A contralateral synchronous heating stress control method for thick-walled component welding

Abstract

This invention relates to a method for controlling stress in the opposite-side synchronous heating during welding of thick-walled components. The method includes: setting a first temperature sensing element near the area to be welded on the thick-walled component to obtain the temperature of the welding zone; setting a heater with controllable power and a second temperature sensing element in the opposite-side region, where the second temperature sensing element obtains the temperature of the opposite-side heating zone; applying an initial power to the heater for feedforward preheating during arc initiation; and dynamically adjusting the output power of the heater based on a preset temperature difference target value using a closed-loop control algorithm, according to the actual difference between the welding zone temperature and the opposite-side heating zone temperature, so that the actual difference approaches the temperature difference target value. This invention provides a novel technical solution that actively and precisely intervenes in the temperature field while the welding thermal cycle is in progress, thereby suppressing stress at its source.

Description

Technical Field

[0001] This invention relates to the field of welding technology, and in particular to a method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components. Background Technology

[0002] In modern large-scale engineering construction, such as high-rise buildings, long-span bridges, marine engineering platforms, and pressure equipment manufacturing, the application of thick-walled steel structural components is becoming increasingly widespread. Among these, the connection nodes between thick-walled box columns and H-beams are typical critical load-bearing areas. Box columns and similar components typically have wall thicknesses of tens or even hundreds of millimeters and enormous overall dimensions, making them physically large heat-carrying bodies. During welding operations, such as welding the relatively small flanges or webs of H-beams to the box column walls, the box column body exhibits extremely strong absorption and conduction capabilities for welding heat, forming a natural, powerful heat dissipation body.

[0003] This structural feature causes the high-temperature heat generated by the welding arc during welding to be rapidly absorbed by the box-shaped column base material and dissipated over distances. The direct consequence is the instantaneous formation of a dramatic temperature gradient of several hundred degrees Celsius between the weld pool and the heat-affected zone near the weld and the column base material just centimeters away. According to the fundamental physical laws of thermal expansion and contraction, this spatially uneven temperature field will inevitably lead to highly inconsistent shrinkage deformation in different regions within the component during subsequent cooling. Because the material is constrained by the surrounding area during cooling and contraction, this inconsistent deformation will eventually be "solidified" within the component as residual stress, simultaneously triggering macroscopic welding deformations such as angular deformation and out-of-plane deformation. Among these, the residual tensile stress formed in the weld and heat-affected zone is particularly harmful, significantly reducing the structure's fatigue strength and brittle fracture resistance, and under certain conditions, may even induce delayed cracking, posing a serious threat to the long-term service safety of the structure.

[0004] To address this technical challenge, existing technologies typically employ auxiliary processes. Traditional methods primarily include preheating before welding and post-weld heat treatment (PWHT). Preheating involves heating the entire component or the area to be welded to a predetermined temperature before welding begins. Its purpose is to slow down the cooling rate after welding and reduce the tendency for hard and brittle structures to form. However, for large components, overall preheating means enormous energy consumption and a long heating time, and it can only reduce the initial temperature gradient, failing to cope with the drastic temperature differences that occur dynamically during welding. Post-weld heat treatment involves placing the entire weldment in a heat treatment furnace after welding, heating it to a certain temperature and holding it there, then slowly cooling it to relax internal stresses. While this method is effective, its drawbacks are also significant: extremely high energy consumption, extremely long processing time (often measured in days), high cost, and for ultra-large components installed on-site, implementing overall heat treatment is almost impractical.

[0005] Both preheating before welding and post-weld heat treatment are essentially passive or reactive stress control methods. They either create initial conditions before welding begins or attempt to remediately eliminate stress after it has formed, failing to proactively intervene in and manage the dynamically changing temperature field during the welding process. Therefore, there is an urgent need for an economical and efficient new technology that can proactively and precisely intervene in the temperature field while the welding thermal cycle is underway, thereby suppressing stress at its source. Summary of the Invention

[0006] In order to solve the technical problems of the prior art to a certain extent, the present invention provides a method for controlling stress by synchronous heating on the opposite side for welding thick-walled components and a cutting production line structure, which can actively and accurately intervene in the temperature field while the welding thermal cycle is in progress, thereby suppressing the stress at the source.

[0007] This invention discloses a method for controlling the stress of simultaneous heating on the opposite side during welding of thick-walled components, comprising the following steps:

[0008] A first temperature measuring element is set near the area to be welded in the thick-walled component to obtain the temperature of the welding area; and a heater with controllable power and a second temperature measuring element are set in the opposite side area to the area to be welded, and the temperature of the opposite side heating area is obtained through the second temperature measuring element.

[0009] When the welding arc is initiated, an initial power is applied to the heater for feedforward preheating;

[0010] During the welding process, based on a preset temperature difference target value, the output power of the heater is adjusted in real time and dynamically according to the actual difference between the temperature of the welding zone and the temperature of the opposite heating zone through a closed-loop control algorithm, so that the actual difference approaches the temperature difference target value.

[0011] After the welding process is completed, the power of the heater is continued to be controlled so that the cooling rate of the opposite heating zone matches the cooling rate of the welding zone, so as to achieve synchronous cooling.

[0012] Specifically, the core of this invention lies in constructing a dynamic thermal balance system spanning the component wall thickness by applying a dynamic and controllable auxiliary heat source opposite the welding heat source, thereby actively managing the temperature field during the welding process. The technical solution of this method can be decomposed into the following interrelated and synergistic technical features:

