Friction stir welding control method and system based on dynamic temperature prediction and control

Abstract

The present application provides a stir friction welding control method and system based on dynamic temperature prediction and control, which relates to the technical field of stir friction welding. The method includes: real-time acquisition of the current welding area temperature corresponding to the current position of the stirring head; prediction of the corresponding predicted temperature value when the stirring head reaches the next welding area based on the current welding area temperature, the moving speed of the stirring head, and the preset welding area length; calculation of the target rotation speed based on the predicted temperature value and the target temperature range, so that the temperature of the next welding area is within the target temperature range when the stirring head reaches the next welding area; calculation of the rotation speed change rate from the current rotation speed to the target rotation speed based on the current rotation speed of the stirring head, the target rotation speed, and the moving speed of the stirring head to control the rotation speed of the stirring head. Implementation of this method can reduce the fluctuation of the temperature of the welding area, improve the accuracy and stability of temperature control, and thus improve the welding quality.

Description

Technical Field

[0001] The present application relates to the technical field of friction stir welding, and in particular to a friction stir welding control method and system based on dynamic temperature prediction and regulation. Background Art

[0002] Friction stir welding, a solid-state joining process, has recently gained widespread application in aerospace, rail transportation, and automotive manufacturing. This process utilizes a high-speed rotating stirrer that moves along the weld seam to frictionally heat and plastically stir the workpieces, achieving high-quality metal connections. The heat input and temperature distribution during welding directly impact the mechanical properties and quality of the weld. Therefore, effective temperature control during welding has long been a key concern in the field of friction stir welding.

[0003] To control the temperature of the welding area, a temperature sensor is typically used to collect real-time data on the current position of the stirring head or the temperature of the welding area. Based on the collected temperature feedback signal, the stirring head's rotation speed or welding movement speed is directly adjusted to maintain the welding area temperature within a predetermined range as much as possible. Existing systems often use closed-loop feedback control based on field temperature signals. This involves adjusting the stirring head parameters in real time based on the deviation between the current temperature and the target temperature using conventional proportional-integral-derivative (PID) control algorithms to meet the basic heat input requirements of the welding process.

[0004] However, due to the inertia of the stirring head and the hysteresis of heat transfer, related technologies are prone to response lag when relying on the current temperature signal to adjust parameters, resulting in increased temperature fluctuations in the weld area. At high welding speeds or when the material's thermophysical properties vary significantly, these technologies are prone to temperature control overshoot or hysteresis, which in turn affects the quality of weld formation and microstructure uniformity, making it difficult to meet the demand for more precise dynamic control of welding temperature. Summary of the Invention

[0005] The present application provides a stir friction welding control method and system based on dynamic temperature prediction and control, which is used to address the problem that the temperature of the welding area is difficult to precisely and stably control due to thermal inertia and control lag under complex working conditions in related technologies.

[0006] In a first aspect, the present application provides a friction stir welding control method based on dynamic temperature prediction and control, which is applied to a friction stir welding control system. The method comprises:

[0007] The temperature sensor set in the welding area collects the current welding area temperature corresponding to the current position of the stirring head in real time;

[0008] Based on the current welding area temperature, the moving speed of the stirring head and the preset welding area length, predicting the predicted temperature value corresponding to the next welding area when the stirring head reaches the next welding area;

[0009] Calculating a target rotation speed based on the predicted temperature value and a target temperature range corresponding to the next welding area so that the temperature of the next welding area is within the target temperature range when the stirring head reaches the next welding area;

[0010] Calculating a rotation speed change rate from the current rotation speed to the target rotation speed based on the current rotation speed of the stirring head, the target rotation speed, and the moving speed of the stirring head;

[0011] The rotation speed of the stirring head is controlled according to the rotation speed change rate so that the stirring head reaches the target rotation speed when it moves to the next welding area.

[0012] Through the above embodiment, the system collects the temperature of the current position of the stirring head in real time through the temperature sensor set in the welding area, and predicts the temperature of the stirring head when it reaches the next welding area based on the moving speed of the stirring head and the preset welding area length. Based on the predicted temperature and the target temperature range, the system dynamically calculates and adjusts the rotation speed of the stirring head. Compared with the traditional closed-loop control method that only relies on current temperature feedback, this method predicts the temperature change trend in advance and performs feedforward control based on the motion parameters of the stirring head to deal with the slow response problem caused by thermal inertia and temperature control lag. It can reduce the temperature fluctuation in the welding area, improve the accuracy and stability of temperature control, thereby improving the uniformity of weld formation and welding quality, and meet the higher requirements of stir friction welding for dynamic fine control of heat input.

[0013] In some embodiments, before the step of collecting the current temperature of the welding area corresponding to the current position of the stirring head in real time by using a temperature sensor provided in the welding area, the method further includes:

[0014] Collect temperature change data of welding area under different welding materials, different welding thicknesses and different welding conditions to obtain a historical temperature database;

[0015] Filtering at least one set of target historical temperature data having the highest similarity to the current welding condition from the historical temperature database;

[0016] A predicted temperature correction value is calculated based on the target historical temperature data and the current welding area temperature, and the predicted temperature correction value is used to correct the predicted temperature value.

