Method for controlling intermetallic compounds at interface of copper-aluminum clad strip

Abstract

The present application relates to the technical field of metal composite material processing control, and discloses a copper-aluminum composite strip interface intermetallic compound control method, which comprises the following steps: obtaining a production line continuous running line speed and calculating an effective holding time; synchronously injecting high-frequency and low-frequency excitation signals into the copper-aluminum composite strip to obtain high-frequency and low-frequency equivalent conductivities; performing differential operation on the two kinds of equivalent conductivities to extract interface micro-strain characteristic values, substituting the characteristic values into an atomic diffusion mapping model to calculate a strain-induced diffusion enhancement factor; substituting a preset target compound layer thickness, the effective holding time and the enhancement factor into an Arrhenius inverse solution kinetics model to obtain a target heating power instruction, calculating a heating frequency and a cooling gas flow instruction according to the characteristic values, and combining to generate a three-dimensional control instruction set; calculating a space delay time according to the line speed, performing first-in first-out data queue shift delay delay, and synchronously driving the heating and cooling modules. The present application realizes dynamic control of the interface compound layer thickness.

Description

Technical Field

[0001] This invention relates to the field of metal composite material processing control technology, specifically a method for controlling intermetallic compounds at the interface of copper-aluminum composite strips. Background Technology

[0002] Copper-aluminum composite strips are increasingly widely used in fields such as power transmission and new energy equipment. In the continuous manufacturing process of copper-aluminum composite strips, the metallurgical bonding quality of the copper-aluminum interface is a key factor that determines the overall performance of the product. The thickness of the intermetallic compound layer at the interface directly affects the bonding strength and conductivity of the strip. Therefore, in the continuous heat treatment process after composite deformation, it is necessary to strictly control the heat input in order to regulate the growth state of the intermetallic compound at the interface.

[0003] Existing continuous heat treatment processes mainly rely on setting fixed heating parameters or simple closed-loop feedback control based on the surface temperature of the strip. In actual production, during the early stages of rolling and other composite deformation, copper and aluminum strips will experience varying degrees of microscopic lattice distortion and strain at their interface. This internal microscopic strain will reduce the diffusion barrier of atoms, thereby accelerating the atomic diffusion rate during the heat treatment process. Existing technologies treat the strip as an ideal model with a uniform internal state and no residual stress. The conventional thermodynamic control logic used in these technologies fails to take into account the promoting effect of local lattice distortion on atomic diffusion.

[0004] At the same time, there is a lack of effective online detection methods for the state of the interface layer inside the composite strip. Due to ignoring the objective differences in the microscopic internal state at different locations along the longitudinal direction of the strip, the use of a uniform heat input will lead to excessive growth of intermetallic compounds in high strain areas, resulting in local embrittlement. In low strain areas, insufficient growth will lead to low bonding strength, ultimately resulting in uneven distribution of the intermetallic compound layer thickness of the entire composite strip.

[0005] Meanwhile, on a continuous production line, there is a fixed physical distance between the front-end detection position of the strip condition and the rear-end heating and cooling execution position. When the existing control method dynamically adjusts the parameters, it lacks a spatial delay compensation mechanism that is closely coupled with the real-time linear speed. Under continuous operation, since the time difference of the strip transmission between workstations is not calculated, the control commands generated by the system based on the front-end condition often cannot be accurately applied to the specific strip section that generates the characteristic data. This spatial control misalignment further reduces the consistency of the interface quality of long strip products. Summary of the Invention

[0006] To address the shortcomings of existing technologies, this invention provides a method for controlling intermetallic compounds at the interface of copper-aluminum composite strips. This method solves the problem of uneven distribution of intermetallic compound layer thickness in the continuous production of copper-aluminum composite strips, which is caused by the failure to consider differences in microscopic lattice distortion and spatial transport delay within the strip.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solution: This invention provides a method for controlling intermetallic compounds at the interface of copper-aluminum composite strips, the method comprising the following steps: Obtain the continuous operating line speed of the production line and calculate the effective heat preservation time; High-frequency and low-frequency excitation signals are synchronously injected into a continuously moving copper-aluminum composite strip to obtain the high-frequency equivalent conductivity and low-frequency equivalent conductivity at a specific longitudinal position. The interface micro-strain characteristic values ​​are extracted by performing a difference operation on the high-frequency equivalent conductivity and the low-frequency equivalent conductivity, and then substituted into the atomic diffusion mapping model to calculate the strain-induced diffusion enhancement factor. The target metallurgical interface intermetallic compound layer thickness, effective heat preservation time and strain-induced diffusion enhancement factor are substituted into the Arrhenius inverse dynamics model to obtain the target heating power command. The target heating frequency command and cooling gas flow command are calculated based on the interface micro-strain characteristic value and combined to generate a three-dimensional control command set. The spatial delay time is calculated based on the continuous operating speed of the production line, and the first-in-first-out data queue shift delay is executed on the three-dimensional control instruction set to drive the induction heating module and the gas cooling module to be synchronously controlled.

[0008] The control method of this invention utilizes the frequency correlation of the electromagnetic skin effect and simultaneously obtains the conductivity distribution of the strip surface and interface layer through dual-frequency eddy current detection. This method uses the surface state characterized by high-frequency signals as a reference benchmark, eliminates surface deformation interference through differential operation, and extracts micro-strain characteristics that reflect the internal lattice distortion of the interface. On this basis, a defect diffusion dynamics mechanism is introduced to convert micro-strain into a strain-induced diffusion enhancement factor, thereby correcting the Arrhenius dynamics model.

[0009] By solving the model in reverse, the thermal input conditions required to compensate for local strain differences are calculated and transformed into control commands for heating power, frequency, and cooling gas flow rate. At the same time, by combining speed measurement and first-in-first-out queue data management to compensate for spatial delay, the commands are ensured to be triggered synchronously when the strip travels to the execution zone, thereby achieving closed-loop control of the thickness of the intermetallic compound layer at the interface.

[0010] Furthermore, the step of obtaining the continuous operating line speed of the production line and calculating the effective heat preservation time specifically includes: The continuous operating line speed of the copper-aluminum composite strip production line and the effective heating section length of the induction heating module are obtained. Divide the effective heating section length by the continuous operating speed of the production line to determine the effective heat preservation time of a specific strip section within the heating zone; A preset lower limit threshold for linear speed is set. When the continuous operating linear speed of the actual production line is lower than the preset lower limit threshold, the previously calculated value is automatically locked to maintain the integrity of the control logic.

