which addresses the technical problems of insufficient control precision and difficulty in defect monitoring during multi-angle submerged arc welding of large steel columns in the prior art, resulting in poor welding quality and stability.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] Example 1, as Figure 1 As shown, this application provides a multi-angle submerged arc welding method for large steel columns, the method comprising: Step S100: Read the submerged arc welding scheme configured for the target large steel column, analyze the welding posture, and establish the welding posture transformation sequence.

[0015] Specifically, the submerged arc welding scheme specifically configured for the target large steel column is first read, and the welding postures involved in the scheme are analyzed to identify specific welding posture types such as flat welding, horizontal welding, vertical welding, and overhead welding. Then, the transformation sequence and time logic of different welding postures in the welding process are sorted out to establish a complete welding posture transformation time sequence. This time sequence will serve as the basis for subsequently determining posture transformation nodes, extracting posture transformation information, and carrying out welding control at each stage.

[0016] Step S200: Determine several attitude transformation nodes based on the welding attitude transformation timing, and extract several attitude transformation information from the submerged arc welding scheme.

[0017] Specifically, based on the established welding posture transformation sequence, several key time points, i.e., several posture transformation nodes, are located when welding postures (such as flat welding, horizontal welding, vertical welding, and overhead welding) change during the welding process. At the same time, several posture transformation information corresponding to these posture transformation nodes are extracted from the submerged arc welding scheme configured for the target large steel column. Each posture transformation information includes at least the steel column posture transformation information and the welding torch posture transformation information, providing a basis for subsequent posture control and monitoring at the transformation nodes.

[0018] Step S300: Decompose each welding posture timing sequence in the welding posture change timing sequence, and perform monitoring attention configuration and welding scheme update configuration under the corresponding posture.

[0019] Specifically, in the established welding posture transformation time sequence, the overall time sequence is decomposed into independent welding posture time sequences corresponding to flat welding, horizontal welding, vertical welding, and overhead welding. For each decomposed welding posture time sequence, the weld direction and weld surface are extracted first, and the flow characteristics of the molten pool metal under the influence of gravity are analyzed to establish the molten pool-gravity influence relationship. Then, the corresponding welding parameters are extracted from the submerged arc welding scheme, and the molten pool flow characteristics are determined based on the two. In addition, welding defects are identified and predicted, and these are used as monitoring targets to complete the monitoring attention configuration. At the same time, the defect compensation plan library is called to perform compensation decision matching, and the welding scheme is updated and configured according to the matching results.

[0020] Step S400: At each welding posture sequence, welding control is performed using the welding scheme update results, and anomaly monitoring is performed simultaneously using the monitoring attention configuration results.

[0021] Specifically, in the welding stages corresponding to the various welding postures such as flat welding, horizontal welding, vertical welding, and overhead welding, the relevant parameters in the welding process are adjusted in real time to implement the welding operation based on the updated welding scheme obtained by matching through the defect compensation plan library. At the same time, in accordance with the monitoring requirements set in the monitoring attention configuration with the goal of predicting welding defects, the actual welding defects are monitored in the welding process. If a defect is detected, the consistency index between the attributes of the real-time welding defect and the predicted welding defect is analyzed. When the index is greater than the preset threshold, the failure warning of the defect compensation plan library is executed.

[0022] Step S500: At the plurality of attitude transformation nodes, attitude transformation is controlled and monitored based on the plurality of attitude transformation information.

[0023] Specifically, at several determined attitude transformation nodes, based on the extracted attitude transformation information including the attitude transformation information of the steel column and the welding torch, the attitudes of the steel column and the welding torch are precisely controlled to complete the attitude transformation, and the transformation process is monitored in real time. After the attitude transformation is controlled and monitored, the stability of the molten pool is also monitored at these nodes and a molten pool stability index is generated. When the index is less than the preset index, the attitude transformation is delayed and continuously monitored to ensure the stability and reliability of the attitude transformation process.

