Intelligent welding method and platform for reinforcing ring of LNG storage tank
Through the intelligent welding device, the rotation module, lifting module and telescopic module, combined with the welding quality prediction model, the problem of insufficient accuracy and strength in the welding ring welding of LNG storage tanks is solved, and the welding efficiency and stability are improved.
Patent Information
- Application Number
- CN202411481385.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-23
- Publication Date
- 2024-11-22
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
During the welding process of LNG storage tank reinforcement ring, there are problems such as insufficient welding accuracy and non-compliance with the welding strength, which is difficult to meet the efficient and safe production needs of modern large storage tanks.
Intelligent welding devices are adopted, including rotation modules, lift modules and telescopic modules. By collecting welding parameters, the embedded welding quality prediction model is used to dynamically adjust the execution parameters and rotation speed of the welding head to ensure the accuracy of the weld quality.
It improves welding efficiency and the stability of weld quality, solves the problem of welding accuracy and strength not meeting standards, and achieves a more efficient welding process.
Smart Images

Figure CN118989830A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent welding, and in particular to an intelligent welding method and platform for a reinforcement ring of an LNG storage tank. Background Art
[0002] With the rapid development of industrial manufacturing and storage and transportation technology, liquefied natural gas storage tanks have been widely used in the field of energy storage and transportation. However, with the increase in the volume of storage tanks and the complexity of manufacturing processes, the welding quality and efficiency requirements of LNG storage tanks have become increasingly higher. Especially in the structure of LNG storage tanks, the reinforcement ring is a key load-bearing component, and its welding process directly affects the overall safety and stability of the tank. The traditional welding method of LNG tank reinforcement rings relies on manual or semi-automatic welding processes, and has problems such as insufficient welding accuracy and substandard welding strength, which makes it difficult to meet the efficient and safe production needs of modern large-scale LNG storage tanks. Summary of the invention
[0003] The present application provides an intelligent welding method and platform for reinforcement rings of LNG storage tanks, which are used to solve the technical problems of insufficient welding accuracy and substandard welding strength when welding reinforcement rings of LNG storage tanks in the prior art.
[0004] In view of the above problems, the present application provides an intelligent welding method and platform for the reinforcement ring of an LNG storage tank.
[0005] In a first aspect of the present application, a smart welding method for a reinforcement ring of an LNG storage tank is provided, the method comprising: Connect a welding device, the welding device including a rotating module, a lifting module, a telescopic module and at least one welding head; determine the reinforcement ring to be welded of the LNG storage tank, and collect welding parameters of the web and the wing plate in the reinforcement ring to be welded, including welding strength and weld position; place the reinforcement ring to be welded on the welding device, and the lifting module and the telescopic module align and position the welding head and the reinforcement ring to be welded according to the weld position, and when the alignment and positioning are completed, configure the execution parameters of the welding head according to the welding strength; the rotation module predicts the execution parameters according to the embedded welding quality prediction model, obtains matching rotation parameters that meet the preset welding quality, and controls the rotation module according to the matching rotation parameters.
[0006] The second aspect of the present application provides an intelligent welding platform for reinforcing rings of LNG storage tanks, the platform comprising: A welding device connection unit, wherein the welding device connection unit is connected to a welding device, wherein the welding device includes a rotating module, a lifting module, a telescopic module and at least one welding head; a welding parameter acquisition unit, wherein the welding parameter acquisition unit determines the reinforcement ring to be welded of the LNG storage tank, and acquires welding parameters of the web and the wing plate in the reinforcement ring to be welded, including welding strength and weld position; an execution parameter configuration unit, wherein the execution parameter configuration unit places the reinforcement ring to be welded on the welding device, and the lifting module and the telescopic module align and position the welding head and the reinforcement ring to be welded according to the weld position, and when the alignment and positioning are completed, configures the execution parameters of the welding head according to the welding strength; a rotation parameter determination unit, wherein the rotation parameter determination unit predicts the execution parameters according to the embedded welding quality prediction model through the rotation module, obtains matching rotation parameters that meet the preset welding quality, and controls the rotation module according to the matching rotation parameters.