[0011] Furthermore, the synchronous injection of high-frequency and low-frequency excitation signals into the continuously moving copper-aluminum composite strip specifically includes: Set the alternating frequency of the high-frequency electromagnetic excitation signal so that the corresponding electromagnetic skin depth is constrained within the outer surface layer of the main heating surface of the copper-aluminum composite strip. Set the alternating frequency of the low-frequency electromagnetic excitation signal so that its corresponding electromagnetic skin depth penetrates the outer surface layer of the main heating surface and extends to the metallurgical bonding interface layer inside the copper and aluminum of the copper-aluminum composite strip. The induced electromotive force signal fed back from the copper-aluminum composite strip is collected and subjected to high-speed analog-to-digital conversion to demodulate the high-frequency equivalent conductivity and low-frequency equivalent conductivity corresponding to the specific longitudinal position.

[0012] Furthermore, the step of extracting the interface micro-strain characteristic values ​​by performing a difference operation on the high-frequency equivalent conductivity and the low-frequency equivalent conductivity specifically includes: Obtain the skin depth weighting coefficients obtained from the calibration of a pre-acquired reference copper-aluminum composite strip without lattice distortion; The low-frequency equivalent conductivity and the high-frequency equivalent conductivity are subjected to a difference matrix operation. The high-frequency equivalent conductivity is used as a background reference and combined with the skin depth weighting coefficient to quantitatively remove the interference contribution of surface macroscopic deformation from the low-frequency equivalent conductivity, and the interface micro-strain characteristic value reflecting the micro-lattice distortion of the copper-aluminum interface is extracted.

[0013] Furthermore, after extracting the interface micro-strain characteristic values, a boundary limiting filtering step is also included: Determine whether the calculated interface micro-strain characteristic value is less than zero. If it is determined to be less than zero, force the interface micro-strain characteristic value at the current coordinate point to be reset to zero.

[0014] Furthermore, the atomic diffusion mapping model employs an empirical polynomial equation calibrated based on crystal defect diffusion kinetics to solve for the strain-induced diffusion enhancement factor, specifically including: The empirical polynomial equation includes a quadratic term, a linear term, and a constant term, wherein the constant term corresponds to the reference property correction value under the condition of no initial strain. The strain-induced diffusion enhancement factor is obtained by substituting the interface micro-strain characteristic values ​​into the empirical polynomial equation and mapping the calculation. When outputting the calculation results, a lower limit value for the strain-induced diffusion enhancement factor is forcibly limited to avoid unsolvable situations in subsequent thermodynamic inverse solution calculations.

[0015] Furthermore, the step of substituting the preset target metallurgical interface intermetallic compound layer thickness, effective holding time, and strain-induced diffusion enhancement factor into the Arrhenius inverse kinematics model to obtain the target heating power command specifically includes: The embedded Arrhenius inverse dynamics model is invoked to solve inversely the optimal target temperature required to suppress or compensate for local strain at a specific longitudinal position. The initial temperature of the copper-aluminum composite strip before entering the multi-physics field interaction zone is obtained by real-time acquisition through an infrared temperature sensor. By combining the energy balance equation, the difference between the optimal target temperature and the initial temperature, and taking into account the average density and specific heat capacity of the copper-aluminum composite strip and the electrothermal conversion efficiency of the induction heating module, is converted into the target heating power command.

[0016] Furthermore, the calculation of the target heating frequency command and cooling gas flow rate command based on the interface micro-strain characteristic values ​​employs a segmented calculation strategy based on the interface strain tolerance threshold, specifically including: When the characteristic value of the interface micro-strain is less than or equal to the preset interface strain tolerance threshold, the basic heating frequency and minimum cooling gas flow rate are output. When the micro-strain characteristic value of the interface is greater than the preset interface strain tolerance threshold, the target heating frequency command is solved piecewise using the preset frequency compensation gain coefficient, the cooling gas flow command is solved piecewise using the preset cooling gas flow compensation gain coefficient, and the solution result is limited to the safe working range of the actuator by the maximum value limiting term embedded in the function.

[0017] Furthermore, the step of performing a first-in-first-out data queue shift delay on the three-dimensional control instruction set specifically includes: The physical distance between the dual-frequency eddy current sensor module and the multi-physics field action zone is read and divided by the continuous operating line speed of the production line to calculate the spatial delay time under ideal conditions. A first-in-first-out data queue is established, the received three-dimensional control command set is pushed into the end, and when the displacement pulse fed back by the speed encoder is received, the stepping pointer inside the data queue is driven to move forward synchronously. When it is determined that the continuous operating speed of the production line is lower than the set safety lower limit or the line stops, the data queue pause pointer shifts to enter the data hold state.

[0018] Furthermore, before the synchronous control of the driving induction heating module and the gas cooling module, a transient driving and feedforward execution compensation step is also included: Obtain the pre-calibrated hardware response compensation time, which is determined by taking the maximum value of the electrical response time of the inverter power supply and the mechanical action time of the servo proportional valve. The hardware response compensation time is converted into lead space steps by combining the real-time continuous operating line speed of the production line. When the copper-aluminum composite strip reaches the multiphysics field action zone by the corresponding advance step, the three-dimensional control instruction set is triggered to pop out from the end of the first-in-first-out data queue in advance, and drive control instructions are issued in parallel.

[0019] This invention provides a method for controlling intermetallic compounds at the interface of copper-aluminum composite strips. It has the following beneficial effects: 1. This invention injects high-frequency and low-frequency excitation signals simultaneously into a moving copper-aluminum composite strip and performs differential calculations on the two equivalent conductivity values ​​obtained. Based on the surface state characterized by the high-frequency signal, the interference of macroscopic deformation of the surface can be quantitatively eliminated from the low-frequency signal, and the microscopic strain characteristic value reflecting the internal lattice distortion of the bonding interface can be extracted. This feature extraction method overcomes the limitation of conventional single surface detection methods that cannot perceive the internal state of the composite strip, and provides accurate basic data for subsequent heating control.

[0020] 2. This invention substitutes the micro-strain eigenvalues ​​into the atomic diffusion mapping model to convert them into strain-induced diffusion enhancement factors, and uses this to modify the Arrhenius inverse dynamics model to solve the target heating power, frequency, and cooling gas flow rate in reverse. This control logic takes into account the promoting effect of local lattice distortion of the strip on atomic diffusion, and can dynamically adjust the required heat input and cooling conditions according to the actual micro-defect state at different locations in the longitudinal direction of the strip, thereby improving the problem of uneven thickness distribution of the intermetallic compound layer in continuous production.