[0017] Through the above-described embodiment, before real-time temperature acquisition, the system first collects and builds a database of historical temperatures for different welding materials, thicknesses, and operating conditions. Based on this database, it selects the historical temperature data most similar to the current operating conditions to correct the predicted temperature. This technical feature enables the system to not only rely on existing real-time measurements and parameter calculations when predicting temperature, but also incorporates historical empirical data to further compensate for potential deviations in the prediction results. This improves the accuracy and robustness of temperature predictions, reduces the impact of material or operating condition fluctuations on the accuracy of the prediction model, and makes welding temperature control more precise and reliable.

[0018] In some embodiments, before the step of collecting the current temperature of the welding area corresponding to the current position of the stirring head in real time by using a temperature sensor provided in the welding area, the method further includes:

[0019] At least one auxiliary temperature sensor is provided in the moving direction of the stirring head in the welding path;

[0020] The auxiliary temperature sensor collects auxiliary temperature data of the welding area in front of the stirring head in real time;

[0021] Calculating the temperature gradient between the current position of the stirring head and the next welding area based on the auxiliary temperature data and the moving speed of the stirring head;

[0022] A feedforward compensation value is calculated based on the temperature gradient, and the feedforward compensation value is used to correct the predicted temperature value.

[0023] Through the above embodiment, the system collects the auxiliary temperature data of the welding area in front of the stirring head in real time, and based on this data and the moving speed of the stirring head, calculates the temperature gradient between the current position and the next area, and then obtains the feedforward compensation value to correct the predicted temperature. This method makes the temperature prediction not only limited to the current position of the stirring head, but also can perceive the temperature change trend of the front area in advance. Through the feedforward compensation mechanism, it can more effectively deal with complex situations such as local temperature changes and uneven heat diffusion of materials, improve the real-time and foresight of dynamic temperature prediction, effectively reduce temperature overshoot or hysteresis, further improve the stability of welding temperature control, and optimize the weld formation effect.

[0024] In some embodiments, before the step of collecting the current temperature of the welding area corresponding to the current position of the stirring head in real time by using a temperature sensor provided in the welding area, the method further includes:

[0025] Before starting welding, the geometric characteristics of the welding path are obtained by scanning the predetermined welding area;

[0026] Identifying characteristic regions in a welding path based on the geometric features, the characteristic regions including a path curvature variation region and a thickness variation region;

[0027] Based on the geometric features of the characteristic region and the welding path, a temperature correction coefficient corresponding to the position of the characteristic region is pre-calculated, and the temperature correction coefficient is used to correct the predicted temperature value.

[0028] Through the above-described embodiment, the system scans the predetermined weld area before welding begins to acquire the geometric characteristics of the weld path. It then identifies characteristic regions within the path (such as areas of varying curvature and thickness), and incorporates targeted temperature correction coefficients into the temperature prediction. This method enables the system to pre-adjust the temperature prediction model for different geometric regions within the weld path, fully accounting for the impact of geometric factors on heat conduction and distribution. By incorporating specialized temperature corrections into characteristic regions of the path, the adaptability and accuracy of temperature control can be improved, preventing abnormal temperature fluctuations caused by geometric variations in the path, and ensuring weld formation consistency and quality even under complex paths.

[0029] In some embodiments, after the step of predicting the predicted temperature value corresponding to the next welding area when the stirring head reaches the next welding area based on the current welding area temperature, the moving speed of the stirring head, and the preset welding area length, the method further includes:

[0030] Determine whether the current welding condition is a characteristic area of ​​the welding path;

[0031] If the current welding condition is a characteristic area, the corresponding temperature correction coefficient is selected to correct the predicted temperature value;

[0032] If the current welding condition is not a characteristic area, determine whether the similarity between the current welding condition and the historical temperature data is greater than a preset similarity threshold;

[0033] If so, the predicted temperature correction value is selected to correct the predicted temperature value;

[0034] If not, the feedforward compensation value is selected to correct the predicted temperature value.

[0035] Through the above embodiment, the system determines that the current welding condition is within a characteristic region and selects a temperature correction coefficient to correct the predicted temperature. Otherwise, it selects a predicted temperature correction value or a feedforward compensation value based on the similarity with historical temperature data. This method intelligently integrates multi-source correction information and dynamically switches to the optimal correction method based on the actual working conditions. This minimizes temperature prediction errors, improves the adaptive capabilities of the temperature control system, and ensures that the welding temperature remains within a reasonable range, thus ensuring high consistency and quality during the welding process.

[0036] In some embodiments, before the step of calculating the target rotation speed based on the predicted temperature value and the target temperature range corresponding to the next welding area so that the temperature of the next welding area is within the target temperature range when the stirring head reaches the next welding area, the step further includes:

[0037] Monitoring temperature changes of a plurality of welding areas within a preset time period in the past to obtain temperature fluctuation values ​​of the welding areas;

[0038] When the temperature fluctuation value exceeds a preset fluctuation threshold, obtaining a deviation amplitude value;

[0039] The width of the target temperature range of the next welding area is reduced according to the adjustment ratio corresponding to the deviation amplitude value.