[0024] In one possible implementation, step S200 further includes: Step S210: Any of the several attitude transformation information includes at least the steel column attitude transformation information and the welding torch attitude transformation information.

[0025] Specifically, among the extracted attitude transformation information, each piece of information encompasses at least two core aspects: first, the steel column attitude transformation information, involving the specific adjustment parameters and execution logic of the large steel column's position and angle at the attitude transformation node; second, the welding torch attitude transformation information, including details of the welding torch's attitude adjustment relative to the steel column, such as the welding torch's tilt angle and its relative position to the weld. These two types of information together constitute the complete basis for attitude transformation, ensuring that the coordinated adjustment of the steel column and welding torch can accurately match the welding requirements during welding attitude transitions.

[0026] In one possible implementation, step S200 further includes: Step S220: Welding postures include at least flat welding, horizontal welding, vertical welding, and overhead welding.

[0027] Specifically, the range of welding postures includes at least four basic and crucial welding postures: flat welding, horizontal welding, vertical welding, and overhead welding. Flat welding refers to welding in a horizontal position; horizontal welding is welding performed laterally on a vertical or inclined plane; vertical welding is welding along a vertical direction; and overhead welding is welding performed below the workpiece. These postures cover a variety of angle scenarios that may be encountered during the welding of large steel columns, providing a clear foundation for subsequent operations such as configuring welding schemes and controlling posture changes based on different postures.

[0028] In one possible implementation, step S300 further includes: Step S310: Analyze the influence of gravity on the molten pool for each welding posture and establish the influence relationship between each molten pool and gravity.

[0029] Step S320: Extract the welding parameters corresponding to each welding posture timing in the submerged arc welding scheme.

[0030] Step S330: Based on the relationship between each molten pool and gravity, and the welding parameters, perform welding defect analysis and compensation for each welding posture sequence under gravity influence, configure monitoring based on welding defects, and update the welding scheme configuration based on the compensation results.

[0031] Specifically, for the decomposed welding posture sequence such as flat welding, horizontal welding, vertical welding, and overhead welding, the weld direction and weld surface information under each posture are first extracted. Then, based on this information, the influence of gravity on the flow of molten pool metal in different welding postures is analyzed. For example, the vertical effect of gravity on the molten pool in flat welding, the lateral flow trend of the molten pool caused by gravity in horizontal welding, the difference in vertical flow of the molten pool caused by gravity in vertical welding, and the obstruction effect of gravity on the formation of the molten pool in overhead welding, etc., are then established to establish the molten pool-gravity influence relationship corresponding to each welding posture sequence, clearly reflecting the correlation between gravity and molten pool state under different postures.

[0032] Welding parameters corresponding to various welding postures such as flat welding, horizontal welding, vertical welding, and overhead welding were extracted and decomposed from the submerged arc welding scheme configured for the target large steel column. These parameters cover key data such as current, voltage, welding speed, and wire extension length required for the welding process under each posture, providing basic parameter support for subsequent twin simulation, defect analysis, and compensation configuration based on the influence relationship between the molten pool and gravity.

[0033] Based on the molten pool-gravity influence relationship corresponding to various welding postures such as flat welding, horizontal welding, vertical welding, and overhead welding, and the welding parameters corresponding to each posture time sequence extracted from the submerged arc welding scheme, the molten pool flow characteristics under each posture are determined through twin simulation. These molten pool flow characteristics are combined with the corresponding weld surface information and input into a trained defect recognizer (which is trained based on defect recognition samples based on molten pool flow characteristics and weld surface) to perform welding defect analysis to generate predicted welding defects. These predicted welding defects are then used as monitoring targets to complete the monitoring attention configuration. At the same time, based on the predicted welding defects, the defect compensation plan library is called to perform compensation decision matching, and the welding scheme is updated and configured according to the compensation results obtained from the matching.