[0013] First, a first temperature sensing element is placed near the area to be welded to capture the critical disturbance signal of the welding zone temperature in real time. Simultaneously, a controllable power heater and a second temperature sensing element are placed on the opposite side of the component (the opposite area). This deployment forms the basis of closed-loop control: the actual temperature on the opposite side is obtained through the second temperature sensing element as a feedback signal, and then the temperature in that area is adjusted through the heater, which acts as an actuator. These two subsystems are spatially aligned and functionally correspond to a monitoring-execution relationship, forming the physical prerequisite for synchronous control on the opposite side. Second, at the initial stage of the welding process, i.e., the instant the welding torch ignites the arc, the system executes a feedforward preheating action. Based on preset welding parameters, an initial power is applied to the heater through a feedforward model. This step embodies predictive control of the welding process. Since arc ignition is a drastic heat input process, relying solely on delayed feedback control is insufficient for instantaneous response, leading to a sharp increase in the initial temperature difference. The role of feedforward preheating is to proactively inject heat into the opposite side before a temperature difference forms, offsetting some of the impact of the upcoming welding heat input. This creates a smoother initial condition for subsequent closed-loop control, thereby improving the speed and stability of the entire control process. Next, during the welding process, the system enters the core closed-loop control stage. This stage uses a preset temperature difference target value as the control benchmark. This target value is not zero, but a small, permissible temperature difference. The control system continuously collects the temperatures of the welding zone and the opposite zone, calculating the actual difference between them. The closed-loop control algorithm (such as a PID algorithm) compares this actual difference with the preset target value, and adjusts the output power of the opposite heater in real time and continuously based on the resulting error. This process is a dynamic optimization and balancing process: when the welding heat input increases, causing the actual temperature difference to increase and deviate from the target value, the controller increases the heating power on the opposite side to "catch up" with the temperature rise of the welding zone; conversely, it decreases the power. This real-time feedback adjustment based on temperature difference is the core mechanism to ensure that the thermal behavior of both sides remains dynamically synchronized throughout the welding process. Finally, when the welding process ends and the component enters the cooling stage, this method does not simply stop all intervention. Instead, power control is continued on the heater on the opposite side to match the cooling rate of the heating zone with the natural cooling rate of the welding zone, achieving synchronous cooling. This step manages the integrity of the welding thermal cycle. Because the final formation of residual stress mainly occurs during the cooling and contraction phase, if the welding zone cools rapidly while the opposite zone has cooled, a harmful temperature gradient will also be generated. Through controlled synchronous cooling, it can be ensured that the contraction behavior of the inner and outer sides of the component tends to be consistent throughout the entire process of losing plasticity, thereby further suppressing and releasing residual stress.

[0014] Preferably, the closed-loop control algorithm is a PID control algorithm, and the formula for calculating the actual difference is:

[0015]

[0016] in, This represents the actual temperature difference. Temperature of the welding zone This refers to the temperature of the heating zone on the opposite side.

[0017] The PID control algorithm is based on error Calculate the heater power adjustment amount.

[0018] in This is the preset target temperature difference value.

[0019] Specifically, in this scheme, the control system first collects the temperature of the welding zone. and the temperature of the opposite heating zone Calculate the actual temperature difference between the two. Subsequently, this actual temperature difference is compared with the target temperature difference value preset in the process. The system error e is obtained through comparison. This error e serves as the input to the PID control algorithm, driving the controller to calculate the adjustment amount for the heater power. This transforms a complex physical process control problem into a classic industrial control problem with mature solutions. The PID control algorithm utilizes its proportional (P) element to respond quickly to the current error, its integral (I) element to eliminate static errors caused by system thermal inertia or continuous heat loss, and its derivative (D) element to predict error trends to suppress overshoot and oscillations. Therefore, concretizing the closed-loop control algorithm into PID control enables the entire control system to possess good dynamic response speed, steady-state control accuracy, and system stability, thereby ensuring that the actual temperature difference is stably and accurately maintained near the preset target value throughout the welding process.

[0020] Furthermore, the PID control algorithm is set as an incremental PID algorithm, which outputs a power adjustment amount for dynamically adjusting the heater power. The calculation formula is:

[0021]

[0022] in,

[0023] This is the proportionality coefficient;

[0024] The integral coefficient;

[0025] Here are the differential coefficients:

[0026] This represents the error at the current moment;

[0027] Error from the previous moment

[0028] This represents the error from the first two time points.

[0029] In other words, this scheme further modifies the PID control algorithm into an incremental PID algorithm, whose output is the adjustment amount of the heater power. It refers to the power, not the absolute value of the power.

[0030] Specifically, within each control cycle, the control system first calculates the power adjustment ΔP based on the current error e(t) and historical error values ​​e(t-1) and e(t-2), using a specific mathematical formula containing proportional, integral, and derivative terms. This adjustment is then added to the heater power value from the previous cycle to obtain the new power setpoint to be applied to the heater in the current cycle. In this mathematical formula, the proportional term (… The response is to the change in error, the integral term ( ) is used to eliminate steady-state error, while the differential term ( This method predicts and suppresses the trend of error changes. The technical advantage of calculating power adjustment amounts rather than absolute power values ​​is that it significantly improves the stability and reliability of the control system. Firstly, because the algorithm outputs power increments, when the control system experiences disturbances (such as brief communication interruptions or controller resets), the output of the actuator (heater) will not undergo a drastic jump from zero to a large absolute value. Instead, after the disturbance disappears, it continues to fine-tune from the previous effective power value. This avoids sudden changes in heating power caused by control interruptions, ensuring a smooth and continuous heating process. Secondly, the incremental algorithm's structure naturally suppresses integral saturation. In traditional absolute PID control, when the error remains large for a long time, the integral term accumulates excessively, leading to a large overshoot when the error reverses. Incremental algorithms integrate the error itself and introduce the difference of the error into the calculation, making the control action pay more attention to the changes in the error. This reduces the overshoot when the system recovers from a large deviation, resulting in a smoother dynamic response in the entire closed-loop control process. In particular, it can achieve a smoother transition when the system starts or stops or the control target is switched, ultimately helping to achieve more accurate and stable control of the temperature difference.

[0031] Preferably, both the first and second temperature sensing elements are K-type armored thermocouples. The first temperature sensing element is located 10-15 mm outside the area to be welded, and the second temperature sensing element is located close to the geometric center of the opposite heating area.

[0032] Specifically, this scheme explicitly uses K-type armored thermocouples as the first and second temperature sensing elements. K-type thermocouples are widely used in industrial temperature measurement due to their wide temperature range, good linearity, and high stability in high-temperature regions, accurately covering temperature changes during the welding process. The armored structure provides physical and electromagnetic shielding for the temperature sensing elements, effectively resisting interference from welding spatter, arc radiation, and strong electromagnetic fields. Simultaneously, the scheme precisely lays out the spatial positions of the temperature sensing points: the first temperature sensing element is placed 10-15mm outside the area to be welded. This position avoids direct damage from the weld pool and accurately captures the temperature of the heat-affected zone, which is most directly related to the formation of residual stress, providing the most critical disturbance signal for control. The second temperature sensing element is placed close to the geometric center of the opposite heating area to ensure that the temperature it acquires represents the core thermal state of the entire opposite controlled area. By defining the sensor type and location as described above, the accuracy, real-time performance, and anti-interference capability of the temperature signal acquisition are guaranteed from the physical source, providing a high-fidelity input for the entire closed-loop control system, which is a prerequisite for achieving high-precision temperature field control.

[0033] Preferably, the heater is an induction heater, and the induction heater includes a digital medium-frequency induction heating power supply and a planar rectangular induction coil, the size of which is larger than the projected area of ​​the area to be welded.