[0040] Through the above-described embodiment, the system monitors temperature fluctuations in multiple welding zones over a period of time. When fluctuations exceed a threshold, it automatically narrows the target temperature range for the next welding zone. This method fully accounts for fluctuations in heat input during the welding process and improves the system's sensitivity to abnormal temperature fluctuations by dynamically adjusting the width of the target temperature range. When temperature fluctuations are large, limiting the target temperature range helps enhance temperature control precision and error correction capabilities, preventing weld quality degradation caused by excessive temperature fluctuations, thereby improving the stability and reliability of the entire welding process.

[0041] In some embodiments, after the step of predicting the predicted temperature value corresponding to the next welding area when the stirring head reaches the next welding area based on the current welding area temperature, the moving speed of the stirring head, and the preset welding area length, the method further includes:

[0042] Comparing the predicted temperature value with the target temperature range to obtain a temperature deviation;

[0043] When the predicted temperature value is greater than the upper limit of the target temperature range, reducing the moving speed of the stirring head by a first preset ratio;

[0044] When the predicted temperature value is less than or equal to the lower limit of the target temperature range, the moving speed of the stirring head is increased at a second preset ratio.

[0045] Through the above-described embodiment, the system compares the predicted temperature with the target temperature range. If the predicted temperature is above the upper limit, the stirring head movement speed is reduced; if it is below the lower limit, the movement speed is increased. This method achieves dual dynamic control of the welding area temperature by adjusting the stirring head movement speed and further coordinating the rotation speed. If the predicted temperature exceeds the target range, the speed is immediately adjusted to quickly restore the temperature range, effectively suppressing temperature deviations and achieving flexible control of welding heat input.

[0046] In a second aspect, the present application provides a friction stir welding control system, the friction stir welding control system comprising: one or more processors and a memory;

[0047] The memory is coupled to the one or more processors, and the memory is used to store computer program code, and the computer program code includes computer instructions. The one or more processors call the computer instructions so that the stir friction welding control system can implement a stir friction welding control method based on dynamic temperature prediction and control provided in the above embodiment, which will not be repeated here.

[0048] In a third aspect, the present application provides a computer-readable storage medium comprising instructions. When the instructions are executed on a friction stir welding control system, the friction stir welding control system can implement a friction stir welding control method based on dynamic temperature prediction and control provided in the above embodiment, which will not be described in detail here.

[0049] Fourthly, the present application provides a computer program product. When the computer program product is run on a stir friction welding control system, the stir friction welding control system can implement a stir friction welding control method based on dynamic temperature prediction and control provided in the above embodiment, which will not be repeated here.

[0050] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0051] 1. Combining dynamic temperature prediction with a multi-source data fusion correction mechanism, this system not only predicts temperature based on the current welding area's real-time temperature, the stirrer's travel speed, and historical data, but also incorporates feedforward compensation for auxiliary temperature sensors, as well as temperature correction coefficients for path geometry and characteristic areas. By comprehensively utilizing real-time data, historical databases, and environmental characteristics, the accuracy and adaptability of temperature prediction are significantly improved, enabling the temperature control system to dynamically and accurately adjust parameters based on complex working conditions. This resolves the hysteresis and instability issues inherent in single-feedback control, effectively ensuring highly consistent welding quality.

[0052] 2. An intelligent, adaptive temperature control mechanism based on working condition recognition intelligently selects the optimal temperature correction method based on multiple factors, including characteristic regions, historical data similarity, and real-time temperature gradients. It also dynamically adjusts both the movement and rotation speeds of the stirring head. When the predicted temperature deviates from the target range, the system adjusts the movement speed in real time to quickly restore the temperature to the ideal range, significantly improving the sensitivity and error correction capabilities of temperature control. This adaptive temperature control logic enables more refined and intelligent dynamic control of the welding process, significantly enhancing process robustness and product consistency.

[0053] 3. An active temperature fluctuation management mechanism has been introduced, enabling real-time monitoring of temperature fluctuations across multiple welding zones. When fluctuations exceed the specified range, the target temperature range is automatically narrowed, improving the system's ability to identify and respond to abnormal fluctuations. By adaptively adjusting the target range width, overshoot and hysteresis during temperature control are effectively reduced, ensuring smoother and more controllable heat input during welding. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 This is a flow chart of a friction stir welding control method based on dynamic temperature prediction and control in an embodiment of the present application;

[0055] Figure 2 This is another flow chart of a friction stir welding control method based on dynamic temperature prediction and control in an embodiment of the present application;

[0056] Figure 3 This is a schematic diagram of the structure of a physical device of the friction stir welding control system in the embodiment of the present application. DETAILED DESCRIPTION

[0057] The terms used in the following examples of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification and appended claims of this application, the singular expressions "a," "an," "said," "above," "the," and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to any or all possible combinations comprising one or more of the listed items.

[0058] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0059] For ease of understanding, the following describes the process of the method provided by this implementation. Figure 1 , which is a flow chart of a stir friction welding control method based on dynamic temperature prediction and control in an embodiment of the present application.