[0034] In one possible implementation, step S310 further includes: Step S311: Extract the weld direction and weld surface for each welding posture sequence.

[0035] Step S312: Analyze the flow characteristics of the molten pool metal under the influence of gravity based on the weld direction and weld surface, and generate the respective molten pool-gravity influence relationships.

[0036] Specifically, for each welding posture sequence decomposed from the welding posture change sequence, such as flat welding, horizontal welding, vertical welding, and overhead welding, the weld direction and weld surface are extracted for the corresponding welding area of ​​each sequence. The weld direction includes spatial orientation such as horizontal, vertical, and inclined, and the weld surface covers its planar shape, tilt angle, and other specific features. This clarifies the specific shape and orientation information of the weld in space under different welding postures, laying the foundation for subsequent analysis of the influence of gravity on the molten pool.

[0037] First, weld seam direction is quantified into spatial vector features under different welding postures such as flat welding, horizontal welding, vertical welding, and overhead welding. Weld seam surface is quantified into planar parameter features, along with corresponding sample data of molten pool metal flow characteristics (such as flow velocity, trajectory, distribution, etc.) to construct a training dataset. Then, a regression model (such as a random forest regression model or a neural network model) is trained using these samples. The quantified features of weld seam direction and weld seam surface are used as input, and the molten pool metal flow characteristic parameters are used as output. The model learns the mapping relationship between weld seam features and molten pool flow characteristics under different postures. After training, the weld seam direction and weld seam surface features corresponding to the welding posture time series to be analyzed are input into the model to obtain the prediction results of molten pool metal flow characteristics under the influence of gravity, thereby generating the molten pool-gravity influence relationship corresponding to each welding posture time series.

[0038] In one possible implementation, step S330 further includes: Step S331: Perform twin simulation based on the welding parameters and the molten pool-gravity influence relationship to determine the flow characteristics of each molten pool.

[0039] Step S332: Identify and compensate for welding defects based on the flow characteristics of each molten pool, and complete the welding scheme update configuration.

[0040] Step S333: Using the welding defect as the monitoring target, complete the monitoring attention configuration.

[0041] Specifically, for various welding postures such as flat welding, horizontal welding, vertical welding, and overhead welding, the corresponding welding parameters (such as current, voltage, and welding speed) extracted from the submerged arc welding scheme, along with the established molten pool-gravity influence relationships (reflecting the influence of gravity on the molten pool under different postures), are used as input conditions for digital twin simulation. By constructing a virtual model consistent with the actual welding scenario, the dynamic flow process of the molten pool metal under the influence of welding parameters and gravity is simulated, thereby determining the molten pool flow characteristics under each welding posture, including key information such as the flow velocity, flow trajectory, and distribution state of the molten pool metal.

[0042] For the molten pool flow characteristics determined under various welding postures such as flat welding, horizontal welding, vertical welding, and overhead welding, a defect recognizer trained based on the molten pool flow characteristics and defect recognition samples collected from the weld surface is used. The molten pool flow characteristics and corresponding weld surface information are combined and input into the defect recognizer to perform welding defect analysis and generate predicted welding defects. Based on the generated predicted welding defects, a defect compensation plan library is called to perform compensation decision matching. According to the compensation decision obtained by matching, the corresponding welding parameters and other contents in the original submerged arc welding scheme are adjusted to complete the updated configuration of the welding scheme.

[0043] Welding defects identified through molten pool flow characteristics (such as defects that may occur in different postures, including flat welding, horizontal welding, vertical welding, and overhead welding) are used as monitoring targets. The types of defects, characteristic parameters, and monitoring frequencies that need to be monitored in each welding posture sequence are clearly defined. For example, for defects such as poor molten pool formation that may be caused by gravity during overhead welding, corresponding image monitoring thresholds and real-time analysis cycles are set to complete the monitoring attention configuration and provide a specific basis for accurate anomaly monitoring in the subsequent welding process.