[0007] One or more technical solutions provided in this application have at least the following technical effects or advantages: The present application is connected with a welding device, which includes a rotating module, a lifting module, a telescopic module and at least one welding head; the reinforcement ring to be welded of the LNG storage tank is determined, and the welding parameters of the web and the wing plate in the reinforcement ring to be welded are collected, including welding strength and weld position; the reinforcement ring to be welded is placed on the welding device, and the lifting module and the telescopic module align and position the welding head with the reinforcement ring to be welded according to the weld position. When the alignment and positioning are completed, the execution parameters of the welding head are configured according to the welding strength; the rotation module predicts the execution parameters according to the embedded welding quality prediction model, obtains the matching rotation parameters that meet the preset welding quality, and controls the rotation module according to the matching rotation parameters. The present invention solves the technical problems of insufficient welding accuracy and substandard welding strength when welding the reinforcement ring of the LNG storage tank in the prior art. Through an intelligent welding device, including a rotating module, a lifting module and a telescopic module, the reinforcement ring of the LNG storage tank is automatically welded. By collecting welding parameters, the execution parameters and rotation speed of the welding head are dynamically adjusted using the embedded welding quality prediction model to ensure the accuracy of the weld quality, thereby achieving the technical effect of improving welding efficiency and weld quality stability. BRIEF DESCRIPTION OF THE DRAWINGS
[0008] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.
[0009] Figure 1A schematic diagram of a process flow of an intelligent welding method for a reinforcement ring of an LNG storage tank provided in an embodiment of the present application; Figure 2 Schematic diagram of the structure of the intelligent welding platform for the reinforcement ring of the LNG storage tank provided in an embodiment of the present application.
[0010] Description of reference numerals: welding device connection unit 11 , welding parameter acquisition unit 12 , execution parameter configuration unit 13 , rotation parameter determination unit 14 . DETAILED DESCRIPTION
[0011] The present application provides an intelligent welding method and platform for the reinforcement ring of an LNG storage tank, aiming to solve the technical problems of insufficient welding accuracy and substandard welding strength when welding the reinforcement ring of the LNG storage tank in the prior art. An intelligent welding device, including a rotating module, a lifting module and a telescopic module, is used to automatically weld the reinforcement ring of the LNG storage tank. By collecting welding parameters and using an embedded welding quality prediction model, the execution parameters and rotation speed of the welding head are dynamically adjusted to ensure the accuracy of the weld quality, thereby achieving the technical effect of improving welding efficiency and the stability of weld quality.
[0012] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application.
[0013] It should be noted that any variations of the terms "include" and "have" are intended to cover non-exclusive inclusions. For example, a process, method, platform, product or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules that are not explicitly listed or inherent to these processes, methods, products or devices.
[0014] Embodiment 1, as Figure 1 As shown, the present application provides an intelligent welding method for a reinforcement ring of an LNG storage tank, the method comprising: Step S100: connecting a welding device, wherein the welding device includes a rotating module, a lifting module, a telescopic module and at least one welding head.
[0015] In an embodiment of the present application, a control system is connected to a welding device. The welding device includes a rotating module, a lifting module, a telescopic module and at least one welding head. The rotating module is responsible for controlling the rotation of the welding head to ensure uniformity during the welding process; the lifting module is used to adjust the height of the welding head so as to accurately align the reinforcement ring to be welded; the telescopic module adjusts the distance of the welding head to adapt to different welding positions. At least one welding head is fixed at the end of the lifting module for performing welding tasks.
[0016] Step S200: determining a reinforcement ring to be welded of an LNG storage tank, and collecting welding parameters of a web plate and a wing plate in the reinforcement ring to be welded, including welding strength and weld position.
[0017] In the embodiment of the present application, when determining the reinforcement ring to be welded of the LNG storage tank, the reinforcement ring to be welded is first automatically identified through the integrated visual recognition technology. Then, the pre-set welding parameters are obtained from the database, including the welding strength and weld position of the web and wing. The welding strength refers to the energy required to ensure the firmness of the weld, while the weld position clarifies the specific location of the welding point at the junction of the web and wing.
[0018] Step S300: placing the reinforcement ring to be welded on the welding device, and the lifting module and the telescopic module align and position the welding head and the reinforcement ring to be welded according to the weld position. After the alignment and positioning is completed, the execution parameters of the welding head are configured according to the welding strength.
[0019] In an embodiment of the present application, the reinforcement ring to be welded is first accurately placed on the welding platform by an automated robotic arm. The robotic arm is equipped with a sensor to detect the position and posture of the reinforcement ring in real time to ensure its stability. Next, a sensor such as a laser rangefinder or a camera is used to obtain the weld position of the reinforcement ring to be welded. According to the weld position, the height of the welding head is adjusted by the lifting module. The lifting module usually uses an electric lifting system that can quickly and accurately adjust the height to ensure that the welding head maintains the correct distance from the docking point of the reinforcement ring to be welded. At the same time, the telescopic module adjusts the horizontal distance between the welding head and the reinforcement ring according to the weld position. This process is controlled by a servo motor to ensure that the welding head can accurately reach the required welding position.