[0034] It is understandable that this scheme utilizes existing induction heater technology, specifically consisting of a digital medium-frequency induction heating power supply and a planar rectangular induction coil. Induction heating technology is characterized by non-contact operation, rapid heating speed, high power density, and ease of precise control. Its response speed can reach the millisecond level, fully meeting the high-frequency adjustment requirements of PID controllers. The digital medium-frequency induction heating power supply accepts standard control signals and can achieve linear and stepless power adjustment, providing execution assurance for the precise output of the control algorithm. The induction coil is designed as a planar rectangle to match the linear characteristics of typical weld seams, and its size is larger than the projected area of ​​the area to be welded to ensure that a uniform thermal field slightly larger than the weld influence zone is formed on the opposite side, thus providing a covering thermal support for the welding area, rather than point-to-point heating. Therefore, the induction heater with this configuration provides an efficient, fast, and precise heating execution unit for this method, ensuring that the power command issued by the control system can be faithfully and rapidly converted into heat input on the opposite side, which is a key execution guarantee for achieving dynamic thermal balance control.

[0035] Furthermore, the initial power is determined by a feedforward model based on preset welding current parameters, and the feedforward preheating automatically starts the instant the welding arc is detected. In other words, this scheme introduces a feedforward preheating step at the instant of welding arc ignition. Specifically, the control system monitors the current state of the welding circuit in real time using external sensors (such as Hall effect current sensors). Once a jump from zero to the welding operating current is detected, an arc ignition event is determined. At the instant of arc ignition, the control system does not wait for feedback from the closed-loop control algorithm, but immediately applies an initial power to the opposite heater based on a preset feedforward model. This feedforward model is essentially a mapping relationship established based on process experience data or thermal simulation, which correlates the input welding parameters (such as welding current, voltage, plate thickness, etc.) with an initial heating power required to effectively suppress the initial thermal shock. The technical effect of introducing this feedforward preheating step is that it effectively compensates for the inherent delay of the closed-loop feedback control system. At the moment welding begins, the temperature in the welding zone rises sharply due to the enormous energy input of the electric arc. If relying solely on feedback control, the system must go through a series of time-lag processes: "measuring the temperature difference - calculating the error - adjusting the power." During this time, a significant temperature gradient has already formed. Feedforward control, an open-loop, predictive control method, proactively and proactively applies a compensatory control action (initial heating power) to the system the instant a disturbance (i.e., arc ignition) is detected. This initial power counteracts the intense heat input generated by the welding arc, thus reducing the actual temperature difference before the closed-loop control system begins to function effectively. The initial peak value is suppressed within a small range. This can also help reduce the initial error e(0) faced by the PID closed-loop algorithm of the subsequent takeover control, avoid the controller from making drastic adjustment actions and possible overshoot oscillations due to excessive initial error, and enable the system to enter a stable tracking state more quickly. This improves the response speed and stability of the entire control system in the initial stage of welding, and ensures that the temperature gradient is effectively managed from the beginning of the welding process.

[0036] Preferably, before arc initiation, a pre-heating step is included: the heater is activated to uniformly heat the entire opposite side panel of the thick-walled component with low power over a large area, uniformly raising the reference temperature of the entire opposite side panel by a preset amount, thereby reducing the initial temperature gradient between the welding zone and the component base material at the moment of arc initiation. Specifically, this scheme further adds a pre-heating step before welding. This step involves using the opposite heater to uniformly heat the entire opposite side panel of the thick-walled component with low power over a large area before the formal welding and synchronous heating begin, thereby raising its reference temperature by a preset amount. The thick-walled component itself is a large heat sink, a characteristic that is unfavorable for welding when cold. This step cleverly transforms its physical properties from a "strong heat sink" to a warm "heat buffer" by pre-injecting a certain amount of heat energy into the component. The technical effect of this method is that it helps to reduce the severe initial temperature difference that should exist between the welding zone and the component base material at the moment of arc initiation. This not only meets the basic requirement of preheating for thick plate welding in traditional processes, but more importantly, it creates an initial working condition with a small temperature difference and easy control for subsequent PID synchronous control, thereby significantly improving the stability and convergence speed of the main control loop and avoiding overshoot or oscillation that may occur in the system during the startup phase due to severe disturbances.

[0037] Preferably, multiple auxiliary temperature measuring elements are arranged around the opposite heating area. The uniformity of pre-charging heating is determined by monitoring the temperature consistency between the auxiliary temperature measuring elements and the second temperature measuring element. Pre-charging heating is stopped when the average temperature of all measuring points reaches the preset pre-charging target temperature. Specifically, this scheme determines the uniformity of pre-charging heating by adding multiple auxiliary temperature measuring elements around the opposite heating area and monitoring the temperature consistency between these auxiliary temperature measuring points and the central temperature measuring point in real time. Pre-charging heating is stopped only when the average temperature of all measuring points reaches the preset target. This technical feature elevates the pre-charging target from a single "temperature target" to "uniform target target." Its technical effect is that it ensures that the initial temperature field on the opposite side of the component is a truly uniform and predictable reference plane before entering the main welding stage, avoiding the introduction of new and unexpected temperature gradients inside the component due to uneven pre-charging. By ensuring the uniformity of the initial conditions, a higher quality guarantee is provided for the accuracy and reliability of all subsequent control stages, thereby making the final stress control effect more consistent and repeatable.

[0038] Preferably, during the welding process, the control objective of the closed-loop control algorithm is to synchronize the temperature change rate of the welding zone with the temperature change rate of the opposite heating zone, and the error calculation formula is:

[0039]

[0040] Where e represents the error in the rate of temperature change. The temperature change rate in the welding zone, The temperature change rate of the opposite heating zone is denoted as , and the control objective is to dynamically approach zero this error.

[0041] This solution is understood to be a further optimization of the previous one, moving away from using a static temperature difference as the control target and instead upgrading it to the synchronization of a dynamic temperature change rate. Specifically, the control system will calculate the temperature change rate of the welding zone in real time. and the temperature change rate of the opposite heating zone The difference between the two is used as the system error input, and the control objective is to dynamically approach zero this error. This optimized scheme helps shift the focus of control from the "state" of the temperature field to its "trend." The root cause of welding residual stress lies in uneven cooling contraction, which is directly caused by the dynamically changing temperature gradient inside the component. Traditional temperature difference control compensates for the already formed temperature gradient, resulting in a certain lag. This scheme, however, directly controls the rate of temperature change on both sides, essentially intervening at the source of the temperature gradient. When fluctuations in welding heat input (such as changes in arc voltage or welder movement speed) cause a change in the heating rate of the welding zone, this control system can immediately detect the change in rate and quickly adjust the heating power on the opposite side to match this change, thus achieving dynamic balance of heat input before the temperature difference significantly expands. Therefore, this scheme provides a more responsive and reliable control scheme that can better maintain the consistency of the temperature field in the wall thickness direction, thereby effectively suppressing stress accumulation at the beginning stage.