[0021] 3. This invention adopts a first-in-first-out data queue shifting and distribution mechanism, combines the production line speed to calculate the spatial delay time, and introduces an advance compensation step based on hardware response time. This design scheme compensates for the physical spatial distance between the detection station and the multi-physics field action area, ensuring that the control command calculated based on the specific strip position can be triggered synchronously when the strip in that section actually reaches the heating and cooling execution area, avoiding the misalignment between the command distribution and the actual action position under dynamic operating conditions. Attached Figure Description

[0022] Figure 1 This is a schematic diagram of the system architecture of the present invention; Figure 2 This is a schematic diagram of the method flow of the present invention; Figure 3 This is a schematic diagram of the process logic for strain space decoupling and diffusion dynamics mapping of the present invention; Figure 4 This is a schematic diagram of the process logic for inversely solving the growth kinetics of intermetallic compounds and generating three-dimensional control commands according to the present invention. Figure 5This is a longitudinal thickness distribution diagram of the intermetallic compound of the present invention; Figure 6 This is a longitudinal distribution diagram of the T-type peel strength of the present invention.

[0023] Among them, 101 is the dual-frequency eddy current sensor module; 102 is the data acquisition module; 103 is the feature decoupling module; 104 is the dynamic mapping module; 105 is the command generation module; 106 is the delay compensation module; 107 is the induction heating module; and 108 is the gas cooling module. Detailed Implementation

[0024] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] See attached document Figure 1 This invention provides a control system for intermetallic compounds at the interface of copper-aluminum composite strips, the system comprising: The system includes a dual-frequency eddy current sensor module 101, a data acquisition module 102, a feature decoupling module 103, a dynamic mapping module 104, an instruction generation module 105, a delay compensation module 106, an induction heating module 107, a gas cooling module 108, and an infrared temperature sensor.

[0026] The system is divided into a feature extraction zone, a spatiotemporal delay buffer zone, and a multi-physics field action zone along the production line direction of the copper-aluminum composite strip. The feature extraction zone is equipped with a dual-frequency eddy current sensor module 101. The multi-physics field action zone is equipped with an induction heating module 107, a gas cooling module 108, and an infrared temperature sensor located at the entrance of the induction heating zone. The infrared temperature sensor is electrically connected to the instruction generation module 105 and is used to collect the initial temperature of the strip before it enters the multi-physics field action zone in real time. The feature extraction zone and the multi-physics field action zone maintain a physical distance in the longitudinal direction of the production line.

[0027] The dual-frequency eddy current sensor module 101 is configured above the copper-aluminum composite strip to be processed. It is used to inject electromagnetic excitation signals of different frequencies into the strip. The dual-frequency eddy current sensor module 101 acquires the conductivity characteristic signals reflecting different thickness levels of the strip in a non-contact manner and transmits the conductivity characteristic signals to the data acquisition module 102.

[0028] The data acquisition module 102 is electrically connected to the dual-frequency eddy current sensor module 101, and is used to record the conductivity response value of the strip at the longitudinal position x in real time, and convert the conductivity response value into a digital signal and input it to the feature decoupling module 103.

[0029] The feature decoupling module 103 stores preset electromagnetic skin depth weighting coefficients, which are used to perform differential operations on high and low frequency conductivity signals. The feature decoupling module 103 extracts the interface micro-strain characteristic values ​​that reflect the micro-lattice distortion of the copper-aluminum bonding interface by stripping the surface macro-deformation signal.

[0030] The kinetic mapping module 104 is connected to the feature decoupling module 103 and internally stores a mapping model calibrated based on atomic diffusion theory. The kinetic mapping module 104 calculates the strain-induced diffusion enhancement factor used to correct the atomic diffusion rate of the interface based on the micro-strain characteristic value of the interface.

[0031] The instruction generation module 105 includes an inverse algorithm for the growth kinetics of intermetallic compounds. Based on the set target layer thickness, effective heat preservation time, and input strain-induced diffusion enhancement factor, the instruction generation module 105 simultaneously calculates the induction heating power, induction heating frequency, and cooling gas flow rate instructions for the current strip section, forming a three-dimensional control instruction set.

[0032] The delay compensation module 106 is equipped with a first-in-first-out data queue. The delay compensation module 106 calculates the spatial delay time based on the real-time linear velocity and physical distance of the strip. The three-dimensional control instruction set is shifted and stored in the data queue to ensure that the time point of instruction issuance is aligned with the time point of the strip segment entering the multi-physics field action zone.

[0033] The output terminals of the induction heating module 107 and the delay compensation module 106 are connected. The induction coil is wrapped around the outer periphery of the strip. The induction heating module 107 receives power commands and frequency commands, and changes the electromagnetic skin depth by adjusting the inverter frequency to achieve directional heat generation on the surface of the strip.

[0034] The gas cooling module 108 is located in the multiphysics field action zone and on the back side of the strip. The gas cooling module 108 includes a servo proportional valve controlled by the flow command. The gas cooling module 108 works in conjunction with the induction heating module 107 to apply forced convection cooling to the back side of the strip and work together with the surface induction heating to establish a quasi-steady-state temperature gradient in the thickness direction of the strip.