[0020] By monitoring the sensor data in real time, the alignment accuracy of the welding head in all directions is ensured. After the alignment is completed, the execution parameters of the welding head are configured according to the preset welding strength. Specifically, the execution parameters corresponding to the welding strength are automatically selected from the stored parameter table. The parameter table pre-sets the values of welding current, welding voltage and welding speed for different welding strengths.
[0021] Furthermore, in the method provided in the embodiment of the application, before the rotation module predicts the execution parameter according to the embedded welding quality prediction model, it also includes: Acquire execution parameter training samples, wherein the execution parameter training samples include welding current samples, welding voltage samples, and welding speed samples; set rotation parameter training samples, wherein the rotation parameter training samples include rotation speed samples; call a cloud processor, and train a welding quality prediction model based on the execution parameter training samples, the rotation parameter training samples, and quality labels identifying welding quality, wherein the quality labels identifying welding quality include weld penetration, weld defects, and weld load; and download the trained welding quality prediction model to the rotation module.
[0022] In the embodiment of the present application, historical welding data is first accessed through a database interface to extract execution parameter training samples, including welding current samples, welding voltage samples, and welding speed samples. Next, rotation parameter training samples are obtained through historical welding test records and experimental data, and the rotation parameter training samples include rotation speed samples.
[0023] The execution parameter training samples and rotation parameter training samples are then packaged and transmitted to the cloud processor through a secure network. The cloud processor first creates a virtual welding environment and simulates the welding process using a computational model based on physical and chemical principles. The simulation takes into account factors such as material properties, heat input, and welding technology. During this process, the cloud processor performs multiple welding simulations based on the input execution parameters and rotation parameters, records the results of each simulation, and generates quality labels for welding quality by comparing and analyzing these results. These labels include indicators such as weld penetration, weld defects, and weld load. In the data comparison and analysis stage, the welding results of each simulation run are compared with the preset welding quality standards or process specifications. For example, the weld penetration in the simulation results is compared with the target penetration value to ensure that the penetration is within the required range; at the same time, the defects in the weld are detected, the number and distribution of defects are calculated, and it is determined whether the weld meets the quality standards. After the comparison and analysis is completed, a quality label is generated for each simulation, such as qualified or unqualified, and these labels are associated with the relevant execution and rotation parameters and stored. Among them, the quality label is qualified only when the weld penetration, weld defects, and weld load all meet the preset quality standards.
[0024] After obtaining the execution parameter training samples, rotation parameter training samples and corresponding quality labels, these data are input into the machine learning model for training. Using machine learning algorithms, such as support vector machines, these training samples are used to learn the relationship between different parameters and welding quality, and finally a welding quality prediction model is obtained.
[0025] After the training is completed, the welding quality prediction model is downloaded to the rotation module. The rotation module can use the model in actual welding operations to dynamically adjust the rotation speed and other related parameters according to the real-time welding parameters, thereby optimizing the welding process and ensuring that the welding quality meets the preset standards.
[0026] Furthermore, the method provided in the application embodiment also includes: The rotation module also includes a rotation deviation prediction model; the axial deviation data samples and radial deviation data samples of the rotation module under the rotation parameter training samples are recorded; welding deviation samples are obtained according to the axial deviation data samples and the radial deviation data samples; the rotation deviation prediction model is trained according to the welding deviation samples, and the matching rotation parameters are constrained and updated according to the rotation deviation prediction model.
[0027] In an embodiment of the present application, the rotation module also includes a rotation deviation prediction model. To obtain the rotation deviation prediction model, first, when executing the rotation parameter training sample, the axial deviation and radial deviation data of the rotation module are recorded in real time by a high-precision sensor to obtain an axial deviation data sample and a radial deviation data sample. The axial deviation data sample is the offset of the rotation module along the welding axis, and the radial deviation data sample is the displacement deviation between the welding head and the weldment in the radial direction.
[0028] Next, the collected axial deviation data samples and radial deviation data samples are processed to calculate the welding deviation for each training sample. Specifically, for each time point, the actual deviation of the welding head is calculated, and the axial and radial deviation data are processed separately to calculate the absolute deviation value for each time point. Then, the deviation data of all collected time points are statistically processed. The deviation is evaluated by calculating the standard deviation or mean of the deviation at all time points to ensure that they can reflect the overall deviation under the rotation parameter.