[0042] Preferably, the rate of temperature change is calculated using the four-point central difference method:

[0043]

[0044] in, The temperature at the current moment. The temperature for the first two sampling periods, The sampling period.

[0045] It is understandable that in digital control systems, the differentiation operation at discrete sampling points easily amplifies the inherent measurement noise in the signal. If a simple difference between two adjacent points is used, the calculated rate of change signal will exhibit severe jitter, leading to oscillations in the control output and making the system unstable. This scheme employs the four-point center difference method, which calculates the slope using the temperature values ​​of the current moment and the previous two sampling periods, with the calculation interval spanning two sampling periods. This method is numerically equivalent to a smoothing filter, effectively suppressing the interference of high-frequency noise on the derivative calculation and obtaining a smoother and more realistic estimate of the temperature change rate. Furthermore, compared to other filtering methods, the center difference method has a theoretical advantage in providing a more accurate approximation of the derivative. Therefore, this specific calculation method provides a stable and reliable input signal for the core algorithm of rate control, forming the technical foundation for ensuring the stable operation of the entire closed-loop control system under high dynamic response, and ensuring the practical feasibility of the rate synchronization control strategy.

[0046] Preferably, the cooling stage after the welding process is completed also includes a high-temperature thermal pulse impact step: when the temperature of the welding zone is within a preset high-temperature range, the heater is controlled to output multiple high-power, short-duration thermal pulses, which generate instantaneous thermal expansion and contraction in the opposite region to promote stress relaxation inside the material; the preset high-temperature range is 600-800°C, the thermal pulse is a high-power pulse with a pulse width of milliseconds, and there is a preset time interval between multiple thermal pulses.

[0047] It is understandable that this scheme, when the component cools to a preset high-temperature range (e.g., 600-800°C), superimposes multiple high-power, short-duration thermal pulses on top of conventional slow cooling control. This temperature range is where most engineering steels exhibit good plasticity and low yield strength, making it a critical window for stress relaxation treatment. Within this range, the applied millisecond-level high-power thermal pulses induce intense and instantaneous thermal expansion and contraction in the opposite region. This effect is mechanically equivalent to a non-contact "thermal shock" or "thermal knock." Each impact excites transient stress waves within the material, providing additional driving energy for dislocation movement and slip within the microstructure. This periodic energy injection effectively disrupts the original stress field balance, promoting microscopic plastic deformation in local areas, thereby releasing and dissipating residual stress stored in the form of elastic strain energy. This method draws on the principle of vibration aging, but uses more easily controlled and integrated thermal energy as the excitation source, enabling in-depth treatment of residual stress without introducing any additional mechanical devices. Its technical effect lies in changing the simple macroscopic temperature field management and going further to the microscopic level to actively promote stress relaxation, thereby further breaking down and reorganizing the harmful stress network during the cooling process, achieving a more significant stress reduction effect than conventional slow cooling.

[0048] Preferably, during the cooling phase after the welding process, asymmetric rate control is used, setting the target cooling rate of the opposite heating zone to a specific proportion of the actual cooling rate of the welding zone:

[0049]

[0050] Where k is a pre-set proportionality coefficient less than 1, so that the cooling rate of the opposite heating zone is always slower than that of the welding zone, so as to form a beneficial stress balance during the cooling process.

[0051] Specifically, this scheme sets the target cooling rate of the opposite heating zone to a specific proportion (coefficient k) less than 1 of the actual cooling rate of the welding zone. This means that throughout the cooling process, the control system actively and precisely ensures that the cooling rate of the opposite region is always a fixed proportion slower than that of the welding zone. The underlying physical logic is that during the cooling phase, the welding zone and its heat-affected zone are the main sources of contraction and tensile stress. Through this scheme, the temperature of the opposite region at any given time will be slightly higher than its temperature under synchronous cooling (k=1). This relatively higher temperature puts it in a state of relative expansion, thereby generating a mild compressive stress field within the component that is opposite to the direction of the contraction and tensile stress in the welding zone, forming a "thermal support" or "thermal balance" effect. This actively introduced beneficial compressive stress can effectively counteract and balance the contraction stress during cooling. Therefore, this scheme elevates cooling control from a "passive following" to an "active guiding" level. By precisely controlling the differences in cooling rates between different regions, it actively constructs a favorable stress distribution within the component, thereby achieving optimized control of the final residual stress state, with effects superior to simple synchronous cooling.

[0052] The technical effect of the contralateral synchronous heating stress control method for welding thick-walled components, as described in this invention, is as follows: By setting up a synchronous heating system on the opposite side of the welding area, it directly acts on the root cause of stress—the non-uniform temperature field. During welding, the welding heat source causes the temperature of the welding area to rise sharply. Without intervention, the heat will be rapidly conducted into the cold base material of the component, forming a large temperature gradient along the wall thickness direction. This invention actively establishes an auxiliary thermal field on the opposite side through contralateral heating, and uses a closed-loop control algorithm based on temperature difference to make the temperature of the opposite region dynamically follow the temperature change of the welding area. In this way, the temperature difference along the component wall thickness direction is maintained in real time and actively within a small preset range. According to the principles of thermophysics, the thermal expansion of a material is proportional to the temperature change. Because the temperature gradient along the component wall thickness direction is significantly reduced, the thermal expansion behavior of the welding area and the opposite region tends to be consistent during the heating process, avoiding the accumulation of internal stress caused by one side expanding violently while the other side remains almost unchanged. During the cooling phase, without intervention after welding, the heat-affected zone (HAZ) will rapidly cool and shrink due to its small size and high temperature, while the large base material cools slowly. This uncoordinated shrinkage is the main cause of huge residual tensile stress. This invention controls the heater on the opposite side, forcing the cooling rate of the opposite region to match that of the welded zone, achieving synchronous shrinkage. This controlled synchronous cooling process ensures that the volume shrinkage of the inner and outer sides of the component remains consistent throughout the transition from a high-temperature plastic state to a low-temperature elastic state. Therefore, the internal stress caused by uneven cooling and shrinkage cannot fully develop and accumulate. Ultimately, because the temperature distribution inside the component is effectively balanced during the critical stages of the entire welding thermal cycle (including heating and cooling), the physical conditions for the formation of harmful residual stress are fundamentally suppressed. Therefore, the level of residual stress inside the component after welding is significantly reduced, and related welding deformations such as angular deformation and out-of-plane deformation can also be effectively controlled. Attached Figure Description

[0053] Figure 1 This is a flowchart of the core solution method of Embodiment 1 of the present invention.