[0035] See attached document Figure 2 This invention provides a method for controlling intermetallic compounds at the interface of copper-aluminum composite strips, the method comprising the following steps: S100, initialize system parameters and calibrate dynamic reference; input and store the continuous running line speed of the copper-aluminum composite strip production line, the physical distance between the dual-frequency eddy current sensor module 101 and the multi-physics field action zone, the reference conductivity of pure metal under no initial strain state and the set target metallurgical bonding interface intermetallic compound layer thickness in the controller; at the same time, calculate the effective heat preservation time of the strip in the heating zone based on the continuous running line speed of the production line and the effective heating section length of the induction heating module 107. S200: Dual-frequency eddy current sensor module 101 and data acquisition module 102 are used to perform online feature extraction of dual-frequency alternating eddy currents. When the copper-aluminum composite strip to be processed continuously passes through the feature extraction area, the dual-frequency eddy current sensor module 101 synchronously injects high-frequency excitation signal and low-frequency excitation signal into the strip. The data acquisition module 102 records and converts the high-frequency equivalent conductivity and low-frequency equivalent conductivity of the strip at a specific longitudinal position in real time to form a digital conductivity response signal. S300, the feature decoupling module 103 and the dynamic mapping module 104 are used to perform strain space decoupling and diffusion dynamic mapping; the feature decoupling module 103 calls the electromagnetic skin depth weighting coefficient to perform differential operation on the high frequency equivalent conductivity and the low frequency equivalent conductivity, strips the conductivity change caused by the macroscopic deformation of the surface layer, and extracts the interface micro-strain characteristic value reflecting the micro-lattice distortion of the copper-aluminum interface; the dynamic mapping module 104 substitutes the interface micro-strain characteristic value into the preset atomic diffusion mapping model and calculates the strain-induced diffusion enhancement factor used to correct the interface atomic diffusion rate; S400, the instruction generation module 105 performs inverse calculation of the intermetallic compound growth kinetics and generates three-dimensional control instructions. The instruction generation module 105 substitutes the thickness of the intermetallic compound layer at the target metallurgical interface, the effective holding time, and the input strain-induced diffusion enhancement factor into the embedded Arrhenius inverse dynamics model to solve for the optimal target temperature required to suppress or compensate for local strain, and converts it into a target heating power instruction in combination with the energy balance equation. At the same time, the instruction generation module 105 calculates the target heating frequency instruction and the cooling gas flow instruction in segments according to the quantitative relationship between the interface micro-strain characteristic value and the set threshold. The target heating power instruction, the target heating frequency instruction, and the cooling gas flow instruction are combined to form a three-dimensional control instruction set for the strip segment at a specific longitudinal position. In S500, the delay compensation module 106 is used to buffer the spatial delay queue and drive the induction heating module 107 and the gas cooling module 108 to synchronously execute feedforward control. The delay compensation module 106 calculates the spatial delay time based on the continuous running speed and physical distance of the production line. The three-dimensional control command set is stepped and shifted in sync with the production speed in the first-in-first-out data queue established by the delay compensation module 106. When the strip segment at a specific longitudinal position actually moves to the multi-physics field action zone as the production line moves, the delay compensation module 106 pops the corresponding three-dimensional control command set from the end of the queue and sends it out in parallel. The induction heating module 107 dynamically adjusts the inverter output according to the target heating power command and the target heating frequency command. The gas cooling module 108 synchronously adjusts the opening of the servo proportional valve according to the cooling gas flow command. The induction heating module 107 and the gas cooling module 108 jointly construct a quasi-steady-state temperature gradient in the strip thickness direction to complete the thermodynamic intervention on the interface of the current strip segment.

[0036] In specific implementation, step S100 may include the following sub-steps: S110, input and store the static process parameters of the copper-aluminum composite strip in the controller; the operator configures and writes the basic process parameters of the production line to the data storage area of ​​the instruction generation module 105 and the delay compensation module 106 through the peripheral input device of the controller or the industrial Ethernet interface. The written process parameters include the continuous running line speed of the copper-aluminum composite strip production line, the width of the copper-aluminum composite strip, the total thickness of the copper-aluminum composite strip, the average density of the copper-aluminum composite strip, the specific heat capacity of the copper-aluminum composite strip, the electrothermal conversion efficiency of the induction heating module 107, and the physical distance between the measurement center line of the dual-frequency eddy current sensor module 101 and the longitudinal geometric center line of the induction heating coil of the induction heating module 107. To ensure that the system does not trigger a calculation dead zone during the start-up, deceleration, or temporary shutdown of the production line, the controller also presets a lower limit threshold for linear speed. In this embodiment, the lower limit threshold ranges from 0.5 meters to 1.5 meters per minute. When the actual operating line speed is lower than the lower limit threshold, the system automatically locks the previous calculated value or switches to standby safe power to ensure the integrity of the control algorithm logic.

[0037] S120, configure the material property reference constant and control boundary parameters in the feature decoupling module 103 and the dynamic mapping module 104; write the reference conductivity of the pure metal in the state without initial strain to the feature decoupling module 103 through the register configuration unit of the controller. The reference conductivity is the standard physical parameter of the copper-aluminum composite strip in the state without lattice distortion. It is obtained by measuring the standard annealed sample at an ambient temperature of 20°C using a standard eddy current tester before production; at the same time, input and store the set target metallurgical bonding interface intermetallic compound layer thickness to the instruction generation module 105. The target metallurgical bonding interface intermetallic compound layer thickness is set according to the bending performance and resistivity technical indicators required in the subsequent application of the copper-aluminum composite strip. The value range is 1μm to 5μm to prevent the interface from becoming brittle or the bonding strength from being insufficient. Furthermore, to support the adaptive iteration of subsequent calculation modules, a set of control constants is also imported into the feature decoupling module 103, the dynamic mapping module 104, and the instruction generation module 105 in this step. The set of control constants includes an interface strain tolerance threshold, a skin depth weighting coefficient, empirical polynomial constant coefficients, a frequency compensation gain coefficient, and a cooling gas flow rate compensation gain coefficient. The interface strain tolerance threshold is determined based on the elastic limit stage endpoint of the material's strain hardening curve and is used as the quantization boundary for determining whether forced convection heat transfer control should be initiated. In this embodiment, the interface strain tolerance threshold ranges from 0.5 MS / m to 2.0 MS / m, the frequency compensation gain coefficient ranges from 5 kHz / (MS / m) to 15 kHz / (MS / m), and the cooling gas flow rate compensation gain coefficient ranges from 10 (m² / m² / s ... 3 / h) / (MS / m) to 20 (m 3 / h) / (MS / m).

[0038] S130, the effective holding time of the heat treatment stage is calculated by the instruction generation module 105; the instruction generation module 105 automatically retrieves a constant pre-pre-stored in the controller's read-only memory, which is the effective heating section length of the induction heating module 107, and the effective heating section length is the physical length of the induction heating coil along the material running direction of the production line; the microprocessor of the instruction generation module 105 performs a division operation, dividing the effective heating section length by the continuous running line speed of the production line obtained in step S110, thereby determining the effective holding time of a specific strip section within the heating zone. The formula for calculating the effective holding time satisfies: ; In the formula, For effective heat preservation time; The effective physical length of the multiphysics field interaction region; The continuous operating speed of the production line.

[0039] The calculated effective heat preservation time is transmitted in real time and stored in the dynamic calculation cache of the instruction generation module 105 as a known time-domain boundary condition for the inverse solution model of intermetallic compound growth in subsequent steps.

[0040] In specific implementation, step S200 may include the following sub-steps: S210, a dual-frequency alternating electromagnetic excitation layout is established on a continuously moving copper-aluminum composite strip using a dual-frequency eddy current sensor module 101; an alternating current is passed through the excitation coil in the dual-frequency eddy current sensor module 101, inducing dual-frequency alternating eddy currents inside the continuously moving strip. During this process, high-frequency electromagnetic excitation signals and low-frequency electromagnetic excitation signals are synchronously injected into the strip; due to the skin effect of the electromagnetic field, the field strength of the alternating electromagnetic field decreases exponentially with the increase of penetration depth when it enters the interior of the metal conductor, and its characteristic penetration depth satisfies the following physical equation: ; In the formula, Electromagnetic skin depth; The frequency is the alternating electromagnetic excitation frequency; The magnetic permeability of the copper-aluminum composite strip; The conductivity of the copper-aluminum composite strip is given.