[0029] When calculating, the , where D is the deviation and N is the total number of data points collected. and are the axial and radial deviations, and is the mean value of the deviation. Through the calculation of this deviation, the welding deviation sample is obtained.
[0030] The rotation deviation prediction model is then trained using welding deviation samples. During training, the relationship between rotation parameters and deviation is learned through machine learning algorithms, such as recurrent neural networks, to generate a model that can predict deviations under specific rotation parameters. This model can effectively predict the possibility of rotation deviation in actual welding processes through training with a large number of samples.
[0031] Finally, in the actual welding process, the rotation parameters are analyzed in real time according to the rotation deviation prediction model obtained through training, and the possible deviation under the current rotation conditions is predicted. If the prediction model shows that the current rotation parameters may cause excessive deviation, the matching rotation parameters are constrained and updated according to the feedback of the model, and the rotation speed or other parameters are automatically adjusted to ensure that the deviation remains within the preset range, thereby improving the welding accuracy and ensuring the quality of the weld.
[0032] Step S400: the rotation module predicts the execution parameters according to the embedded welding quality prediction model, obtains matching rotation parameters that meet the preset welding quality, and controls the rotation module according to the matching rotation parameters.
[0033] In the embodiment of the present application, the rotation module predicts the execution parameters according to the embedded welding quality prediction model, inputs the execution parameters into the embedded welding quality prediction model, and predicts the quality index of the weld. The welding quality data output by the model includes the penetration depth of the weld, the welding strength, and the potential defects in the weld. The model evaluates whether the current welding conditions meet the preset welding quality requirements by predicting these indicators.
[0034] The preset welding quality is set by technical experts based on the welding process specifications, covering the penetration, strength, and defect-free standards of the weld. For example, the penetration of the weld must reach a specific range, and the weld cannot have obvious defects. The preset welding quality requirements serve as a benchmark for the model to evaluate welding quality and are used to determine whether the current weld is qualified.
[0035] After the model outputs the prediction results of the welding quality, the rotation parameters of the rotation module, such as the rotation speed and angle, are adjusted according to the results. Matching rotation parameters refers to the combination of rotation speed and rotation angle that can meet the preset welding quality requirements. Iterate multiple rotation parameter combinations and try to adjust the uniformity of welding heat input and weld quality through different combinations of rotation speed, acceleration, etc. until the optimal rotation parameters that meet the requirements of weld penetration, weld strength, etc. are found. For example, if the prediction results show that the weld penetration is not enough, reduce the rotation speed and increase the heat input to deepen the penetration; if pores appear in the weld, adjust the rotation acceleration to reduce the stress in the weld. Once the matching rotation parameters are determined, these parameters are transmitted to the rotation module through the control unit to adjust the rotation speed and angle of the welding head in real time.
[0036] Furthermore, the method provided in the application embodiment also includes: If the welding head includes multiple welding heads, the multiple welding heads are arranged in sequence on an annular fixed track, and the annular fixed track is sleeved on the cylindrical surface of the rotating module; after the track distribution positions of the multiple welding heads are determined, the multiple welding heads are locked.
[0037] In the embodiment of the present application, if the welding device includes multiple welding heads, these welding heads will be installed in sequence on a circular fixed track. The circular fixed track is sleeved on the cylindrical surface of the rotating module, allowing the welding heads to perform synchronous welding operations around the reinforcement ring to be welded.
[0038] To install multiple welding heads, first, determine the distribution position of each welding head on the circular track according to the weld distribution and welding process requirements of the reinforcement ring to be welded. Once the distribution positions of multiple welding heads on the circular track are determined, the position of each welding head is accurately adjusted by positioning sensors or laser positioning equipment. In this process, the spatial position of each welding head is monitored in real time and compared with the preset track distribution to ensure that there is no deviation.
[0039] After the distribution position of all welding heads is determined, the welding heads are locked by the automatic locking mechanism. The locking process is usually performed by electric or hydraulic devices to ensure the stability of each welding head during the welding process and avoid position deviation. Once the welding head is fixed in place, the synchronous control stage of the welding operation is entered. Multiple welding heads work together to ensure that the welding process of each welding point is consistent, and the welding current, voltage, speed and other parameters are synchronized at each welding point to avoid inconsistent welding results.
[0040] Furthermore, in the method provided in the embodiment of the application, if the welding head includes a plurality of welding heads, it further includes: The welding positions of the multiple welding heads are analyzed according to the weld positions to determine the corresponding multiple welding positions; the lifting module and the telescopic module determine the lifting displacement and telescopic displacement between each welding head and the reinforcement ring to be welded according to the multiple welding positions, and perform alignment control according to the lifting displacement and telescopic displacement.