[0060] S101. Using a temperature sensor provided in the welding area, the current welding area temperature corresponding to the current position of the stirring head is collected in real time.

[0061] Among them, the welding area refers to the part of the workpiece where the stirring head currently acts, that is, the local area where the stirring head rotates and friction to generate heat and realize material connection; the stirring head refers to the tool used to generate friction heat and plastic stirring in stir friction welding, usually consisting of a shoulder and a stirring needle, and its rotation and movement directly affect the welding heat input.

[0062] Specifically, a temperature sensor monitors the temperature below the stirrer shoulder or on the workpiece surface in real time at a preset sampling frequency (e.g., 10 times per second). It converts the temperature signal into an electrical signal (e.g., voltage or current) and transmits it to the control system's signal processing module. The signal processing module performs preprocessing on the raw signal, including filtering and amplification, to remove noise interference. This module then generates a value that accurately reflects the current weld area temperature, tags it with a corresponding timestamp, and stores the information about the stirrer's position in a temporary system database for subsequent use.

[0063] S102 : Based on the current welding area temperature, the moving speed of the stirring head and the preset welding area length, predict the predicted temperature value corresponding to the next welding area when the stirring head reaches the next welding area.

[0064] Among them, the preset welding area length refers to dividing the entire welding path into several continuous temperature control intervals. The length of each interval is pre-set according to the welding process requirements (such as 50mm / interval) and is used to determine the distance from the current position to the next area; the next welding area refers to the next temperature control interval in the moving direction of the stirring head, which is adjacent to the current area and has no overlap.

[0065] Specifically, the control system first obtains the current welding area temperature from the temporary database , the current moving speed v of the stirring head and the preset welding area length L. According to the heat conduction theory, assuming that the heat input is uniform during the welding process and the material thermal physical parameters (such as thermal conductivity and specific heat capacity) are constant, the time t=L / v required for the stirring head to move from the current position to the next area is calculated. Then, based on the first-order inertia link model (where K is the temperature change rate coefficient, obtained by fitting historical data or calculating the heat transfer equation), the temperature of the next welding area In addition, the system can also introduce the influence factor of the stirring head rotation speed on the heat input, such as (ω is the current rotation speed, α is the heat input coefficient) to improve the prediction accuracy.

[0066] Alternatively, the system can employ an analytical method based on differential heat transfer equations. This method establishes a heat conduction model for the weld zone, treating the stirrer as a mobile heat source. The Fourier heat conduction equation is used to solve the temperature distribution in the next zone, and the predicted temperature is calculated based on boundary conditions (such as the initial workpiece temperature and ambient heat dissipation).

[0067] Optionally, the system can also adopt a data-driven machine learning model. Using the sample data of "current temperature - moving speed - area length - actual temperature" in the historical temperature database, train a neural network (such as an LSTM network) or a regression model, and directly output the predicted temperature value by inputting the current parameters; the model can be updated regularly with new data to improve the generalization ability.

[0068] It can be understood that other methods can also be used to achieve temperature prediction. For example, by combining finite element simulation to calculate the thermal field distribution in real time and predicting the temperature of the next area according to the moving trajectory of the stirring head, which is not limited here.

[0069] S103. Calculate the target rotation speed based on the predicted temperature value and the target temperature range corresponding to the next welding area, so that the temperature of the next welding area is within the target temperature range when the stirring head reaches the next welding area.

[0070] Among them, the target temperature range refers to the temperature interval ([T_min, T_max)) preset according to the welding process requirements to ensure that the weld microstructure properties and forming quality meet the standards; the target rotation speed refers to the rotation speed ω that the stirring head needs to reach when it arrives at the area to make the temperature of the next area fall within the target range.

[0071] Specifically, the system first compares the predicted temperature value with the target temperature range (T_min < t < T_max). If the predicted temperature value is greater than the target temperature range, it is determined that the predicted temperature is too high, and the rotation speed needs to be reduced to reduce the heat input. The target rotation speed ωt is calculated by the proportional control algorithm, such as where k1 is the adjustment coefficient and ωc is the current rotation speed. Similarly, if the predicted temperature value is less than the target temperature range, it is determined that the predicted temperature is too low, and the rotation speed needs to be increased to increase the heat input. At this time .

[0072] Optionally, the system can adopt a model predictive control (MPC) algorithm to establish a dynamic model of the stirring head rotation speed and temperature, optimize the rotation speed sequence of multiple future sampling periods within a finite time domain to ensure that the predicted temperature is within the target range, and at the same time meet the actuator rate limit; update the control input in real time through the rolling optimization strategy.

[0073] It can be understood that other methods can also be used to calculate the target rotation speed. For example, a rule-based fuzzy control algorithm can be used to generate the target rotation speed by setting a fuzzy rule table of "reduce speed when temperature is high" and "increase speed when temperature is low", combined with the predicted temperature deviation and the deviation change rate, which is not limited here.