[0041] Based on this physical relationship, the penetration range of the electromagnetic field in the thickness direction of the strip can be adjusted by setting different alternating electromagnetic excitation frequencies. In order to accurately separate the macroscopic deformation of the surface layer and the internal interface strain, the alternating frequency of the high-frequency electromagnetic excitation signal is preferably 100kHz to 500kHz. Within this frequency range, the corresponding electromagnetic skin depth is constrained within the outer surface layer of the main heating surface of the copper-aluminum composite strip, which is used to extract the macroscopic deformation and residual strain information near the outer surface layer caused by the previous cold rolling. The alternating frequency of the low-frequency electromagnetic excitation signal is preferably 10kHz to 50kHz. Since the low-frequency signal has stronger penetration ability, its corresponding electromagnetic skin depth can penetrate the outer surface layer of the main heating surface and extend to the metallurgical bonding interface layer inside the copper and aluminum, which is used to collect the microscopic dislocation distortion and atomic lattice deformation characteristics at the interface.

[0042] S220, the data acquisition module 102 captures and digitizes the secondary alternating electromagnetic signal fed back from the strip; when a micro-segment of the strip with specific strain characteristics continuously passes over the feature extraction area, the detection coil in the dual-frequency eddy current sensor module 101 is affected by the internal eddy current secondary magnetic field and generates an induced electromotive force change; the data acquisition module 102 performs high-speed analog-to-digital conversion on the induced electromotive force signal to obtain high-frequency impedance components and low-frequency impedance components; the conversion of high and low frequency impedance signals to conductivity values ​​is performed based on existing orthogonal phase demodulation and conductivity inversion methods; The high-frequency equivalent conductivity and low-frequency equivalent conductivity corresponding to the specific longitudinal position of the current strip are thus demodulated and calculated. In order to prevent the spatiotemporal mapping misalignment caused by strip speed fluctuations or local slippage, the data acquisition module 102 performs hardware synchronous triggering with the speed encoder on the production line, and records the impedance signal with a fixed longitudinal displacement step pulse as a reference. The data acquisition module 102 converts the discrete conductivity values ​​into digital conductivity time-series response signals that correspond one-to-one with the longitudinal spatial coordinates of the strip, and sends them to the feature decoupling module 103 in real time.

[0043] See attached document Figure 3 In specific implementation, step S300 may include the following sub-steps: S310, the feature decoupling module 103 performs interface micro-strain feature decoupling calculation based on electromagnetic skin difference; the feature decoupling module 103 receives the digital conductivity time-series response signal input by the data acquisition module 102 in real time. The digital conductivity time-series response signal includes the high-frequency equivalent conductivity and low-frequency equivalent conductivity of the strip at a specific longitudinal position. During the plastic processing of metal conductors, macroscopic mechanical stress such as cold rolling causes work hardening and macroscopic residual deformation of the outer surface layer of the strip, while the copper-aluminum interface is mainly enriched with microscopic lattice distortion and dislocation defects. These two deformation characteristics at different spatial levels will have different degrees of weakening effect on the macroscopic electromagnetic conductivity of the conductor. High-frequency electromagnetic excitation signals can only penetrate the surface layer due to the skin effect, so the change in conductivity they provide is mainly caused by surface processing deformation. Low-frequency electromagnetic excitation signals, however, can penetrate to the interface layer, and the change in conductivity they provide includes the combined effect of surface processing deformation and interface micro-distortion. The feature decoupling module 103 retrieves the skin depth weighting coefficients pre-stored in its internal non-volatile memory, performs a difference matrix operation on the low-frequency equivalent conductivity and the high-frequency equivalent conductivity, and uses the surface high-frequency equivalent conductivity signal as a background reference. It then quantitatively separates and eliminates the interference contribution of the macroscopic cold-rolling deformation of the surface layer to the overall resistivity from the fully penetrated low-frequency equivalent conductivity signal, thereby separating and extracting the interface micro-strain characteristic values ​​that mainly reflect the micro-lattice distortion of the copper-aluminum interface. Its algebraic decoupling formula satisfies: ; In the formula, The characteristic value of interfacial microstrain at a specific longitudinal location; The low-frequency equivalent conductivity at a specific longitudinal position; For the specific longitudinal position coordinates of the copper-aluminum composite strip; For skin depth weighting coefficients; It represents the high-frequency equivalent conductivity at a specific longitudinal position.

[0044] The range of the skin depth weighting coefficient is determined based on the thickness ratio of the copper layer to the aluminum layer in the strip substrate and the attenuation gradient of the high and low frequency excitation signals on both sides of the heterometallic interface layer. Its preferred value is constrained to be in the range of 0.85 to 1.15. Specifically, the skin depth weighting coefficient is obtained by calibrating the conductivity ratio of a reference copper-aluminum composite strip without lattice distortion at high and low frequencies. To avoid the dead zone problem caused by negative differential results due to electromagnetic interference in the production line or transient surge noise from sensors, which could lead to the collapse of subsequent dynamic algorithm logic, the feature decoupling module 103 is equipped with boundary limiting filtering logic. When the calculated interface micro-strain characteristic value is less than zero, the controller forcibly resets the interface micro-strain characteristic value of the current coordinate point to zero, ensuring the stability and continuity of data flow to subsequent modules.

[0045] S320, the interface micro-strain characteristic value is converted into strain-induced diffusion enhancement factor by the dynamic mapping module 104; after the characteristic decoupling module 103 completes the electromagnetic skin difference operation, the calculated interface micro-strain characteristic value is output to the dynamic mapping module 104 in real time; in the solid-state metallographic transformation and interface dynamics principle, micro-lattice distortion and high-density dislocations will form a fast diffusion channel for atoms, reduce the apparent diffusion activation energy required for copper atoms and aluminum atoms to penetrate each other and react in the solid phase, and cause the intermetallic compound nuclei at the interface to nucleate and grow at a rate greater than that of the reference state; In terms of physical mechanism, the increase in impedance caused by lattice distortion is proportional to the dislocation density, and the increase in dislocation density provides a channel with low activation energy for atomic diffusion. To quantitatively evaluate the amplification effect of this local microscopic strain field on the atomic diffusion rate, the kinetic mapping module 104 substitutes the received interface microscopic strain characteristic values ​​into a preset atomic diffusion mapping model. The atomic diffusion mapping model uses an empirical polynomial equation based on crystal defect diffusion kinetics calibration to map the microscopic electromagnetic property characterization into thermodynamic correction parameters, and calculates the strain-induced diffusion enhancement factor used to correct the apparent atomic diffusion rate at the interface. The empirical polynomial equation satisfies the following mathematical relationship: ; In the formula, The strain-induced diffusion enhancement factor at a specific longitudinal location; The constant coefficients of the quadratic empirical polynomial; The characteristic value of interfacial microstrain at a specific longitudinal location; The constant coefficients of the empirical polynomial for the first-order term; The constant term is the constant coefficient of the empirical polynomial.