[0041] In the embodiment of the present application, the weld position of the reinforcement ring to be welded is first obtained by using sensors such as a laser rangefinder and a camera, and the three-dimensional coordinates of the starting point, end point and path of the weld are recorded. The acquired weld position information is input into the CAD / CAM software for processing, and a weld path map is generated through computer vision algorithms and image processing technology, and the weld is analyzed to automatically divide the weld into multiple welding segments. Each welding head is responsible for different weld segments to ensure that the welding tasks are evenly distributed without overlap or omission.
[0042] Next, determine the welding position, which refers to the position to which the welding head needs to move during the welding process. Specifically, first match the three-dimensional coordinates of the weld segment with the range of motion of the welding head. Through the three-dimensional coordinate system, each welding point of the weld segment will be marked as a specific welding position. Then use the inverse kinematics algorithm to calculate the path and movement required for the welding head to get from its current position to the weld position. Based on the three-dimensional coordinates of the weld, the algorithm takes into account the range of motion and posture of the welding head to determine the specific position that the welding head needs to reach. If there are multiple welding heads, ensure that the welding positions of each welding head do not overlap and can cover all weld segments.
[0043] When the welding position is determined, the movement of the welding head is controlled by the lifting module and the telescopic module. First, the lifting displacement is calculated according to the welding position to ensure that the welding head is aligned with the height of the weld in the vertical direction. At the same time, the telescopic displacement is calculated to ensure that the welding head can reach the accurate position of the weld in the horizontal direction. Then, the motion control algorithm is used to calculate the movement trajectory of each welding head based on the three-dimensional coordinate difference between the current position of the welding head and the welding position. The system sends these displacement instructions to the control module of the welding head to ensure that the welding head can reach the required welding position.
[0044] During the welding operation, the PID controller is used to achieve precise alignment control of the welding head. By real-time monitoring of the sensor feedback information, the lifting module and telescopic module of the welding head are continuously adjusted to ensure that the welding head is always aligned with the weld. If the sensor detects that the position of the welding head deviates from the weld during the welding process, fine-tuning is immediately performed to correct the position of the welding head to ensure the stability of the weld quality.
[0045] Furthermore, the method provided in the application embodiment also includes: The number of welding heads is configured through an activation function; a preset welding efficiency is set; and the synchronous welding efficiency of multiple welding heads is obtained; the activation function takes the preset welding efficiency as a target and the synchronous welding efficiency of the multiple welding heads as an input variable to obtain a response result, and the response result is the number of welding heads that need to be started.
[0046] In the embodiment of the present application, the number of welding heads is configured by an activation function. The preset welding efficiency is set by a technical expert according to the requirements of the welding task. The efficiency value is defined according to production requirements and welding process requirements to ensure that a specific number of weld tasks are completed within a given time. For example, the preset welding efficiency may mean that a certain length of weld is completed per minute.
[0047] Next, collect the synchronous welding efficiency of multiple welding heads, that is, the welding efficiency that each welding head can achieve when multiple welding heads work synchronously. The synchronous welding efficiency is the amount of welding that each welding head can complete within a certain period of time. The efficiency values of these welding heads are obtained through historical data or real-time monitoring.
[0048] An activation function is then used to determine the number of welding heads that need to be started. An activation function is a mathematical model that takes a preset welding efficiency as a target and the synchronous welding efficiency of each welding head as an input variable. The role of the activation function is to automatically calculate the number of welding heads that need to be started based on the required welding task and the currently available welding head efficiency. Specifically, the activation function is , where N represents the number of welding heads that need to be started. It is the preset welding efficiency, such as the length of the weld to be completed per minute. It is the average synchronous welding efficiency of each welding head, which indicates the length of weld that each welding head can complete in a certain period of time. When the calculated N is not an integer, round it up to ensure that the number of welding heads is an integer.
[0049] The response result calculated by the activation function is the number of welding heads that need to be started.
[0050] Furthermore, in the method provided in the embodiment of the application, obtaining the synchronous welding efficiency of multiple welding heads also includes: The number K of welding heads on the rotating module is obtained; according to the number K of welding heads, synchronous welding groups are set in sequence and incremented, and a synchronous welding group set is output; an efficiency test is performed according to the synchronous welding group set, and the synchronous welding efficiency of multiple welding heads is output based on the efficiency test result.