[0054] Figure label: None. Detailed Implementation

[0055] To more clearly illustrate the technical solution of the present invention, the present invention will be described in detail below with reference to specific embodiments. It should be noted that the following embodiments are only used to explain the present invention and are not intended to limit the scope of protection of the present invention.

[0056] Example 1

[0057] This embodiment discloses a method for controlling stress by synchronous heating on the opposite side during welding of thick-walled components. Taking a common node in large steel structure engineering as an example, this embodiment specifically involves welding a 40mm thick H-beam flange plate to the wall plate of an 80mm thick Q355B box column in the form of a full penetration angle joint.

[0058] Combination Figure 1 As shown, firstly, to implement the control method of this invention, corresponding hardware deployment is required. A first temperature sensing element is installed approximately 10-15 mm outside the welding area of ​​the box-shaped column, near the root of the bevel of the H-beam flange. This first temperature sensing element is used to acquire the welding zone temperature in real time during the welding process. On the inner wall panel of the box-shaped column, strictly opposite to the area to be welded, i.e., the opposite side, a heater with controllable power and a second temperature sensing element are installed. The second temperature sensing element is used to acquire the temperature of the opposite heating zone in real time. In this embodiment, to ensure the accuracy of temperature measurement and stability in a high-temperature welding environment, both the first and second temperature measuring elements are preferably K-type armored thermocouples. The arrangement of the first temperature measuring element can effectively reflect the temperature of the heat-affected zone most relevant to the formation of residual stress, while avoiding direct damage from the welding arc and molten metal. The second temperature measuring element is located at the geometric center of the opposite heating area to accurately reflect the core temperature of that area.

[0059] In this embodiment, the heater is an induction heater, which mainly consists of a digital medium-frequency induction heating power supply and a matching planar rectangular induction coil. Induction heating is characterized by non-contact operation, high thermal efficiency, rapid power response, and easy precise control, making it very suitable for this embodiment. The induction coil is fixedly installed on the inner wall of the box-shaped column, and its coverage area is designed to be slightly larger than the projected area of ​​the external area to be welded on the inner wall to ensure that the opposite area can be fully and uniformly heated. The induction heating power supply is connected to a central control system (such as a PLC) through a control interface, allowing its output power to be adjusted dynamically in real time.

[0060] After completing the above hardware deployment, the stress control process of this invention can be executed. Before the welding operation begins, the process engineer, through the human-machine interface of the control system, presets a target temperature difference value based on the welding process qualification results. For example, the temperature can be set to 80°C depending on the material and thickness of the component.

[0061] The moment the welder starts the welding torch and the welding arc ignites, the control system detects the generation of welding current through an external current sensor and immediately initiates the feedforward preheating step. In this step, the control system uses a pre-established feedforward model (i.e., a feedforward parameter matching model) that correlates preset process parameters such as the welding current with the required initial heating power. Based on this model, the system provides an initial power to the induction heater on the opposite side for synchronous heating. The technical advantage of this approach is that it can quickly respond to the impact of welding heat input, proactively establishing a basic thermal balance at the beginning of the welding process. This avoids the instantaneous formation of a severe temperature gradient along the thickness of the component wall due to the rapid unidirectional conduction of welding heat, creating stable initial conditions for subsequent closed-loop control.

[0062] After the feedforward preheating is initiated, the system quickly switches to the closed-loop control phase of the welding process. During this phase, the control system performs the following operations at a high frequency (e.g., every 20ms): real-time acquisition of the welding zone temperature obtained from the first temperature sensing element. The temperature of the opposite heating zone obtained by the second temperature sensing element. According to the formula:

[0063]

[0064] Calculate the current actual temperature difference Then, the actual temperature difference ( ) and the preset temperature difference target value ( Compare and calculate the error. A PID closed-loop control algorithm calculates the adjustment amount of the induction heater power in real time based on the magnitude and polarity of the error (e), and outputs a control signal to dynamically adjust the heater's output power. Specifically, when the temperature in the welding zone rises too quickly, causing an actual temperature difference ( When the temperature difference increases and exceeds the target value, the PID controller increases the output power of the opposite induction heater to accelerate the heating rate on the opposite side and offset the increase in temperature difference; conversely, if the actual temperature difference is less than the target value, the heating power is reduced accordingly. Through this continuous and dynamic closed-loop negative feedback regulation, the temperature difference between the welding zone and the opposite zone is always precisely maintained near the preset target value throughout the welding process. The technical effect of this stage is that by actively managing and maintaining a small and stable temperature gradient through the wall thickness direction, it helps to suppress the accumulation of residual stress inside the component caused by uneven thermal expansion and contraction, thereby significantly reducing the residual stress and angular deformation of the component after welding.

[0065] After the welding process ends, i.e., once the welding arc is detected to have extinguished, the control flow enters the controlled cooling phase. At this point, the induction heater on the opposite side does not immediately stop operating. The control system continues to run, but its control objective shifts to matching the cooling rate of the opposite heating zone with the natural cooling rate of the welding zone. The system continuously monitors temperature changes on both sides and adjusts the power of the induction heaters accordingly, providing supplemental heating to the opposite side in a gradually decreasing manner, thereby achieving synchronous cooling of the two zones. The technical advantage of this approach is that it avoids the generation of enormous tensile stress between the welded zone and the still-high-temperature base material due to rapid contraction during the cooling phase. By guiding the entire component cross-section to cool in a holistic, slow, and uniform manner, more time is provided for stress relaxation in the material at high temperatures, further releasing and homogenizing residual stresses that may have formed during welding, ultimately ensuring that the component has a lower residual stress level and less welding deformation after cooling to room temperature.

[0066] Example 2

[0067] This embodiment further defines and optimizes the specific implementation of the closed-loop control algorithm described in Embodiment 1. In the closed-loop control stage of the welding process described in Embodiment 1, to further improve control accuracy, response speed, and system stability, the PID closed-loop control algorithm preferably adopts an incremental PID algorithm, as follows:

[0068] Unlike conventional positional PID algorithms that output absolute power values, incremental PID algorithms output the heater power adjustment (ΔP), which is the power value that should be increased or decreased in the current control cycle relative to the previous cycle. This implementation method avoids drastic changes in control values ​​when the control system switches or experiences significant disturbances, thus ensuring the smoothness and safety of the control process.

[0069] In this embodiment, the specific calculation formula of the incremental PID control algorithm is as follows:

[0070]

[0071] in:

[0072] ΔP is the power adjustment calculated by the control system within one control cycle;

[0073] This is the proportionality coefficient. The integral coefficient is... These are the differential coefficients. The values ​​of these three coefficients are determined in advance through experimental tuning or simulation analysis based on the specific component materials, dimensions, and welding processes to achieve the optimal control effect.