[0046] The empirical polynomial constant coefficients were obtained through a combination of transmission electron microscopy dislocation density statistical experiments and diffusion couple high-temperature tracer experiments. Specifically, they were obtained by measuring the change in interfacial conductivity of the strip under different deformations and the corresponding atomic diffusion coefficients, and then performing polynomial fitting using the least squares method. These coefficients were used to establish a quantitative conversion relationship between microscopic physical damage and transient physicochemical diffusion coefficients. The preferred range for the quadratic term empirical polynomial constant coefficients is 0.02 to 0.08, and the preferred range for the linear term empirical polynomial constant coefficients is 1.25 to 3.45. The constant term empirical polynomial constant coefficients correspond to the baseline property correction values ​​under no initial strain conditions and are fixed at 1.0. Meanwhile, to prevent the polynomial from calculating extreme values ​​that do not conform to physical reality under large strain noise, the kinetic mapping module 104 is equipped with output control dead zone protection logic, which forcibly limits the lower limit of the calculated strain-induced diffusion enhancement factor to 1.0, thereby avoiding the occurrence of negative temperatures or no solution in subsequent thermodynamic inverse calculations. After completing the strain-induced diffusion enhancement factor calculation for a single point segment, the kinetic mapping module 104 injects the parameter into the input register of the instruction generation module 105 in real time in the form of a data packet, as the underlying thermodynamic boundary input for subsequent calculation of asymmetric multiphysics control instructions.

[0047] See attached document Figure 4 In specific implementation, step S400 may include the following sub-steps: S410, the instruction generation module 105 is used to perform inverse kinetic calculations in the macroscopic thermodynamic energy dimension; the instruction generation module 105 obtains in real time the thickness of the intermetallic compound layer at the target metallurgical interface, the effective holding time, and the input strain-induced diffusion enhancement factor; during the solid-phase diffusion reaction, the thickness of the intermetallic compound growth layer exhibits a nonlinear kinetic response relationship with the diffusion coefficient and the holding time; in order to control the compound growth thickness within the preset target range under the premise that microscopic lattice distortion intensifies interfacial atomic diffusion.

[0048] The microprocessor of the instruction generation module 105 calls the embedded Arrhenius inverse dynamics model to solve for the optimal target temperature required for the current strip segment. Subsequently, in order to convert this optimal target temperature into an explicit physical quantity that can be directly responded to by the hardware actuator, the instruction generation module 105, in conjunction with the heat balance equation, calculates the target heating power command used to intervene in the local interface. Its specific inverse dynamics and energy balance calculation formulas satisfy: ; ; In the formula, The target heating temperature of the strip at a specific longitudinal position; The activation energy for diffusion at the copper-aluminum interface; It is the ideal gas constant; The strain-induced diffusion enhancement factor at a specific longitudinal location; Pre-diffusion factor; For effective heat preservation time; The thickness of the intermetallic compound layer at the target metallurgical interface; Target heating power command; The average density of the copper-aluminum composite strip; Specific heat capacity of copper-aluminum composite strip; The width of the copper-aluminum composite strip; This refers to the total thickness of the copper-aluminum composite strip; The continuous operating speed of the production line; The initial temperature of the strip before it enters the multiphysics field interaction region; The electrothermal conversion efficiency of the induction heating module 107.

[0049] To ensure the completeness of the calculation logic, the instruction generation module 105 includes denominator and logarithmic term value verification logic before performing the aforementioned logarithmic operation. When the thickness of the intermetallic compound layer at the target metallurgical interface approaches zero, a preset minimum control thickness value is automatically used for substitution to avoid the generation of mathematical singularities. In this embodiment, the minimum control thickness value ranges from 0.1 μm to 0.5 μm; the preferred range for the aforementioned copper-aluminum interface diffusion activation energy is 110 kJ / mol to 130 kJ / mol, and the preferred range for the pre-diffusion factor is 0.0001 m. 2 / s to 0.0005m 2 / s, these physical constants were determined by cross-sectional metallographic microscopic observation of copper-aluminum diffusion samples at different temperatures, and by performing linear regression fitting in combination with experimental data on growth thickness; the initial temperature of the strip before entering the multiphysics field action zone was extracted in real time online by an infrared temperature sensor placed at the entrance of the induction heating area to ensure the dynamic accuracy of the thermal balance calculation.

[0050] S420, the instruction generation module 105 performs segmented calculations of the spatial heat source distribution and forced convection heat transfer boundary. To meet the adaptive adjustment requirements under different local deformation damage levels, the control strategy establishes discretized segmented control boundaries based on the severity of interface distortion. The instruction generation module 105 performs quantitative condition judgments on the received interface micro-strain characteristic values ​​and preset interface strain tolerance thresholds. When the interface micro-strain characteristic value is less than or equal to the interface strain tolerance threshold, it indicates that there are few micro-defects inside the strip, and the system executes the conventional uniform heating mode, outputting the basic heating frequency and minimum cooling gas flow rate. When the interface micro-strain characteristic value is greater than the interface strain tolerance threshold, it indicates that micro-distortion has accelerated atomic diffusion, and the system adaptively switches to a strong deformation compensation mode. By introducing an adaptive gain coefficient, the target heating frequency instruction and cooling gas flow rate instruction are solved segmentally to dynamically adjust the spatial heat source distribution. The segmented calculation formula satisfies: ; ; In the formula, Command for the target heating frequency at a specific longitudinal position; Based on the heating frequency; The characteristic value of interfacial microstrain at a specific longitudinal location; This refers to the interfacial strain tolerance threshold. This is the upper limit of the induction heating frequency; This is the frequency compensation gain coefficient; Command for cooling gas flow rate at a specific longitudinal position; Minimum cooling gas flow rate; This is the upper limit for gas flow rate; This is the gain coefficient for cooling gas flow rate compensation.