[0051] In the embodiment of the present application, firstly, the number K of welding heads currently installed on the rotating module is automatically detected by the sensor or control module integrated on the rotating module. For example, if 5 welding heads are detected on the rotating module, then K=5.
[0052] After obtaining the number of welding heads K, different synchronous welding groups are set in increasing order. Here, an increasing combination algorithm is used to generate combinations from 1 welding head to all K welding heads in sequence. Each combination represents a synchronous welding group. Specifically, starting from 1 head, the number of heads is gradually increased until the maximum number K. For example, if K=5, 5 welding groups are generated in sequence, welding group No. 1 contains 1 welding head, welding group No. 2 contains 2 welding heads, welding group No. 3 contains 3 welding heads, welding group No. 4 contains 4 welding heads, and welding group No. 5 contains 5 welding heads. These welding groups constitute a synchronous welding group set.
[0053] Next, each of these synchronous welding groups is subjected to an efficiency test. The purpose of the efficiency test is to measure the working efficiency of each synchronous welding group through actual operation or simulated operation. During the test, each welding group is started one by one, and 1 welding head, 2 welding heads... until all K welding heads work simultaneously. During the test, the length of the weld completed by each welding group per unit time is recorded. For example, the number of meters that each welding group can weld in one minute is tracked to determine its welding efficiency. After the efficiency test is completed, the collected test data is processed and the synchronous welding efficiency of multiple welding heads is finally output. The synchronous welding efficiency refers to the welding efficiency when different numbers of welding heads work simultaneously, expressed as the total length welded per unit time.
[0054] Furthermore, in the method provided in the embodiment of the application, the welding positions of the plurality of welding heads are analyzed according to the weld positions, and further includes: Record multiple status data sets of the multiple welding heads; perform abnormality identification according to the multiple status data sets to determine the abnormal welding head; and reconfigure the welding positions of other welding heads at the weld position based on the abnormal welding head.
[0055] In the embodiment of the present application, each welding head is first monitored in real time by a sensor to record state data sets of multiple welding heads, which include parameters such as welding current, voltage, and welding speed.
[0056] Next, the state data set is analyzed through an anomaly detection algorithm, and abnormal welding heads are identified by comparing the current state of each welding head with the preset normal working range. For example, when the welding current of a welding head is detected to be out of the normal range, the welding head is marked as abnormal. Once the abnormal welding head is identified, its work is suspended and other normally working welding heads are dynamically adjusted. Through the path planning algorithm, the welding tasks are reallocated, and the unfinished weld tasks of the abnormal welding head are handed over to other heads to continue to execute.
[0057] In the embodiments of the present application, in summary, the embodiments of the present application have at least the following technical effects: The present application is connected with a welding device, which includes a rotating module, a lifting module, a telescopic module and at least one welding head; the reinforcement ring to be welded of the LNG storage tank is determined, and the welding parameters of the web and the wing plate in the reinforcement ring to be welded are collected, including welding strength and weld position; the reinforcement ring to be welded is placed on the welding device, and the lifting module and the telescopic module align and position the welding head with the reinforcement ring to be welded according to the weld position. When the alignment and positioning are completed, the execution parameters of the welding head are configured according to the welding strength; the rotation module predicts the execution parameters according to the embedded welding quality prediction model, obtains the matching rotation parameters that meet the preset welding quality, and controls the rotation module according to the matching rotation parameters. The present invention solves the technical problems of insufficient welding accuracy and substandard welding strength when welding the reinforcement ring of the LNG storage tank in the prior art. Through an intelligent welding device, including a rotating module, a lifting module and a telescopic module, the reinforcement ring of the LNG storage tank is automatically welded. By collecting welding parameters, the execution parameters and rotation speed of the welding head are dynamically adjusted using the embedded welding quality prediction model to ensure the accuracy of the weld quality, thereby achieving the technical effect of improving welding efficiency and weld quality stability.
[0058] Embodiment 2, based on the same inventive concept as the intelligent welding method for the reinforcement ring of the LNG storage tank in the previous embodiment, Figure 2 As shown, the present application provides an intelligent welding platform for reinforcing rings of LNG storage tanks. The platform and method embodiments in the present application are based on the same inventive concept. The platform includes: A welding device connection unit 11, wherein the welding device connection unit 11 is connected to a welding device, wherein the welding device includes a rotating module, a lifting module, a telescopic module and at least one welding head; a welding parameter acquisition unit 12, wherein the welding parameter acquisition unit 12 determines the reinforcement ring to be welded of the LNG storage tank, and collects welding parameters of the web and the wing plate in the reinforcement ring to be welded, including welding strength and weld position; an execution parameter configuration unit 13, wherein the execution parameter configuration unit 13 places the reinforcement ring to be welded on the welding device, and the lifting module and the telescopic module align and position the welding head and the reinforcement ring to be welded according to the weld position, and when the alignment and positioning are completed, configure the execution parameters of the welding head according to the welding strength; a rotation parameter determination unit 14, wherein the rotation parameter determination unit 14 predicts the execution parameters according to the embedded welding quality prediction model through the rotation module, obtains matching rotation parameters that meet the preset welding quality, and controls the rotation module according to the matching rotation parameters.