[0044] In one possible implementation, step S332 further includes: For flat welding, horizontal welding, vertical welding, and overhead welding, defect identification samples based on molten pool flow characteristics and weld surface are collected to train the defect identifyer.

[0045] The flow characteristics of each molten pool are combined with the corresponding weld surface information and input into the defect identifier for defect analysis to generate predicted welding defects.

[0046] Based on the predicted welding defects, the defect compensation plan library is invoked to perform compensation decision matching and complete the welding scheme update configuration.

[0047] Specifically, for four welding postures—flat welding, horizontal welding, vertical welding, and overhead welding—a defect identifier is trained using a combined algorithm of Convolutional Neural Network (CNN) and Support Vector Machine (SVM). First, defect samples containing molten pool flow characteristics (such as velocity field distribution and eddy current morphology) and weld surface information (such as three-dimensional contour parameters) are collected for each posture. These samples are then labeled (e.g., defect types such as incomplete penetration, porosity, and weld beads). Next, CNN is used to extract features from the molten pool flow characteristic images, obtaining high-dimensional feature vectors. Simultaneously, the weld surface parameters are quantized and fused with the feature vectors extracted by the CNN to form a comprehensive feature set. Then, the comprehensive feature set is input into the SVM classifier for training. By adjusting the kernel function parameters, the model is optimized to learn the mapping relationship between molten pool flow characteristics, weld surface information, and welding defects under different postures. Finally, the resulting defect identifier can classify the input combination of molten pool flow characteristics and weld surface information, achieving accurate identification of welding defects.

[0048] The flow characteristics of the molten pool (such as the flow velocity, trajectory, and distribution of the molten metal) determined through twin simulation under various welding postures such as flat welding, horizontal welding, vertical welding, and overhead welding are combined with the corresponding extracted weld surface information (such as the planar shape and tilt angle of the weld surface) to form comprehensive data containing the characteristics of both. Then, this comprehensive data is input into a pre-trained defect identifyer (which is trained based on the molten pool flow characteristics and weld surface defect samples under the above four postures). The identifyer analyzes the comprehensive data according to the learned defect identification rules, determines the type, location, and severity of welding defects that may occur under the current welding posture, and finally generates predicted welding defects.

[0049] Based on the predicted welding defects (such as incomplete penetration, weld beads, porosity, etc.) generated by the defect identifier under different welding postures such as flat welding, horizontal welding, vertical welding, and overhead welding, the system automatically retrieves a preset defect compensation plan library. This plan library stores compensation strategies corresponding to various welding defects, including adjustment schemes for parameters such as welding current, voltage, and welding speed, as well as fine-tuning suggestions for welding torch posture. By matching the type and characteristics of the predicted welding defects with the plans in the plan library, the optimal compensation decision is selected. Finally, based on the compensation decision, the welding parameters and operation procedures for the corresponding postures in the original submerged arc welding scheme are updated, thereby completing the updated configuration of the welding scheme to effectively prevent and compensate for the predicted welding defects.

[0050] In one possible implementation, step S500 further includes: Step S510: Monitor the stability of the molten pool at the several attitude transformation nodes and generate molten pool stability indicators.

[0051] Step S520: When the stability index of the molten pool is less than the preset index, delay control and continuous monitoring of attitude change are performed.

[0052] Specifically, multidimensional data such as temperature field distribution, shape parameters (e.g., length, width, and depth of the molten pool), and metal flow velocity are collected in real time at several attitude transformation nodes and standardized. Then, principal component analysis (PCA) is used to reduce the dimensionality of these multidimensional data and extract key feature vectors reflecting the stability of the molten pool. Next, a fuzzy comprehensive evaluation algorithm is used, combined with preset feature weights (e.g., temperature uniformity weight, shape stability weight, and flow stability weight), to comprehensively score the key feature vectors. Finally, the comprehensive score is mapped to a molten pool stability index of 0-100, where a higher score indicates better molten pool stability.