[0074] Specifically, the system may also dynamically evaluate and regulate the temperature stability during the welding process before calculating the target rotation speed based on the predicted temperature value and the target temperature range: first, continuously monitor the temperature change data of multiple welding areas within a preset time period (such as 5 minutes) in the past, and obtain the temperature fluctuation value by calculating the range of the temperature sequence (the difference between the maximum and minimum values). For example, temperature data is collected 10 times per second, and 1000 data points in the last 300 seconds are taken to calculate ΔT=T_max-T_min; when the fluctuation value exceeds the preset threshold (such as 20°C), the deviation amplitude value (ΔT_excess=ΔT-threshold) is further calculated. For example, if the measured fluctuation value of 25°C exceeds the threshold by 5°C, then ΔT_excess=5°C, and according to the preset deviation amplitude-adjustment ratio mapping relationship (such as each 1°C excess corresponds to a 2% reduction in the target range width), the shrinkage of the target temperature range of the next welding area is calculated. For example, the original range width is 40°C, and the adjustment ratio is 5%. When the target temperature range is narrowed, the new width becomes 38°C, thereby improving the temperature control accuracy by narrowing the target temperature range width, suppressing the temperature fluctuation caused by unstable heat input, and ensuring the precision and stability of temperature control in the subsequent welding area.

[0075] S104 : Calculate the rotation speed change rate from the current rotation speed to the target rotation speed based on the current rotation speed of the stirring head, the target rotation speed, and the moving speed of the stirring head.

[0076] Specifically, the control system first obtains the current rotational speed, target rotational speed, and movement speed, and then obtains the preset welding area length from step S102. Based on the kinematic relationship, the time required for the stirring head to move from the current position to the next area is calculated as t = L / v. Then, based on the rotational speed change Δω = ωt - ωc, the rotational speed change rate k = Δω / L is calculated.

[0077] S105 , controlling the rotation speed of the stirring head according to the rotation speed change rate so that the stirring head reaches the target rotation speed when it moves to the next welding area.

[0078] Specifically, the control system dynamically calculates the target speed, ωc1 = ωc + ks, based on the rotational speed change rate k and the real-time travel distance s of the mixing head. The system uses a closed-loop control algorithm (such as PID control) to adjust the motor drive signal so that the actual speed tracks the calculated value. For example, when the mixing head travels a distance s1, the target speed is ωc + ks1. If the actual speed is lower than this value, the system increases the motor drive power; if it is higher, the drive power is reduced until the speed precisely matches ωt when the mixing head reaches the next area.

[0079] In the above embodiment, the system uses a temperature sensor installed in the welding area to collect the temperature of the current position of the stirring head in real time. It combines the movement speed of the stirring head and the preset welding area length to predict the temperature when the stirring head reaches the next welding area, and dynamically calculates and adjusts the rotation speed of the stirring head based on the predicted temperature and the target temperature range. Compared with the traditional closed-loop control method that only relies on current temperature feedback, this method predicts the temperature change trend in advance and performs feedforward control based on the motion parameters of the stirring head to address the slow response problem caused by thermal inertia and temperature control lag. It can reduce the temperature fluctuation in the welding area and improve the accuracy and stability of temperature control, thereby improving the uniformity of weld formation and welding quality, and meet the higher requirements of stir friction welding for dynamic and fine control of heat input.

[0080] After combining the above scenarios, the following is a more detailed description of the process of the method provided by this implementation. Figure 2 , is another flow chart of a stir friction welding control method based on dynamic temperature prediction and control in an embodiment of the present application.

[0081] S201. Obtain the current welding status.

[0082] Specifically, the system uses an integrated sensor network and data interface to collect and integrate multidimensional data from the welding process in real time, forming a complete description of the current working conditions. Data sources include: real-time temperature detected by temperature sensors, stirring head movement and rotation speeds fed by encoders, workpiece material and thickness parameters (obtained through code scanning or manual input), and welding path coordinate information (derived from pre-imported CAD models or real-time scan data). The system encapsulates this data into working condition data packets according to a preset format and stores them in a cache for subsequent logic calls.

[0083] S202: Determine whether the current welding condition is a characteristic area of ​​the welding path.

[0084] Specifically, the system first obtains the current position coordinates of the stirring head (via a servo motor encoder or visual positioning system) and extracts the curvature radius and thickness values ​​corresponding to the current position from the pre-stored welding path geometric feature data. Based on the preset threshold, if the curvature radius is less than 100mm or the thickness change exceeds 1mm / 50mm, it is determined to be a characteristic area (such as a curve or a thickness mutation section). For example, if the current position coordinates correspond to a path curvature radius of 80mm (<100mm threshold), it is marked as a curvature change area; if the thickness suddenly changes from 3mm to 5mm and the change distance is 40mm (<50mm threshold), it is marked as a thickness change area.

[0085] Furthermore, when the system determines that the current welding condition is a characteristic area of ​​the welding path, it enters step S203 to correct the predicted temperature value using the temperature correction coefficient; conversely, when the system determines that the current welding condition is not a characteristic area of ​​the welding path, it skips step S203 and directly executes step S204 to determine the similarity threshold.

[0086] S203: Select a corresponding temperature correction coefficient to correct the predicted temperature value.