[0059] Furthermore, the platform is also used to implement the following functions: Acquire execution parameter training samples, wherein the execution parameter training samples include welding current samples, welding voltage samples, and welding speed samples; set rotation parameter training samples, wherein the rotation parameter training samples include rotation speed samples; call a cloud processor, and train a welding quality prediction model based on the execution parameter training samples, the rotation parameter training samples, and quality labels identifying welding quality, wherein the quality labels identifying welding quality include weld penetration, weld defects, and weld load; and download the trained welding quality prediction model to the rotation module.
[0060] Furthermore, the platform is also used to implement the following functions: The rotation module also includes a rotation deviation prediction model; the axial deviation data samples and radial deviation data samples of the rotation module under the rotation parameter training samples are recorded; welding deviation samples are obtained according to the axial deviation data samples and the radial deviation data samples; the rotation deviation prediction model is trained according to the welding deviation samples, and the matching rotation parameters are constrained and updated according to the rotation deviation prediction model.
[0061] Furthermore, the platform is also used to implement the following functions: If the welding head includes multiple welding heads, the multiple welding heads are arranged in sequence on an annular fixed track, and the annular fixed track is sleeved on the cylindrical surface of the rotating module; after the track distribution positions of the multiple welding heads are determined, the multiple welding heads are locked.
[0062] Furthermore, the platform is also used to implement the following functions: The welding positions of the multiple welding heads are analyzed according to the weld positions to determine the corresponding multiple welding positions; the lifting module and the telescopic module determine the lifting displacement and telescopic displacement between each welding head and the reinforcement ring to be welded according to the multiple welding positions, and perform alignment control according to the lifting displacement and telescopic displacement.
[0063] Furthermore, the platform is also used to implement the following functions: The number of welding heads is configured through an activation function; a preset welding efficiency is set; and the synchronous welding efficiency of multiple welding heads is obtained; the activation function takes the preset welding efficiency as a target and the synchronous welding efficiency of the multiple welding heads as an input variable to obtain a response result, and the response result is the number of welding heads that need to be started.
[0064] Furthermore, the platform is also used to implement the following functions: The number K of welding heads on the rotating module is obtained; according to the number K of welding heads, synchronous welding groups are set in sequence and incremented, and a synchronous welding group set is output; an efficiency test is performed according to the synchronous welding group set, and the synchronous welding efficiency of multiple welding heads is output based on the efficiency test result.
[0065] Furthermore, the platform is also used to implement the following functions: Record multiple status data sets of the multiple welding heads; perform abnormality identification according to the multiple status data sets to determine the abnormal welding head; and reconfigure the welding positions of other welding heads at the weld position based on the abnormal welding head.
[0066] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. The processes depicted in the accompanying drawings do not necessarily require the specific order and continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.
[0067] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the protection scope of the present application.
[0068] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.
Claims
1. Intelligent welding method for reinforcement ring of LNG storage tank, characterized in that: The method comprises: Connecting a welding device, the welding device comprising a rotating module, a lifting module, a telescopic module and at least one welding head; Determine the reinforcement ring to be welded of the LNG storage tank, and collect welding parameters of the web and wing plates in the reinforcement ring to be welded, including welding strength and weld position; The reinforcement ring to be welded is placed on the welding device, and the lifting module and the telescopic module align and position the welding head and the reinforcement ring to be welded according to the weld position. After the alignment and positioning is completed, the execution parameters of the welding head are configured according to the welding strength; The rotation module predicts the execution parameters according to the embedded welding quality prediction model, obtains matching rotation parameters that meet the preset welding quality, and controls the rotation module according to the matching rotation parameters.