[0053] When the molten pool stability index generated by monitoring is lower than the preset standard index, it indicates that the molten pool state has not yet met the stability requirements for attitude change. In order to avoid welding quality problems caused by attitude change, the attitude change delay control is automatically triggered, and the original attitude change operation is suspended. At the same time, the temperature, shape, metal flow and other states of the molten pool are continuously monitored in real time until the molten pool stability index rises back to the preset index or above, and then the attitude change process is resumed to ensure the stability of the welding process and the welding quality.

[0054] In one possible implementation, step S400 further includes: Step S410: Monitor whether welding defects occur according to the monitoring attention configuration. If so, analyze the consistency index of the attributes of real-time welding defects and predicted welding defects.

[0055] Step S420: If the attribute consistency index is greater than the preset threshold, execute the failure warning of the defect compensation plan library.

[0056] Specifically, based on the welding defect monitoring indicators (such as defect type, size threshold, etc.) set in the monitoring configuration, high-definition industrial cameras, infrared temperature sensors, and other equipment are used to collect real-time data on the weld area during flat welding, horizontal welding, vertical welding, and overhead welding processes, acquiring monitoring data such as images and temperature. Then, the collected data is processed through image processing algorithms and data analysis models to determine whether welding defects have occurred. If a real-time welding defect is detected, the attribute information of the defect (such as type, location coordinates, size parameters, morphological features, etc.) is extracted, and the attribute information of the previously generated predicted welding defects is retrieved. Finally, the attribute information of the two types of defects is compared and calculated using a cosine similarity algorithm or a weighted scoring method to obtain an attribute consistency index reflecting the degree of consistency between the two.

[0057] After calculating the consistency index between the attributes of real-time welding defects and predicted welding defects, the index is compared with a preset threshold. If the consistency index is greater than the preset threshold, it indicates that the difference between the two in terms of attributes is large. This means that the compensation decision in the defect compensation plan library that matches the predicted welding defects may not be able to effectively deal with the actual real-time welding defects, and the plan library has a risk of failure. At this time, the failure warning of the defect compensation plan library is immediately executed. By issuing alarm signals and displaying warning information on the control terminal, relevant personnel are prompted to check, evaluate and update the defect compensation plan library to ensure the effectiveness of compensation decisions in subsequent welding processes.

[0058] Example 2, based on the same inventive concept as the multi-angle submerged arc welding method for large steel columns in the foregoing examples, such as... Figure 2As shown, this application provides a multi-angle submerged arc welding system for large steel columns. The system and method embodiments in this application are based on the same inventive concept. The system includes: The timing establishment module 10 is used to read the submerged arc welding scheme configured for the target large steel column, analyze the welding posture, and establish the welding posture transformation timing.

[0059] The attitude transformation information extraction module 20 is used to determine several attitude transformation nodes based on the welding attitude transformation timing and extract several attitude transformation information from the submerged arc welding scheme.

[0060] The update configuration module 30 is used to decompose each welding posture timing sequence in the welding posture change timing sequence, and to perform monitoring attention configuration and welding scheme update configuration under the corresponding posture.

[0061] The anomaly monitoring module 40 is used to perform welding control by utilizing the welding scheme update results at each welding posture time sequence, and to simultaneously perform anomaly monitoring by utilizing the monitoring attention configuration results.

[0062] The attitude change monitoring module 50 is used to control and monitor attitude changes based on the attitude change information at the plurality of attitude change nodes.

[0063] Furthermore, the system is also used to implement the following functions: Any of the several attitude transformation information includes at least the steel column attitude transformation information and the welding torch attitude transformation information.

[0064] Furthermore, the system is also used to implement the following functions: Welding postures include at least flat welding, horizontal welding, vertical welding, and overhead welding.

