A method for real-time monitoring of temperature in a core region of double-sided hybrid friction stir welding

Abstract

The present application belongs to the technical field of friction stir welding process monitoring, and relates to a kind of real-time monitoring method for core zone temperature of double-sided composite friction stir welding.The present application comprehensively considers the heat generated by friction between upper and lower shoulder, the heat generated by friction between pin and metal, and the heat generated by plastic deformation of metal during welding process, uses coupled Euler-Lagrange simulation technology to establish a DS-FSW process simulation model, obtains the temperature distribution of welding process, and combines improved random dual coordinate ascent algorithm to build a correlation model between extreme temperature of core zone and temperature of feature points on the edge surface of shoulder, so as to realize the characterization of extreme temperature of core zone based on surface temperature.An online monitoring system integrating surface infrared real-time temperature measurement, parameter adjustment and visualization function is developed to realize accurate online monitoring of core zone temperature of welding.The present application provides a high-precision intelligent monitoring method for real-time control of DS-FSW process parameters, and significantly improves the welding quality and process stability of medium-thick plate.

Description

Technical Field

[0001] This invention belongs to the field of process monitoring technology for double-sided composite friction stir welding (DS-FSW), and relates to a real-time temperature monitoring method for the core area of ​​double-sided composite friction stir welding. It is a method for monitoring the core area temperature of double-sided composite friction stir welding (DS-FSW) based on the correlation between the surface temperature of the weldment and the core area temperature. Background Technology

[0002] Friction stir welding (FSW) is a solid-state plastic joining technology invented by the Welding Institute in the UK in the 1990s. It achieves plastic joining of materials through frictional heat generated between the stirring head and the workpiece, and has been widely used in aerospace, rail transportation, and other fields. However, traditional single-sided FSW has significant technical bottlenecks when welding medium-thick plates: a nonlinear temperature gradient along the thickness direction leads to defects such as inhomogeneous microstructure in the thermo-mechanically affected zone, incomplete root penetration, and excessive weld deformation. Therefore, researchers have proposed DS-FSW technology, which uses simultaneous welding with stirring heads on both sides of the workpiece to equalize the temperature field along the thickness direction. Its bidirectional thermal effect promotes full plastic flow of the material along the thickness, eliminating incomplete root penetration defects; at the same time, symmetrical heat input suppresses asymmetrical deformation, significantly improving the uniformity and dimensional stability of the weld microstructure, providing a new path for high-quality welding of thick plates.

[0003] The temperature of the core region of the DS-FSW (including the weld nugget area and thermomechanical influence zone located below the stirring head shoulder) affects the plastic flow and microstructure evolution of the weld material. Excessive or insufficient temperature can cause welding defects, thus affecting weld quality. Therefore, it is necessary to characterize and control the temperature of the core region in a timely manner. However, effective monitoring of the temperature in the DS-FSW core region faces the following technical challenges: (1) The coupling mechanism of the two heat sources is complex, and the synchronous heat generation of the upper and lower shoulders and the stirring pin makes temperature field modeling difficult; (2) Existing contact temperature measurement methods (such as embedded thermocouples) have inherent defects such as damaging the workpiece or stirring head structure and response lag; (3) Non-contact infrared temperature measurement, due to the obstruction of the double-shoulder structure, can only obtain surface temperature data of a limited area at the shoulder edge. Current domestic and international research has not yet broken through the in-situ measurement technology of the temperature field in the DS-FSW core region. Summary of the Invention

[0004] To address the problems of existing technologies, this invention provides a method for real-time monitoring of the core temperature in double-sided composite friction stir welding (DS-FSW) based on surface feature point temperatures, and develops a DS-FSW core temperature monitoring system to achieve real-time monitoring of the core temperature. This invention innovatively integrates infrared thermal imager temperature measurement technology, thermo-mechanical coupled finite element simulation, and the improved stochastic dual coordinate ascent (ISDCA) algorithm to construct an intelligent core temperature monitoring system for the DS-FSW process, solving the technical challenge of real-time characterization of the core temperature in the welding process of medium and thick plates.

[0005] The technical solution of the present invention is as follows:

[0006] A method for real-time temperature monitoring of the core area in double-sided composite friction stir welding, comprising the following steps:

[0007] Step 1: Taking into account the heat generation in the contact area between the upper and lower shoulders of the DS-FSW and the base material, as well as the heat generated by the plastic deformation of the metal, a simulation model of the DS-FSW process based on the coupled Euler-Lagrange method is established.