2. The intelligent welding method for the reinforcement ring of the LNG storage tank according to claim 1, characterized in that: Before the rotation module predicts the execution parameter according to the embedded welding quality prediction model, the method includes: Acquire execution parameter training samples, wherein the execution parameter training samples include welding current samples, welding voltage samples, and welding speed samples; Setting a rotation parameter training sample, wherein the rotation parameter training sample includes a rotation speed sample; Calling a cloud processor to train a welding quality prediction model based on the execution parameter training sample, the rotation parameter training sample and the quality label identifying the welding quality, wherein the quality label identifying the welding quality includes weld penetration, weld defects and weld load; The trained welding quality prediction model is downloaded to the rotation module.
3. The intelligent welding method for the reinforcement ring of the LNG storage tank according to claim 2, characterized in that: The rotation module also includes a rotation deviation prediction model; Recording the axial deviation data samples and radial deviation data samples of the rotation module under the rotation parameter training samples; Acquire a welding deviation sample according to the axial deviation data sample and the radial deviation data sample; A rotation deviation prediction model is trained according to the welding deviation samples, and the matching rotation parameters are constrained and updated according to the rotation deviation prediction model.
4. The intelligent welding method for the reinforcement ring of the LNG storage tank according to claim 1, characterized in that: If the welding head includes a plurality of welding heads, the plurality of welding heads are sequentially arranged on an annular fixed track, and the annular fixed track is sleeved on the cylindrical surface of the rotating module; After the track distribution positions of the multiple welding heads are determined, the multiple welding heads are locked.
5. The intelligent welding method for the reinforcement ring of the LNG storage tank according to claim 1, characterized in that: If the welding head comprises a plurality of welding heads, the method further comprises: Analyzing the welding positions of the plurality of welding heads according to the weld positions to determine the corresponding plurality of welding positions; The lifting module and the telescopic module determine the lifting displacement and telescopic displacement between each welding head and the reinforcement ring to be welded according to the multiple welding positions, and perform alignment control according to the lifting displacement and telescopic displacement.
6. The intelligent welding method for the reinforcement ring of the LNG storage tank according to claim 1, characterized in that: The number of welding heads is configured by an activation function; Set preset welding efficiency; Get the synchronous welding efficiency of multiple welding heads; The activation function takes the preset welding efficiency as a target and the synchronous welding efficiency of the multiple welding heads as an input variable to obtain a response result, and the response result is the number of welding heads that need to be started.
7. The intelligent welding method for the reinforcement ring of the LNG storage tank according to claim 6, characterized in that: Acquiring the synchronous welding efficiency of multiple welding heads, the method includes: Obtain the number K of welding heads on the rotating module; According to the number K of welding heads, synchronous welding groups are set in increasing order, and a synchronous welding group set is output; An efficiency test is performed according to the synchronous welding group set, and the synchronous welding efficiency of multiple welding heads is output based on the efficiency test result.
8. The intelligent welding method for the reinforcement ring of the LNG storage tank according to claim 5, characterized in that: Analyzing the welding positions of the plurality of welding heads according to the weld positions also includes: Recording a plurality of status data sets of the plurality of welding heads; Perform abnormality identification according to the multiple status data sets to determine an abnormal welding head; Based on the abnormal welding head, the welding positions of other welding heads are reconfigured at the weld position.
9. The intelligent welding platform for the reinforcement ring of LNG storage tanks is characterized by: The platform includes: A welding device connecting unit, wherein the welding device connecting unit is connected to a welding device, wherein the welding device includes a rotating module, a lifting module, a telescopic module and at least one welding head; A welding parameter acquisition unit, wherein the welding parameter acquisition unit determines a reinforcement ring to be welded of the LNG storage tank, and acquires welding parameters of a web plate and a wing plate in the reinforcement ring to be welded, including welding strength and weld position; An execution parameter configuration unit, wherein the execution parameter configuration unit places the reinforcement ring to be welded on the welding device, and the lifting module and the telescopic module align and position the welding head and the reinforcement ring to be welded according to the weld position, and after the alignment and positioning is completed, configures the execution parameters of the welding head according to the welding strength; A rotation parameter determination unit predicts the execution parameters according to an embedded welding quality prediction model through the rotation module, obtains matching rotation parameters that meet preset welding quality, and controls the rotation module according to the matching rotation parameters.
Citation Information
Patent Citations
Method and system for bisymmetric heat source welding based on flexible arc plasma
CN108465961A
Welding line forming quality monitoring method
CN108500498A
Automatic welding machine based on machine vision and control method thereof
CN112935473A
Welding monitoring device and welding monitoring system
CN113125013A
Strengthening parameter generation method and system based on welding process prediction and storage medium
CN115609180A
Cited By
Large-span water conservancy gate and manufacturing process thereof
CN120480349A