[0065] Furthermore, the system is also used to implement the following functions: The influence of gravity on the molten pool is analyzed for each welding posture timing, and the influence relationship between each molten pool and gravity is established. The welding parameters corresponding to each welding posture timing are extracted from the submerged arc welding scheme. Based on the influence relationship between each molten pool and gravity and the welding parameters, the welding defects under the influence of gravity are analyzed and compensated for for each welding posture timing. The monitoring and attention configuration is carried out according to the welding defects, and the welding scheme is updated and configured according to the compensation results.

[0066] Furthermore, the system is also used to implement the following functions: The weld direction and weld surface are extracted for each welding posture sequence; the flow characteristics of the molten pool metal under the influence of gravity are analyzed based on the weld direction and weld surface, and the molten pool-gravity influence relationship is generated for each weld.

[0067] Furthermore, the system is also used to implement the following functions: Based on the welding parameters and the molten pool-gravity influence relationship, a twin simulation is performed to determine the flow characteristics of each molten pool; welding defects are identified and compensated based on the flow characteristics of each molten pool, and the welding scheme is updated and configured; the welding defects are used as monitoring targets to complete the monitoring and attention configuration.

[0068] Furthermore, the system is also used to implement the following functions: For flat welding, horizontal welding, vertical welding, and overhead welding, defect identification samples based on molten pool flow characteristics and weld surface are collected to train a defect identifier; the molten pool flow characteristics and corresponding weld surface information are combined and input into the defect identifier for defect analysis to generate predicted welding defects; based on the predicted welding defects, the defect compensation plan library is called to perform compensation decision matching and complete the welding scheme update configuration.

[0069] Furthermore, the system is also used to implement the following functions: Melt pool stability is monitored at several attitude transformation nodes to generate a melt pool stability index. When the melt pool stability index is less than a preset index, attitude transformation delay control and continuous monitoring are performed.

[0070] Furthermore, the system is also used to implement the following functions: According to the monitoring configuration, monitor whether welding defects occur. If so, analyze the consistency index between the attributes of real-time welding defects and predicted welding defects. If the attribute consistency index is greater than a preset threshold, execute the failure warning of the defect compensation plan library.

[0071] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0072] The above description is only a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.

[0073] This specification and accompanying drawings are merely illustrative examples of this application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Clearly, those skilled in the art can make various alterations and modifications to this application without departing from its scope. Therefore, if such modifications and variations fall within the scope of this application and its equivalents, this application intends to include such modifications and variations.

Claims

1. A multi-angle submerged arc welding method for large steel columns, characterized in that, include: Read the submerged arc welding scheme configured for the target large steel column, analyze the welding posture, and establish the welding posture transformation timing. Based on the welding posture transformation timing, several posture transformation nodes are determined, and several posture transformation information is extracted from the submerged arc welding scheme; The welding posture transformation time sequence is decomposed into each welding posture time sequence, and the monitoring attention configuration and welding scheme update configuration are performed under the corresponding posture. At each welding posture sequence, welding control is performed using the welding scheme update results, and anomaly monitoring is performed simultaneously using the monitoring attention configuration results. At the aforementioned attitude transformation nodes, attitude transformation is controlled and monitored based on the aforementioned attitude transformation information.

2. The multi-angle submerged arc welding method for large steel columns as described in claim 1, characterized in that, Any of the several attitude transformation information includes at least the steel column attitude transformation information and the welding torch attitude transformation information.

3. The multi-angle submerged arc welding method for large steel columns as described in claim 1, characterized in that, Welding postures include at least flat welding, horizontal welding, vertical welding, and overhead welding.