[0074] The error at the current moment, i.e. ,in, The preset temperature difference target value, This is the actual temperature difference collected and calculated at the current moment;

[0075] This is the error from the previous moment, i.e., the error from the previous control moment;

[0076] This represents the error between the first two time points, i.e., the error between the first two control time points.

[0077] Within each discrete control cycle of the control system, the specific execution logic is as follows: First, the system calculates the error at the current moment. ; then, and the error stored from the previous moment Errors from the first two time points Substituting into the above formula, the power adjustment amount ΔP is calculated. Finally, this adjustment amount is added to the power output value P(t-1) of the previous moment to obtain the target power output value P(t) = P(t-1) + ΔP at the current moment. This P(t) value is then limited (ensuring it is between 0% and 100%) and converted into a control signal, which is sent to the induction heating power supply.

[0078] The technical advantage of this algorithm lies in its proportional term. It responds to changes in the error, rather than the absolute value of the error, which makes the system respond more quickly to disturbances and less prone to large oscillations. Integral term Used to eliminate steady-state errors in the system, ensuring that the actual temperature difference converges precisely to the target value during long-term stable welding. Differential term. By predicting the trend of error changes, it plays a role in early suppression and damping, effectively preventing overshoot caused by drastic temperature changes and significantly enhancing the reliability of the control system. In summary, by employing this specific incremental PID algorithm, smoother, more stable, and more precise dynamic control of the welding temperature field can be achieved, thereby more effectively suppressing the generation of welding residual stress.

[0079] In this embodiment, regarding the scaling factor ( ), integral coefficient ( ) and differential coefficients ( The method for determining () is explained below with supplementary information:

[0080] This supplementary explanation uses the Ziegler-Nichols critical oscillation method, which is widely used in the field of industrial control, as an example to elaborate on the specific steps for obtaining a set of excellent PID parameters suitable for the control system described in this invention:

[0081] Step 1: System Test Preparation Before parameter tuning, the control system needs to be placed in pure proportional control mode. Specifically, in the control algorithm, the integral coefficient ( ) and differential coefficients ( The initial values ​​of all parameters are set to 0. At this time, the output of the controller is determined only by the proportional term.

[0082] Step 2: Determine the critical proportional gain (Ku)

[0083] Initiate the welding heat input (simulation can be performed using actual welding or a heat source with equivalent power), and simultaneously activate the opposite side heating control system. From a small scaling factor ( ) value (e.g. Start with 0.1 and gradually increase. The set value. During this process, the actual temperature difference is closely monitored ( The response curve of ). With As the temperature increases, the system becomes more sensitive to temperature fluctuations, until at a certain point... When the value is reached, the actual temperature difference ( The curve of ) exhibits a continuous, constant-amplitude oscillation. This oscillation neither decays nor diverges over time. The proportional gain value that just induces this critical stable oscillation is recorded; this value is the critical proportional gain, denoted as Ku.

[0084] Step 3: Determine the critical oscillation period (Tu)

[0085] When the system oscillates stably at the critical proportional gain (Ku), the time between two consecutive peaks (or troughs) on the oscillation curve is measured. This time is the critical oscillation period, denoted as Tu.

[0086] Step 4: Calculate the initial values ​​of the PID parameters according to the tuning rules.

[0087] Based on the measured critical proportional gain (Ku) and critical oscillation period (Tu), the recommended initial values ​​of the PID parameters are calculated according to the Ziegler-Nichols tuning rule table. For the standard PID controller used in this embodiment, the classic calculation formula is:

[0088]

[0089] Ti = 0.5 × Tu

[0090] Td = 0.125 × Tu

[0091] Where Ti is the integral time and Td is the derivative time. Since the control formula in Embodiment 2 of this invention directly uses the integral coefficient (Ki) and derivative coefficient (Kd), it needs to be converted according to the discrete sampling period (Ts) of the control system (e.g., 20ms or 0.02s as described in Embodiment 1).

[0092]

[0093]

[0094] The above calculations , , Used as the initial parameter settings for the PID controller.

[0095] Step 5: Online Fine-tuning and Optimization

[0096] The above steps yield a set of reliable initial parameters that ensure the system's basic stability and responsiveness. In actual operating conditions, to achieve better control performance, online fine-tuning can be performed based on these initial values. For example, perform a complete welding test on the test piece and observe the temperature difference control curve: if it is found that the system eliminates steady-state error slowly, the initial parameters can be appropriately increased. Value; if the system response has a large overshoot, it can be appropriately reduced. Value and (or) increase The value can be increased to enhance damping; if the overall system response is sluggish, it can be appropriately increased. Value. Through several such iterative optimizations, a set of optimal PID parameter combinations for a specific welding object can be quickly obtained.

[0097] The above is a clear, logically rigorous, and engineering-effective method for PID parameter tuning.

[0098] Example 3

[0099] This embodiment is a further optimization scheme proposed based on Embodiment 1 for the preparation stage before the formal start of welding. It adds a heat capacity pre-charging step, which aims to create a more stable and favorable initial thermal field environment for subsequent precise synchronous control.

[0100] Based on the hardware deployment described in Embodiment 1, this embodiment optimizes the control process. Before the welder is ready to begin welding, the system does not immediately enter a waiting state for arc ignition; instead, a pre-charging step is executed. After the operator initiates this process through the human-machine interface, the control system first commands the opposite induction heater to start operating at a preset, low, constant power (e.g., 20% of its rated power). The purpose of heating at this stage is not to achieve dynamic synchronization with the welding area, but to preheat the entire opposite wall panel of the box-shaped column over a wide area in a gentle and uniform manner.

[0101] To ensure the uniformity of this pre-charging process and make it quantifiable and controllable, this embodiment expands the deployment of the sensing unit. Specifically, in addition to the second temperature sensing element located at the geometric center of the opposite heating area, multiple auxiliary temperature sensing elements are added at key locations around it. For example, an auxiliary K-type armored thermocouple of the same model as the second temperature sensing element is added approximately 300 mm above, below, left, and right of the central temperature sensing point. During the pre-charging phase, the control system simultaneously collects temperature data from these five temperature sensing points (one central point and four auxiliary points).

[0102] The termination condition of the pre-charging process is determined by these temperature measurement data. The control system calculates the arithmetic mean temperature of these five measurement points in real time and compares it with a preset pre-charging target temperature (e.g., 120°C). Simultaneously, the system monitors the maximum temperature difference between these five measurement points to assess the uniformity of heating. When the calculated average temperature reaches 120°C and the temperature difference between points is within the allowable range, the control system determines that the heat capacity pre-charging stage is complete, immediately stops heating the opposite heater, and puts the system into a ready-to-weld state, at which point welding operations can begin.