[0087] Before starting welding, the system obtains the geometric features of the welding path by scanning the predetermined welding area; identifies the characteristic area in the welding path based on the geometric features; and pre-calculates the temperature correction coefficient corresponding to the position of the characteristic area based on the characteristic area and the geometric features of the welding path to correct the predicted temperature value.

[0088] Specifically, before the control system collects the temperature of the welding area at the current position of the stirring head in real time through the temperature sensor, it needs to complete the preprocessing of the geometric characteristics of the welding path: first, before starting welding, the predetermined welding area is scanned using equipment such as a 3D laser scanner and a structured light camera to obtain point cloud data of the welding path. After noise reduction and gridding, a three-dimensional geometric model is generated, from which geometric features such as the path curvature radius and the workpiece thickness distribution are extracted (such as a bend with a curvature radius less than 100mm and an area with a sudden thickness change exceeding 1mm); then, based on the geometric features, the characteristic areas in the welding path are identified, and by setting a curvature radius threshold (such as 100mm) and a thickness change threshold (such as 1mm / 50mm), the path curvature change area (such as a continuous bending section) and the thickness change area (such as a sudden thickness change transition section) are marked, and the position of the characteristic area is stored in the form of a coordinate interval; finally, according to the characteristic area type and geometric parameters (such as curvature radius, thickness change), a preset formula (such as the curvature coefficient calculation formula) or finite element simulation is used to fit the temperature correction coefficient. For example, when the curvature radius is 50mm, the curvature coefficient is 0.9, and the thickness is 5mm relative to 3mm. The curvature coefficient ratio is 1.07, and the temperature correction coefficient is associated with the position of the characteristic area and stored, which is used to make targeted corrections to the predicted temperature value during the welding process to compensate for the influence of geometric factors on heat conduction.

[0089] S204: Determine whether the similarity between the current welding condition and the historical temperature data is greater than a preset similarity threshold.

[0090] Specifically, the system extracts key parameters from the current operating conditions: material type (e.g., aluminum alloy 6061), thickness (4mm), travel speed (60mm / min), and rotation speed (1000rpm). These parameters are encoded as a feature vector V_c = [6061, 4, 60, 1000]. Records of the same material type are then filtered from the historical temperature database. The thickness and speed parameters are normalized (e.g., thickness normalization = 4 / 6 = 0.67, travel speed normalization = 60 / 100 = 0.6). The cosine similarity algorithm is then used to calculate similarity with each historical record. For example, if a historical record V_h = [6061, 3, 50, 1200] is normalized and Sim = 0.85, which exceeds the 0.8 threshold, it is considered acceptable as reference data.

[0091] Furthermore, if the detected similarity is greater than the preset similarity threshold, the process proceeds to step S205 to correct the predicted temperature value using the predicted temperature correction value; otherwise, the process proceeds to step S206 to directly correct the predicted temperature value using the feedforward compensation value.

[0092] S205: Select a predicted temperature correction value to correct the predicted temperature value.

[0093] The system collects temperature change data of the welding area under different welding materials, different welding thicknesses and different welding conditions to obtain a historical temperature database; at least one set of target historical temperature data with the highest similarity to the current welding condition is screened out from the historical temperature database; and a predicted temperature correction value is calculated based on the target historical temperature data and the current welding area temperature to correct the predicted temperature value.

[0094] Specifically, before the control system uses the temperature sensor to collect the temperature of the welding area at the current position of the stirring head in real time, it must first build a historical temperature database and filter and correct the data accordingly. Through multiple sets of welding tests, for different welding materials (such as aluminum alloy, titanium alloy, etc.), welding thickness (such as 2mm, 5mm, etc.) and welding working condition combinations (including parameters such as stirring head rotation speed, moving speed, shoulder pressure, etc.), the temperature sensor is used to collect the temperature change data of the welding area at a high frequency (such as 20 times per second), and the process parameters and timestamps are recorded simultaneously. After denoising and normalization, they are classified and stored according to the "material-thickness-working condition" dimension to form a historical temperature database containing temperature curves, process parameters and weld quality evaluation; then, after welding is started, the current welding working condition parameters (such as material type, workpiece thickness, moving speed, etc.) are obtained and encoded into feature vectors. The records with the closest thickness and speed parameters under the same material type are retrieved in the historical database through algorithms such as Euclidean distance and cosine similarity, and at least one set of target historical temperature data with the highest similarity to the current working condition is screened out (such as selecting the top 3 in descending order of similarity). Finally, the initial temperature value corresponding to the current position in the target historical data is extracted and compared with the currently measured initial temperature of the workpiece to calculate the temperature deviation. Combined with the temporal trend of the historical temperature curve and the current temperature change rate, the predicted temperature correction value is obtained (such as the compensation amount calculated by linear deviation or dynamic curve error). This is used to calibrate the temperature value predicted based on the real-time temperature and motion parameters in order to compensate for the prediction deviation caused by factors such as material differences and environmental changes.

[0095] S206: Select a feedforward compensation value to correct the predicted temperature value.