[0008] Step 2: Analyze the temperature field distribution of the DS-FSW process simulation model and extract the dataset of extreme temperatures (peak and minimum temperatures) in the core area and the corresponding temperature of feature points on the shoulder edge surface during the welding process;

[0009] Step 3: Establish a core region extreme temperature characterization model based on the ISDCA algorithm. This model consists of two parts: a basic prediction model and a model adaptive optimization using a dynamic adapter.

[0010] The basic prediction model uses temperature data from surface feature points on the advancing and retreating sides during welding as input. Normalization and trimming layers are used to standardize the raw data, eliminating dimensional differences between data points and limiting the interference of outliers. The model then uses a stochastic coordinate ascent algorithm to optimize parameters and incorporates an early stopping mechanism during training. By monitoring the loss changes on the validation set in real time, training automatically terminates if the validation loss fails to improve for several consecutive cycles, preventing overfitting. After training, the basic model outputs the extreme temperature characterization results of the welding core area. The model parameters are stabilized through regularization constraints, providing a reliable baseline prediction for subsequent adaptive optimization.

[0011] With the continuous accumulation of experimental data, to further enhance the model's adaptability to real-world data, a dynamic adapter adaptively optimizes and corrects the basic prediction model using real-time measured surface feature point temperatures and extreme temperatures in the welding core area. The dynamic adapter is innovatively designed as a structure combining a feature extractor and a parameter generator. The feature extractor captures key features (such as weights and biases) from the input data, while the parameter generator generates the necessary scaling factors and bias adjustment terms in real time. These parameters are dynamically updated using a backpropagation algorithm with noisy estimation, achieving deep integration of the basic prediction model with the latest experimental data. The scaling factor, after processing with the Sigmoid function, is strictly constrained within the range of [0.8, 1.2], and the bias adjustment term is constrained within the range of [-1, 1], ensuring that the model's representation results remain within a reasonable range. The adaptively optimized temperature representation model can more accurately represent the extreme temperatures in the core area during actual welding.

[0012] Step 4: Establish a temperature monitoring system for the core area of ​​the DS-FSW, which integrates real-time temperature measurement, parameter adjustment, and visualization functions for feature points on the surface of the weldment based on an infrared thermal imager;

[0013] Step 5: Encapsulate the extreme temperature characterization model of the core area and embed it into the DS-FSW core area temperature monitoring system;

[0014] Step 6: Conduct DS-FSW experiment. During the welding process, the infrared thermal imager measures the surface feature point temperature of the shoulder edge of the stirring head in real time. The surface feature point temperature is extracted using global variables and input into the core area extreme temperature characterization model. The model outputs the core area extreme temperature characterization value and displays it on the front-end interface of the monitoring system. When the core area extreme temperature exceeds the solid-liquid phase temperature threshold of the weldment material, the over-limit alarm function is triggered, providing a reference for welding process control.

[0015] The beneficial effects of this invention are:

[0016] This invention proposes a DS-FSW core region temperature monitoring method based on the correlation between the surface temperature and the core region temperature of the weldment. It establishes a core region extreme temperature characterization model based on an improved stochastic dual coordinate ascent algorithm, develops a DS-FSW core region extreme temperature monitoring system, and achieves in-situ characterization of the core region temperature in double-sided composite friction stir welding for the first time. It realizes real-time monitoring of the core region temperature during the welding process and conducts field experiments to verify the effectiveness of the proposed method, which can provide a reference for welding process control. Attached Figure Description

[0017] Figure 1 This is a schematic diagram of the extraction of extreme temperatures in the core area;

[0018] Figure 2This is a schematic diagram of surface feature point temperature extraction;

[0019] Figure 3 This is a flowchart of the process for establishing a characterization model for extreme temperatures in the core area;

[0020] Figure 4 It is the extreme temperature result of the core area characterized by the monitoring system. Detailed Implementation

[0021] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and technical solutions.

[0022] This embodiment establishes a DS-FSW process simulation model for a 12mm thick 2219 aluminum alloy, extracts extreme temperature data of the surface and core regions, establishes the correlation between the two, establishes a characterization model of the extreme temperature of the core region based on the correlation, and builds a core region temperature monitoring system. Combined with real-time measurement of surface temperature using an infrared thermal imager, real-time monitoring of the core region temperature is achieved. The specific steps are as follows:

[0023] (1) A simulation model of the DS-FSW process of 12mm thick 2219 aluminum alloy was established based on the coupled Euler-Lagrange method.