[0103] The technical effect of this heat capacity pre-charging step can be understood as follows: a large, thick-walled component itself is a large heat sink. If welding is performed directly in a cold state, the heat from the welding zone will be rapidly conducted to the low-temperature base material, forming an extremely drastic temperature gradient at the moment the welding arc begins. This drastic gradient is not only the main source of residual stress but also adversely affects the initial response of the subsequent closed-loop control system, easily leading to severe overshoot or oscillation of the control quantity. By implementing this heat capacity pre-charging step, it is equivalent to transforming the large "heat sink" of the component into a "heat capacity buffer" with a certain heat reserve in advance. When welding officially begins, the initial temperature difference between the welding zone and the base material has been significantly reduced, thus creating a smooth and stable initial operating condition for the temperature difference-based closed-loop control described in Example 1. This allows the control system to intervene more smoothly and achieve faster and more accurate tracking control of the target temperature difference. At the same time, by measuring temperature at multiple points and using the average temperature as the criterion, the uniformity of the entire reference temperature field is ensured, avoiding the introduction of new stress sources due to uneven preheating, and providing a more reliable guarantee for suppressing welding residual stress from the source.

[0104] Example 4

[0105] This embodiment is a further optimization of the method described in Embodiment 1, specifically addressing the controlled cooling stage after welding. During the cooling process of the component from high temperature to low temperature, a high-temperature thermal pulse impact method is coupled to achieve enhanced release of residual stress. The specific implementation process is as follows:

[0106] Upon completion of the welding process, i.e., after the welding arc is detected to have extinguished, the control system enters a controlled cooling mode. In this mode, the control system continuously monitors the temperature of the welding zone. ), and monitor it in real time. When the control system detects When the temperature drops and enters a preset high-temperature range, such as 800°C to 600°C, the system will automatically trigger and execute a preset high-temperature thermal pulse impact subroutine. This temperature range is selected based on the material properties; within this range, Q355B steel still retains good plasticity, which is beneficial for stress relaxation.

[0107] The specific execution logic of this subroutine is as follows: The control system sends a series of discrete, high-energy heating commands to the induction heater on the opposite side. For example, the control system can command the heater to output a thermal pulse with a width of 200 milliseconds at 90% of its rated power. After a single pulse ends, the system enters a preset waiting interval, such as 8 seconds. The purpose of this interval is to allow the thermal effect generated by the pulse to fully act within the material and gradually subside, preparing for the next impact. After this interval ends, the system sends the same thermal pulse again. This "pulse-interval" cycle is repeated a preset number of times, such as 10 times. After completing all preset number of thermal pulse impacts, the subroutine ends, and the control system immediately returns to the normal controlled cooling mode until the component temperature drops below the safety threshold.

[0108] The technical effect of this high-temperature thermal pulse impact step lies in the fact that, within the high-temperature range of 600°C to 800°C, metallic materials still possess high plasticity and low creep resistance, representing a golden window for stress release. By applying a transient, high-power thermal pulse to the opposite side, a dramatic thermal expansion can be induced in a localized area, its effect similar to a non-contact "thermal shock." This periodic impact effectively promotes the movement and rearrangement of dislocations within the material's internal microstructure, accelerating the stress relaxation process. This is equivalent to interrupting and homogenizing stress concentration at the microscopic level during the critical period when the macroscopic cooling stress field is forming, thereby achieving a further and deeper reduction in residual stress on top of conventional slow cooling, effectively improving the fatigue performance and dimensional stability of the component.

[0109] Example 5

[0110] This embodiment is a further optimization of the closed-loop control strategy in the welding process based on the method described in Embodiment 1. The core of this optimization lies in transforming the control target from a static temperature difference value to a dynamic temperature change rate, thereby achieving more sensitive and fundamental control over the dynamic balance of the temperature field. The specific implementation process is as follows:

[0111] After the welding process begins, the control system no longer uses a fixed temperature difference target value. Instead of using it as a control benchmark, the control target was adjusted to be set to reduce the rate of temperature change in the opposite heating zone ( ). Real-time and closely track the temperature change rate of the welding zone. Therefore, the error input (e) of the PID control algorithm is redefined as the difference between the rates of temperature change in the two regions, i.e.:

[0112]

[0113] The task of the control system is to adjust the heater power in real time so that the error (e) dynamically approaches zero.

[0114] Furthermore, in this embodiment, in order to reliably implement this control logic in an industrial controller (such as a PLC), the calculation of the temperature change rate needs to be completed using precise numerical methods within discrete sampling periods. The control system continuously executes a control loop with a fixed sampling period Δt (e.g., 20ms as described in Embodiment 2). Within each period, the system first acquires and filters the temperature of the welding zone (…). ) and the temperature of the opposite heating zone ( The instantaneous value of ) is then calculated. Subsequently, the four-point center difference method is used to calculate the respective temperature change rate, which, compared to the simple two-point difference method, has better noise suppression and smaller phase delay. The specific calculation formula is as follows:

[0115]

[0116] in, The temperature value obtained in the current sampling period. These are the temperature values ​​obtained in the first two sampling periods. The control system uses this formula to calculate... and This leads to the rate error e(t), which is then used as the input to the PID closed-loop control algorithm to calculate the power adjustment of the opposite heater.

[0117] This control strategy based on synchronized temperature change rate has the advantage of fundamentally changing the control response mode. Traditional temperature difference control compensates for an already formed result (temperature difference) with a lag, while the method in this embodiment intervenes in advance against the trend that leads to the temperature difference (i.e., the difference in heating rate). When the heating rate of the welding zone changes due to factors such as arc fluctuations and welder hand movements during the welding process, this control scheme can detect this dynamic trend in a timely manner and quickly adjust the heating power on the opposite side to match the change, thereby eliminating it in the initial state before a significant temperature difference accumulates. This direct control of the dynamic process makes the system more reliable in the face of various disturbances during the welding process, enabling more accurate dynamic balance of the temperature field, and thus more effectively suppressing stress throughout the entire process of stress formation.