[0096] At least one auxiliary temperature sensor is set in the moving direction of the stirring head in the welding path; the auxiliary temperature data of the welding area in front of the stirring head is collected in real time through the auxiliary temperature sensor; based on the auxiliary temperature data and the moving speed of the stirring head, the temperature gradient between the current position of the stirring head and the next welding area is calculated; based on the temperature gradient, a feedforward compensation value is calculated to correct the predicted temperature value.

[0097] Specifically, before the control system collects the temperature of the welding area at the current position of the stirring head in real time through the temperature sensor, it needs to complete the preprocessing of auxiliary temperature sensing and feedforward compensation. First, according to the welding process requirements, at least one auxiliary temperature sensor (such as an infrared sensor or thermocouple) is installed on the welding path within a range of 50-200mm in front of the stirring head's moving direction. The sensor can be arranged at a single point or in an equidistant array (such as a 3-sensor group with a spacing of 30mm). It is connected to the control system via a high-temperature resistant cable or wireless transmission to ensure that its detection area covers the next welding area that the stirring head is about to reach. During the welding process, the auxiliary temperature sensor collects the temperature data of the welding area in front in real time at a frequency of 10-50Hz. After filtering, amplification and analog-to-digital conversion by the signal conditioning module, auxiliary temperature data with a timestamp and position mark is generated (such as the temperature 200℃ at 100mm in front). Then, the control system combines the current stirring head movement speed (such as 60mm / min) and the auxiliary temperature data to calculate the temperature gradient between the current position and the next area. If it is a single sensor, the gradient G = (front temperature - current temperature) / distance. If it is multiple sensors, the slope of the position-temperature curve is fitted by linear regression to obtain G (unit: ℃ / mm). Finally, according to the temperature gradient G Based on the preset welding area length L (e.g., 50 mm), the feedforward compensation value ΔT_ff=G×L (e.g., ΔT_ff=100°C when G=2°C / mm) is calculated and added to the temperature value predicted based on the real-time temperature and motion parameters to compensate for heat conduction lag and temperature change trend in advance, thereby improving the accuracy of the predicted temperature.

[0098] Specifically, after calculating the predicted temperature for the next weld zone based on the current weld zone temperature, the stir head speed, and the preset weld zone length, the system compares the predicted temperature with the target temperature range. If the predicted temperature exceeds the upper limit of the target temperature range, the system calculates the upper deviation and reduces the stir head speed by a first preset ratio (e.g., a 5% reduction for deviations ≤10°C and a 10% reduction for deviations >10°C) to reduce heat input per unit time. For example, if the current speed is 100 mm / min, the system reduces it by 5% to 95 mm / min. If the predicted temperature is less than or equal to the lower limit of the target temperature range, the system calculates the lower deviation and increases the stir head speed by a second preset ratio (e.g., a 4% increase for deviations ≤10°C and an 8% increase for deviations >10°C) to increase heat input per unit length. For example, if the current speed is 80 mm / min, the system increases it by 4% to 83.2 mm / min. Through this two-way dynamic adjustment mechanism, combined with rotational speed control, the system achieves precise control of weld heat input, ensuring that the weld zone temperature remains stable within the process requirements.

[0099] The friction stir welding control system of the embodiment of the present invention is applied to electronic equipment. Figure 3 A schematic diagram of the architecture of an electronic device suitable for implementing an embodiment of the present invention is shown.

[0100] It should be noted that Figure 3 The electronic device shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0101] Those skilled in the art will appreciate that all or part of the steps in the various methods of the above embodiments can be accomplished by instructions (computer programs) or by controlling related hardware through instructions (computer programs), and the instructions can be stored in a computer-readable storage medium and loaded and executed by a processor. The electronic device of this embodiment includes a storage medium and a processor, wherein the storage medium stores a plurality of instructions, which can be loaded by the processor to execute any step of the method provided in the embodiment of the present invention.

[0102] Specifically, the storage medium and the processor are electrically connected directly or indirectly to achieve data transmission or interaction. For example, these elements can be electrically connected to each other via one or more signal lines. The storage medium stores computer-executable instructions for implementing the data access control method, including at least one software function module that can be stored in the storage medium in the form of software or firmware. The processor executes various functional applications and data processing by running the software programs and modules stored in the storage medium. The storage medium can be, but is not limited to, random access memory (RAM), read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), etc. The storage medium is used to store programs, and the processor executes the programs after receiving the execution instructions.

[0103] Furthermore, the software programs and modules in the above-mentioned storage medium may also include an operating system, which may include various software components and / or drivers for managing system tasks (such as memory management, storage device control, power management, etc.), and may communicate with various hardware or software components to provide an operating environment for other software components. The processor may be an integrated circuit chip having signal processing capabilities. The above-mentioned processor may be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc., which may implement or execute the various methods, steps, and logic flow diagrams disclosed in this embodiment. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0104] Since the instructions stored in the storage medium can execute the steps of any method provided in the embodiments of the present invention, the beneficial effects of any method provided in the embodiments of the present invention can be achieved. Please refer to the previous embodiments for details and will not be repeated here.