[0024] The geometric models of the stirring head and the weldment were created using UG. The geometric parameters of the stirring head are shown in Table 1.

[0025] Table 1 Geometric parameters of the stirring head

[0026]

[0027] The weldment model includes a 100mm×70mm×12mm Lagrangian solid and a 100mm×70mm×14mm Eulerian solid. After the geometric models of the weldment and the stirring head are established, they are imported into Abaqus finite element simulation software for assembly. During assembly, the bottom surfaces of the two are aligned, and the stirring head is located 20mm from the centerline on the left side of the weldment. The bottom surface of the stirring pin coincides with the upper surface of the Lagrangian solid. The Lagrangian solid is used for material assignment, and the Eulerian solid is used in the simulation calculation. The mesh elements of the stirring head are tetrahedral, with a total of 21,712 nodes and 115,310 elements. The mesh elements of the weldment are hexahedral, with a total of 341,901 nodes. The total number of mesh elements was reduced from 1,050,000 to 300,000 through local refinement. The heat generated during the welding process comes from frictional heat and plastic deformation heat. The initial temperature is set to 25℃, and three heat dissipation methods are considered: heat conduction (conductivity coefficient 100), heat convection, and heat radiation. The thermal conductivity of the weldment and the fixture is set to 100 W / m. 2 At ℃, the thermal conductivity between the weldment and air is 300 W / m. 2The contact method between the stirring head and the weldment is set to universal contact. In the initial stage of welding, sliding friction is the main heat generation method; when the temperature exceeds a certain value, the material becomes viscous and undergoes plastic flow, with shear friction becoming dominant. The Johnson-Cook constitutive model is used to describe the relationship between the rheological stress of 2219 aluminum alloy and temperature, strain, and strain rate. Finally, based on the established process simulation model, the temperature field under multiple sets of different process parameters is analyzed.

[0028] (2) Data sets of extreme temperatures in the core area and characteristic point temperatures on the shoulder edge surface during the welding process were obtained based on the DS-FSW process simulation model.

[0029] By analyzing the temperature field distribution under different process parameters, it was found that the peak temperature in the core region of the DS-FSW is located below the shoulder of the stirring head on the weld cross-section, while the lowest temperature in the core region is located at the bottom surface of the stirring pin on the weld cross-section. Figure 1 As shown. The selected surface feature points are located on both sides of the shoulder, as... Figure 2 As shown, this position ensures that temperature measurement errors caused by heat radiation from the stirring head are avoided when using an infrared thermal imager for temperature measurement during the welding process. 144 sets of surface feature point temperature data and corresponding core area extreme temperature data were extracted as a dataset during the welding feed stage of the process simulation under different process parameters.

[0030] (3) Establish a core region extreme temperature characterization model based on the ISDCA algorithm.

[0031] The process of building this model is as follows: Figure 3 As shown. The basic prediction model is established using the stochastic coordinate ascent algorithm, and its representation is as follows:

[0032]

[0033] in, The core region extreme temperature output by the basic model, with input feature dimension n=2, x i The corresponding surface feature point temperature measurement value, w i b represents the weight of the input quantity, and b represents the bias.

[0034] During training, surface feature point temperatures are used as input and core region extreme temperatures are used as output. The training / validation / test sets are divided into a 100:22:22 ratio. Mean squared error is used as the loss function, with a convergence threshold of 0.001, an initial learning rate of 0.001, weight initialization range of [0, 0.01], initial bias of 0, and a maximum of 100 iterations. An early stopping mechanism is implemented: if the validation loss no longer decreases after 5 consecutive iterations, training is stopped to prevent overfitting. Finally, the model parameters with the best generalization performance are saved as the base prediction model.

[0035] To adapt to the experimental data, a dynamic adapter is introduced to further improve the model's generalization performance. As experimental data continues to be generated, the adapter generates a scaling factor α(x)∈[0.8,1.2] and a bias term β(x)∈[-1,1] in a feedforward manner to correct the output of the base prediction model.

[0036]

[0037] The temperature characterization model trained using the above methods can output adaptively optimized extreme temperature characterization values ​​for the core region.

[0038] (4) Development and experimental verification of the monitoring system.

[0039] A dual-sided composite FSW core area extreme temperature monitoring system was developed using C#. The system's software incorporates a class file for running the core area extreme temperature characterization model. During welding, a global variable calorimeter is used to measure the temperature of characteristic points on the weldment surface. The core area extreme temperature characterization model is then invoked and displayed on the front-end interface. When the core area extreme temperature exceeds the solid-liquid phase temperature threshold of the weldment material, an over-limit alarm function is triggered for timely adjustment.