[0051] The aforementioned basic heating frequency ranges from 20kHz to 30kHz, and the upper limit of the induction heating frequency ranges from 50kHz to 80kHz. This range ensures that the depth of electromagnetic energy concentration on the surface is less than one-third of the total thickness of the strip. The minimum cooling gas flow rate ranges from 5m³ / h. 3 / h to 10m 3 / h, which is mainly used to maintain the basic dustproof pressure of the nozzle, and the upper limit of the gas flow rate is 40m. 3 / h to 60m 3 / h, to provide sufficient back-side cooling rate; by embedding a maximum value limit term in the piecewise function, a hardware saturation upper limit is set to ensure that the solution result is always within the safe operating range of the actuator.

[0052] S430, implement the three-dimensional physical field collaborative reconstruction of the quasi-steady-state temperature gradient in the thickness direction; after obtaining the target heating power command, target heating frequency command and cooling gas flow command, the command generation module 105 reconstructs these spatially decoupled but physically highly coordinated control parameters into a unified three-dimensional control command set; utilizing the skin depth sensitivity of electromagnetic induction heating to frequency, and the hysteresis of heat conduction in the metal strip in a short response time, the tending-to-uniform thermal equilibrium state is forcibly broken in the thin strip; when the strong deformation compensation mode is triggered, the increased target heating frequency command causes the electromagnetic heating layer generated by the induction heating module 107 to shrink towards the surface of the main heating surface of the strip, at which time the heating center is far away from the copper-aluminum interface; Meanwhile, high-flow-rate cooling gas performs strong convective heat transfer on the back side of the strip, establishing an artificial heat trap along the heat transfer path from the interface layer to the back side. Through the longitudinal spatial synergy of directional heat generation on the surface and forced heat dissipation on the back side, an asymmetric quasi-steady-state temperature gradient is established in the thickness direction of the copper-aluminum composite strip. This helps to maintain a high temperature sufficient for softening annealing of the matrix on the outer surface layer while keeping the actual temperature at the copper-aluminum interface below a safe temperature that inhibits the growth of intermetallic compounds, thereby achieving precise control over the thickness of the brittle phase at the interface.

[0053] In specific implementations, step S500 provided by the present invention may include the following sub-steps: S510, the delay compensation module 106 performs dynamic allocation of the spatiotemporal delay of the first-in-first-out data queue; the delay compensation module 106 receives in real time the three-dimensional control command set calculated and output by the command generation module 105, which corresponds to the specific longitudinal position of the strip in the feature extraction area; since there is a fixed physical distance between the detection center line of the dual-frequency eddy current sensor module 101 and the actual action center of the induction heating module 107 and the gas cooling module 108, the command cannot be executed immediately after the command is calculated, and spatial and temporal alignment is required; the delay compensation module 106 reads the physical distance recorded during system initialization and the real-time continuous line speed of the production line, and calculates the spatial delay time under ideal conditions, the calculation formula of which satisfies: ; In the formula, For spatial delay time; This refers to the physical distance between the dual-frequency eddy current sensor module 101 and the multiphysics field interaction zone. The continuous operating speed of the production line.

[0054] To overcome the cumulative error in time calculation caused by fluctuations in the production line's operating speed, the delay compensation module 106, in conjunction with a high-precision speed encoder on the main drive roller, discretizes the spatial delay time into the control cycle steps corresponding to the physical displacement. The delay compensation module 106 allocates a continuous address space in the controller's memory as a first-in-first-out data queue, and pushes the received three-dimensional control instruction set along with the current longitudinal position coordinates as a data structure to the end of the data queue. When a displacement pulse is received from the encoder, the step pointer inside the data queue moves forward synchronously. To ensure the integrity of the algorithm logic, the delay compensation module 106 incorporates speed lower limit detection and queue overflow protection logic. The preferred range of the aforementioned safety lower limit is 0.5 meters to 1.5 meters per minute. The maximum allocation depth of the aforementioned first-in-first-out data queue is calculated by dividing the physical distance by the single-step displacement resolution of the speed encoder. When the continuous operating speed of the production line is lower than the set safety lower limit or the line stops, the data queue pauses pointer shifting and stops pushing in new data, entering a data holding state, thereby avoiding division-by-zero dead zones and memory overflow crashes.

[0055] S520 utilizes the delay compensation module 106 to perform transient drive and parallel feedforward execution control of the hardware actuator; when a strip micro-segment at a specific longitudinal position is about to reach the multi-physics field action zone as the production line moves, the delay compensation module 106 prepares to pop the instruction set at the front of the data queue; in order to offset the inherent physical response delay of the underlying actuator, the delay compensation module 106 combines the current real-time continuous operating line speed of the production line, converts the pre-calibrated hardware response compensation time into the corresponding advance spatial steps, and triggers the instruction pop-up action in advance when the strip reaches the corresponding advance steps before reaching the multi-physics field action zone; the hardware response compensation time is determined by taking the maximum value of the electrical response time of the inverter power supply and the mechanical action time of the servo proportional valve, with a preferred value range of 10 milliseconds to 50 milliseconds; After the instruction set is popped up, the delay compensation module 106 sends drive instructions in parallel through the industrial bus. On the one hand, it sends the target heating power instruction and the target heating frequency instruction to the induction heating module 107. The digital signal processor inside the induction heating module 107 adjusts the switching frequency and duty cycle of the insulated gate bipolar transistor of the high-frequency inverter topology according to the received instructions to realize the transient switching of radio frequency energy and skin depth. On the other hand, it sends the cooling gas flow instruction to the gas cooling module 108 to drive the servo proportional valve in the gas circuit system to adjust the opening degree and accurately output the required back-side cooling gas flow. The specific phase-locked loop frequency tracking and power regulation closed loop inside the induction heating module 107, as well as the gas pressure and flow rate tracking control of the gas cooling module 108, are carried out according to existing industrial automatic control methods. Through the above-mentioned spatiotemporal alignment and parallel feedforward drive, the asymmetric multiphysics field can match the high deformation defect region corresponding to the strip in a timely manner, and complete the control action for the growth of intermetallic compounds.

[0056] To aid in understanding the technical solution of this invention, the following is an application example of controlling the intermetallic compound at the interface of cold-rolled copper-aluminum composite strip in a continuous annealing process.