4. The multi-angle submerged arc welding method for large steel columns as described in claim 2, characterized in that, The welding posture change sequence is decomposed into individual welding posture sequences, and monitoring and attention configurations and welding scheme update configurations are performed for the corresponding postures, including: The influence of gravity on the molten pool is analyzed for each welding posture, and the influence relationship between each molten pool and gravity is established. Extract the welding parameters corresponding to each welding posture timing in the submerged arc welding scheme; Based on the relationship between each molten pool and gravity, and the welding parameters, we perform welding defect analysis and compensation under gravity for each welding posture sequence. We then configure monitoring based on the welding defects and update the welding scheme configuration based on the compensation results.

5. The multi-angle submerged arc welding method for large steel columns as described in claim 4, characterized in that, The influence of gravity on the molten pool is analyzed for each welding posture, and the influence relationship between each molten pool and gravity is established, including: The weld direction and weld surface are extracted for each welding posture sequence; Based on the analysis of the weld direction and weld surface, the flow characteristics of the molten pool metal under the influence of gravity are analyzed, and the respective molten pool-gravity influence relationships are generated.

6. The multi-angle submerged arc welding method for large steel columns as described in claim 4, characterized in that, Based on the aforementioned molten pool-gravity influence relationship and the aforementioned welding parameters, welding defect analysis and compensation are performed on the timing of the aforementioned welding postures under the influence of gravity. Monitoring and attention configurations are implemented based on the welding defects, and the welding scheme is updated and configured based on the compensation results, including: Based on the welding parameters and the molten pool-gravity influence relationship, twin simulations are performed to determine the flow characteristics of each molten pool. Welding defects are identified and compensated based on the flow characteristics of each molten pool, and the welding scheme is updated and configured. Using the aforementioned welding defects as the monitoring target, a monitoring configuration is completed.

7. The multi-angle submerged arc welding method for large steel columns as described in claim 6, characterized in that, Welding defects are identified and compensated based on the flow characteristics of each molten pool, and the welding scheme is updated and configured, including: For flat welding, horizontal welding, vertical welding, and overhead welding, defect identification samples based on molten pool flow characteristics and weld surface are collected to train the defect identifyer; The flow characteristics of each molten pool are combined with the corresponding weld surface information and input into the defect identifier for defect analysis to generate predicted welding defects. Based on the predicted welding defects, the defect compensation plan library is invoked to perform compensation decision matching and complete the welding scheme update configuration.

8. The multi-angle submerged arc welding method for large steel columns as described in claim 1, characterized in that, After controlling and monitoring attitude changes based on the attitude change information at the plurality of attitude change nodes, the method further includes: Molten pool stability is monitored at the aforementioned attitude transformation nodes to generate molten pool stability indices. When the stability index of the molten pool is less than the preset index, delay control and continuous monitoring of attitude change are performed.

9. The multi-angle submerged arc welding method for large steel columns as described in claim 7, characterized in that, Utilize monitoring attention configuration results for anomaly monitoring, including: According to the monitoring configuration, monitor whether welding defects occur. If so, analyze the consistency index of the attributes of real-time welding defects and predicted welding defects. If the attribute consistency index is greater than the preset threshold, the failure warning of the defect compensation plan library will be executed.

10. A multi-angle submerged arc welding system for large steel columns, characterized in that, The system is used to implement the multi-angle submerged arc welding method for large steel columns according to any one of claims 1-9, the system comprising: The timing establishment module is used to read the submerged arc welding scheme configured for the target large steel column, analyze the welding posture, and establish the welding posture transformation timing. The attitude transformation information extraction module is used to determine several attitude transformation nodes based on the welding attitude transformation timing and extract several attitude transformation information from the submerged arc welding scheme. The configuration update module is used to decompose each welding posture timing sequence in the welding posture change timing sequence, and to perform monitoring attention configuration and welding scheme update configuration under the corresponding posture. The anomaly monitoring module is used to control welding by utilizing the welding scheme update results at various welding postures and simultaneously monitor anomalies by utilizing the monitoring attention configuration results. The attitude change monitoring module is used to control and monitor attitude changes based on the attitude change information at the plurality of attitude change nodes.

Images