[0040] Five welding experiments were conducted on 12mm thick 2219 aluminum alloy under different process parameters using a double-sided composite friction stir welding (FSW) system. Surface temperature was acquired using an infrared thermal imager, and the extreme temperature of the core region was characterized in-situ using a double-sided composite FSW core region extreme temperature monitoring system, verified using thermocouples. The extreme temperature data of the core region obtained from the thermocouples will be compared with the extreme temperature results characterized by the monitoring system. Figure 4 As shown, the maximum relative error of the peak temperature in the welding core area is 3.96%, and the maximum relative error of the minimum temperature in the core area is 4.70%, which verifies the effectiveness of the proposed monitoring method.

Claims

1. A method for real-time monitoring of temperature in a double-sided composite friction stir welding nugget, comprising: The specific steps are as follows: Step 1: Considering the heat generation in the contact area between the upper and lower shoulder of double-sided composite friction stir welding (DS-FSW) and the base material and the heat generation due to metal plastic deformation, a DS-FSW process simulation model based on the coupled Euler-Lagrange method is established; Step 2: Analyze the temperature field distribution of the DS-FSW process simulation model, and extract the data set of the extreme temperature of the core area and the corresponding shoulder edge surface feature point temperature during the welding process; Step 3: Establish a core area extreme temperature representation model based on the improved stochastic dual coordinate ascent algorithm (ISDCA), including a basic prediction model and a model self-adaptive optimization using a dynamic adapter; The basic prediction model takes the temperature data of the surface feature points on the advancing side and the retreating side during the welding process as input, standardizes the original data through normalization and clipping layers, then optimizes the parameters using the stochastic coordinate ascent algorithm, and introduces an early stopping mechanism during training; After training, the basic model outputs the extreme temperature representation results of the welding core area, and the model parameters are kept stable through regularization constraints, providing a reliable baseline prediction for subsequent model self-adaptive optimization; The dynamic adapter performs self-adaptive optimization and correction on the basic prediction model by measuring the surface feature point temperature and the welding core area extreme temperature data in real time; the dynamic adapter includes a feature extractor and a parameter generator, where the feature extractor is responsible for capturing key features in the input data, including weights and biases; The parameter generator generates the required scale factor and bias adjustment term in real time; these parameters are dynamically updated through the backpropagation algorithm with noise estimation; Step 4: Establish a DS-FSW core area temperature monitoring system, which integrates real-time measurement of surface feature point temperature based on an infrared thermal imager, parameter adjustment, and visualization functions; Step 5: Package and embed the core area extreme temperature representation model into the DS-FSW core area temperature monitoring system; Step 6: Perform DS-FSW experiments, and in the welding process, the infrared thermal imager measures the surface feature point temperature of the shoulder edge of the stir head in real time, and the global variable is used to extract the surface feature point temperature and input it into the core area extreme temperature representation model. The model outputs the core area extreme temperature representation value and displays it on the front-end interface of the monitoring system. When the core area extreme temperature exceeds the solid-liquid phase temperature threshold of the welding material, an out-of-limit alarm function is triggered, providing a reference for welding process control.

2. The method for real-time monitoring of the temperature of the core zone of double-sided composite friction stir welding according to claim 1, characterized in that, In step 3, the basic prediction model is established using the stochastic coordinate ascent algorithm, and the expression is: (1) where, Core zone extremum temperature output by the base model, input feature dimension n = 2, x i Corresponding surface feature point temperature measurement value, w i Weight of the input quantity, and b is the bias.

3. The method of claim 1, wherein the method is characterized by: In step 3, the basic prediction model optimized and corrected by the dynamic adapter outputs: (2) wherein, is an adaptive optimized core zone extreme temperature representation value output by the trained temperature representation model; is a scale factor generated by the dynamic adapter in a feedforward manner; is a bias term generated by the dynamic adapter in a feedforward manner.

4. The method of real-time monitoring of the temperature of the core zone of double-sided composite friction stir welding according to claim 1, characterized in that, In step 3, the scale factor is constrained in the interval [0.8, 1.2] after being processed by the Sigmoid function, and the bias adjustment term is constrained in the interval [-1, 1].

5. The method of real-time monitoring of the temperature of the core zone of double-sided composite friction stir welding according to claim 1, characterized in that, In step 2, the core area extreme temperature includes the peak temperature and the minimum temperature.

Images