[0105] The above description is merely a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A friction stir welding control method based on dynamic temperature prediction and control, applied to a friction stir welding control system, characterized in that: The method comprises: The temperature sensor set in the welding area collects the current welding area temperature corresponding to the current position of the stirring head in real time; Based on the current welding area temperature, the moving speed of the stirring head and the preset welding area length, predicting the predicted temperature value corresponding to the next welding area when the stirring head reaches the next welding area; Calculating a target rotation speed based on the predicted temperature value and a target temperature range corresponding to the next welding area so that the temperature of the next welding area is within the target temperature range when the stirring head reaches the next welding area; Calculating a rotation speed change rate from the current rotation speed to the target rotation speed based on the current rotation speed of the stirring head, the target rotation speed, and the moving speed of the stirring head; The rotation speed of the stirring head is controlled according to the rotation speed change rate so that the stirring head reaches the target rotation speed when it moves to the next welding area.

2. The method according to claim 1, characterized in that Before the step of collecting the current temperature of the welding area corresponding to the current position of the stirring head in real time by using a temperature sensor provided in the welding area, the method further includes: Collect temperature change data of welding area under different welding materials, different welding thicknesses and different welding conditions to obtain a historical temperature database; Filtering at least one set of target historical temperature data having the highest similarity to the current welding condition from the historical temperature database; A predicted temperature correction value is calculated based on the target historical temperature data and the current welding area temperature, and the predicted temperature correction value is used to correct the predicted temperature value.

3. The method according to claim 2, characterized in that Before the step of collecting the current temperature of the welding area corresponding to the current position of the stirring head in real time by using a temperature sensor provided in the welding area, the method further includes: At least one auxiliary temperature sensor is provided in the moving direction of the stirring head in the welding path; The auxiliary temperature sensor collects auxiliary temperature data of the welding area in front of the stirring head in real time; Calculating the temperature gradient between the current position of the stirring head and the next welding area based on the auxiliary temperature data and the moving speed of the stirring head; A feedforward compensation value is calculated based on the temperature gradient, and the feedforward compensation value is used to correct the predicted temperature value.

4. The method according to claim 3, characterized in that Before the step of collecting the current temperature of the welding area corresponding to the current position of the stirring head in real time by using a temperature sensor provided in the welding area, the method further includes: Before starting welding, the geometric characteristics of the welding path are obtained by scanning the predetermined welding area; Identifying characteristic regions in a welding path based on the geometric features, the characteristic regions including a path curvature variation region and a thickness variation region; Based on the geometric features of the characteristic region and the welding path, a temperature correction coefficient corresponding to the position of the characteristic region is pre-calculated, and the temperature correction coefficient is used to correct the predicted temperature value.

5. The method according to claim 4, characterized in that After the step of predicting the predicted temperature value corresponding to the next welding area when the stirring head reaches the next welding area based on the current welding area temperature, the moving speed of the stirring head and the preset welding area length, the method further includes: Determine whether the current welding condition is a characteristic area of ​​the welding path; If the current welding condition is a characteristic area, the corresponding temperature correction coefficient is selected to correct the predicted temperature value; If the current welding condition is not a characteristic area, determine whether the similarity between the current welding condition and the historical temperature data is greater than a preset similarity threshold; If so, the predicted temperature correction value is selected to correct the predicted temperature value; If not, the feedforward compensation value is selected to correct the predicted temperature value.

6. The method according to claim 1, wherein Before the step of calculating the target rotation speed based on the predicted temperature value and the target temperature range corresponding to the next welding area so that the temperature of the next welding area is within the target temperature range when the stirring head reaches the next welding area, the method further includes: Monitoring temperature changes of a plurality of welding areas within a preset time period in the past to obtain temperature fluctuation values ​​of the welding areas; When the temperature fluctuation value exceeds a preset fluctuation threshold, obtaining a deviation amplitude value; The width of the target temperature range of the next welding area is reduced according to the adjustment ratio corresponding to the deviation amplitude value.

7. The method according to claim 1, characterized in that After the step of predicting the predicted temperature value corresponding to the next welding area when the stirring head reaches the next welding area based on the current welding area temperature, the moving speed of the stirring head and the preset welding area length, the method further includes: Comparing the predicted temperature value with the target temperature range to obtain a temperature deviation; When the predicted temperature value is greater than the upper limit of the target temperature range, reducing the moving speed of the stirring head by a first preset ratio; When the predicted temperature value is less than or equal to the lower limit of the target temperature range, the moving speed of the stirring head is increased at a second preset ratio.

8. A friction stir welding control system, characterized in that: The friction stir welding control system includes: one or more processors and memories; The memory is coupled to the one or more processors, and is configured to store computer program codes, wherein the computer program codes include computer instructions. The one or more processors call the computer instructions to enable the friction stir welding control system to execute the method according to any one of claims 1 to 7.

9. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on a friction stir welding control system, the friction stir welding control system is caused to execute the method according to any one of claims 1 to 7.

10. A computer program product, characterized in that When the computer program product is run on a friction stir welding control system, the friction stir welding control system is enabled to perform the method according to any one of claims 1 to 7.

Images