[0118] Example 6

[0119] This embodiment is a further, more refined control strategy proposed based on the method described in Embodiment 5, specifically for the controlled cooling stage after welding. The core of this strategy lies in shifting the control objective of the cooling stage from simple rate synchronization to an active, asymmetric rate guidance, in order to achieve a higher level of counterbalancing and equilibrium during stress formation. The specific process is as follows:

[0120] After the welding process ends, i.e., after the control system determines that the arc has extinguished by monitoring the welding circuit current, the system control logic automatically and seamlessly switches from the welding synchronization control mode described in Example 5 to the asymmetric rate cooling mode of this example. The control objective in this mode is to maintain a pre-set proportional relationship between the cooling rate of the opposite heating zone and the actual cooling rate of the welding zone, specifically manifested as the following dynamic objective:

[0121]

[0122] Where k is a pre-set proportionality coefficient less than 1, so that the cooling rate of the opposite heating zone is always slower than that of the welding zone, so as to form a beneficial stress balance during the cooling process.

[0123] To achieve this dynamic objective in a closed-loop control system, the control system transforms the relationship to be applicable to a conventional PID controller. Within each control cycle, the system first calculates the current actual cooling rate of the welding zone using the filtering and differential methods described in Example 5. The system then multiplies this value by a pre-defined scaling factor k of 0.95, based on the material and process, to obtain a dynamically changing target cooling rate setpoint, namely: This setpoint represents the ideal cooling rate that the opposite heating zone should achieve at the current moment. Simultaneously, the system also calculates the actual cooling rate feedback value of the opposite heating zone using the same method. .

[0124] Furthermore, the error input of the PID control algorithm , Defined as the difference between the target setpoint and the actual feedback value:

[0125]

[0126] The controller uses this error As input, the adjustment amount of the opposite heater power is calculated using an incremental PID algorithm. If the actual cooling rate on the opposite side is too fast (i.e., (The absolute value of the error is greater than the absolute value of the target speed). When the value is positive, the PID controller will increase the heating power accordingly to "slow down" the cooling; conversely, if the cooling is too slow, it will decrease the heating power. Through this high-frequency closed-loop regulation, the system can force the actual cooling rate of the opposite side. Accurately and dynamically track its target rate setpoint ( This control process will continue until the temperature in the welding zone drops below the safe threshold of 150°C, at which point the main shrinkage process has been completed.

[0127] The technical advantage of this asymmetric rate cooling method lies in its shift from a passive "following" approach to an active "guiding" effect in controlling the cooling phase. During the critical stage where harmful tensile stress arises in the weld zone due to cooling contraction, this method precisely controls the opposite region to maintain a relatively higher temperature and a slower cooling rate, artificially and controllably sustaining a gentle thermal expansion in that area. This thermal expansion creates a beneficial compressive stress field within the component, opposite to the direction of the welding contraction tensile stress, essentially acting as a dynamic "thermal support." This thermal support counteracts and balances the welding tensile stress in real time as it forms, achieving more effective residual stress control. Ultimately, this results in lower residual stress levels and greater dimensional stability within the component after complete cooling.

Claims

1. A method for controlling stress by simultaneous heating on opposite sides during welding of thick-walled components, characterized in that, Includes the following steps: A first temperature measuring element is set near the area to be welded in the thick-walled component to obtain the temperature of the welding area; and a heater with controllable power and a second temperature measuring element are set in the opposite side area to the area to be welded, and the temperature of the opposite side heating area is obtained through the second temperature measuring element. When the welding arc is initiated, an initial power is applied to the heater for feedforward preheating; During the welding process, based on a preset temperature difference target value, the output power of the heater is adjusted in real time and dynamically according to the actual difference between the temperature of the welding zone and the temperature of the opposite heating zone through a closed-loop control algorithm, so that the actual difference approaches the temperature difference target value. The control objective of the closed-loop control algorithm is to synchronize the temperature change rate of the welding zone with the temperature change rate of the opposite heating zone. The error calculation formula is as follows: ； Where e is the error in the rate of temperature change. The temperature change rate in the welding zone, The control objective is to make the error dynamically approach zero, given the rate of temperature change in the opposite heating zone. After the welding process is completed, the power of the heater is continued to be controlled so that the cooling rate of the opposite heating zone matches the cooling rate of the welding zone, so as to achieve synchronous cooling.

2. The method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components according to claim 1, characterized in that, The closed-loop control algorithm is a PID control algorithm, and the formula for calculating the actual difference is: ； in, This represents the actual temperature difference. Temperature of the welding zone This refers to the temperature of the heating zone on the opposite side. The PID control algorithm is based on error Calculate the heater power adjustment, where This is the preset target temperature difference value.

3. The method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components according to claim 1, characterized in that, Both the first and second temperature sensing elements are K-type armored thermocouples. The first temperature sensing element is located 10-15 mm outside the area to be welded, and the second temperature sensing element is located close to the geometric center of the opposite heating area.

4. The method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components according to claim 1, characterized in that, The heater is an induction heater, and the induction heater includes a digital medium-frequency induction heating power supply and a planar rectangular induction coil. The size of the induction coil is larger than the projected area of ​​the area to be welded.

5. The method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components according to claim 1, characterized in that, Before the welding arc is initiated, a heat capacity pre-charging step is also included: the heater is activated to uniformly heat the entire opposite side wall of the thick-walled component with low power over a large area, and the reference temperature of the entire opposite side wall is uniformly increased by a preset amount to reduce the initial temperature gradient between the welding zone and the component base material at the moment of welding arc initiation.

6. The method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components according to claim 5, characterized in that, Multiple auxiliary temperature measuring elements are set around the opposite heating area. The uniformity of pre-charging heating is determined by monitoring the temperature consistency between the auxiliary temperature measuring elements and the second temperature measuring element. Pre-charging heating is stopped when the average temperature of each measuring point reaches the preset pre-charging target temperature.

7. The method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components according to claim 1, characterized in that, The rate of temperature change was calculated using the four-point central difference method: ； in, The temperature at the current moment. The temperature for the first two sampling periods. The sampling period.

8. The method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components according to claim 1, characterized in that, The cooling stage after the welding process is completed also includes a high-temperature thermal pulse impact step: when the temperature of the welding area is within the preset high-temperature range, the heater is controlled to output multiple high-power, short-duration thermal pulses, and the thermal pulses generate instantaneous thermal expansion and contraction in the opposite area to promote stress relaxation inside the material. The preset high temperature range is 600-800°C, the thermal pulse is a high-power pulse with a pulse width of milliseconds, and there is a preset time interval between multiple thermal pulses.

9. The method for controlling stress by simultaneous heating on the opposite side during welding of thick-walled components according to claim 1, characterized in that, During the cooling phase after welding, asymmetric rate control is used, setting the target cooling rate of the opposite heating zone to a specific proportion of the actual cooling rate of the welding zone: ； Where k is a pre-set proportionality coefficient less than 1, so that the cooling rate of the opposite heating zone is always slower than that of the welding zone, so as to form a beneficial stress balance during the cooling process.

Images