[0057] The total thickness of the cold-rolled copper-aluminum composite strip is set to 2.0 mm, with a copper layer thickness of 0.5 mm and an aluminum layer thickness of 1.5 mm. The width of the composite strip is 200 mm. The controller receives basic parameters, the continuous running linear speed of the strip is set to 0.2 m / s, the effective heating length of the multi-physics field action zone is calibrated to 1.8 m, the physical distance from the center of the dual-frequency eddy current sensor module 101 to the center of the heating zone is calibrated to 6.0 m, and the electrothermal conversion efficiency input value of the induction heating module 107 is 0.65.

[0058] The instruction generation module 105 calculates the effective heat preservation time of the strip as 9.0s based on the running linear speed and effective heating length. The system inputs the thickness of the intermetallic compound layer at the target metallurgical bonding interface as 2.0μm and sets the interface strain tolerance threshold to 1.0MS / m.

[0059] During continuous operation, the dual-frequency eddy current sensor module 101 injects alternating electromagnetic excitation into the strip. The data acquisition module 102 measures the high-frequency equivalent conductivity at a specific longitudinal position of the strip as 42.5 MS / m and the low-frequency equivalent conductivity as 42.875 MS / m. The feature decoupling module 103 calls the skin depth weighting coefficient with a value of 0.95 to perform differential calculation and calculates the interface micro-strain characteristic value as 2.5 MS / m.

[0060] The dynamic mapping module 104 inputs polynomial coefficients, with the quadratic coefficient set to 0.04, the linear coefficient set to 1.8, and the constant term set to 1.0. After calculation, the strain-induced diffusion enhancement factor corresponding to the coordinate point is output as 5.75.

[0061] The instruction generation module 105 calls the inverse dynamics model. In order to counteract the diffusion enhancement effect and maintain the target layer thickness of 2.0 μm, the calculation results show that the local heat treatment target temperature needs to be reduced to 395℃. Based on the energy balance equation, the output target heating power instruction is 85kW.

[0062] The instruction generation module 105 compares the measured interfacial microstrain characteristic value of 2.5 MS / m with the tolerance threshold of 1.0 MS / m. Since the characteristic value is greater than the tolerance threshold, the system executes a segmented calculation instruction. The basic heating frequency is set to 20 kHz, the frequency compensation gain coefficient is 12 kHz / (MS / m), the target heating frequency is 38 kHz, and the minimum cooling gas flow rate is set to 10 m³ / s. 3 / h, the flow compensation gain coefficient is 15 (m 3 The calculated cooling gas flow rate is 32.5 m³ / h / (MS / m). 3 / h, the above three parameters together form a three-dimensional control command set for this coordinate micro-segment.

[0063] The delay compensation module 106 calculates a spatial delay time of 30s based on the physical distance and linear velocity. After deducting the 0.04s response compensation time set by the hardware actuator, the delay compensation module 106 triggers a command when the actual running time of the strip reaches 29.96s. The induction heating module 107 adjusts the inverter output frequency to 38kHz and the power to 85kW, while the gas cooling module 108 adjusts the proportional valve output to 32.5m. 3 / h of cold air creates a specific temperature gradient in the thickness direction of the strip, completing the thermodynamic compensation for this micro-segment of coordinates.

[0064] Table 1: Comparison of Measured Parameters of Continuously Annealed Copper-Aluminum Composite Strip 50 0.38 1.84 1.96 41.2 39.8 100 0.82 1.91 2.04 39.5 40.1 150 2.54 4.37 2.11 18.4 38.5 200 3.12 5.12 2.08 11.7 37.9 250 0.45 1.88 1.95 40.8 41.0 According to Table 1 and Figure 5 and Figure 6 It can be seen that the increase in the interfacial microstrain characteristic value is directly related to the abnormal thickening of the intermetallic compound. At the longitudinal positions of 150m and 200m, the microstrain characteristic values ​​extracted by dual-frequency feature decoupling reached 2.54MS / m and 3.12MS / m, respectively, exceeding the set system tolerance threshold. The traditional treatment method of fixed heating power and symmetrical heat transfer resulted in interface runaway in these two intervals, and the thickness of the intermetallic compound increased to 4.37μm and 5.12μm, respectively. The corresponding physical property test showed that the formation of the thick and brittle phase caused the macroscopic T-shaped peel strength of the material to decrease to 18.4N / mm and 11.7N / mm, and brittle fracture of the interface occurred.

[0065] For the same initial defect strip, the measured thickness of the intermetallic compound at each longitudinal sampling point ranges from 1.95 μm to 2.11 μm, with the overall average value close to the target control line of 2.0 μm. Due to the interference of macroscopic deformation of the stripped surface, the system accurately identifies the internal lattice distortion in the 150 m and 200 m sections. The instruction generation module 105 automatically increases the induction heating frequency to move the heat source to the surface, while increasing the flow rate of the back cooling gas to construct a heat trap.

[0066] Through the coordinated action of the aforementioned physical fields, the high-temperature accumulation effect at the interface is suppressed. The experimental results show that the proposed scheme maintains the peel strength value in the microscopic high-deformation damage area, and the overall physical performance data converges within the range of 37.9 N / mm to 41.0 N / mm. The system achieves continuous and stable operation of strip segments with different initial lattice states without slowing down the production line, eliminating large fluctuations in the compound thickness, and confirming the engineering effectiveness of the diffusion kinetics inverse solution algorithm and spatiotemporal feedforward control in suppressing brittle phases at the interface.

[0067] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for controlling intermetallic compounds at the interface of copper-aluminum composite strips, characterized in that, The method includes the following steps: Obtain the continuous operating line speed of the production line and calculate the effective heat preservation time; High-frequency and low-frequency excitation signals are synchronously injected into a continuously moving copper-aluminum composite strip to obtain the high-frequency equivalent conductivity and low-frequency equivalent conductivity at a specific longitudinal position. The interface micro-strain characteristic values ​​are extracted by performing a difference operation on the high-frequency equivalent conductivity and the low-frequency equivalent conductivity, and then substituted into the atomic diffusion mapping model to calculate the strain-induced diffusion enhancement factor. The target metallurgical interface intermetallic compound layer thickness, effective heat preservation time and strain-induced diffusion enhancement factor are substituted into the Arrhenius inverse dynamics model to obtain the target heating power command. The target heating frequency command and cooling gas flow command are calculated based on the interface micro-strain characteristic value and combined to generate a three-dimensional control command set. The spatial delay time is calculated based on the continuous operating speed of the production line, and the first-in-first-out data queue shift delay is executed on the three-dimensional control instruction set to drive the induction heating module and the gas cooling module to be synchronously controlled.

2. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 1, characterized in that, The process of obtaining the continuous operating line speed of the production line and calculating the effective heat preservation time specifically includes: The continuous operating line speed of the copper-aluminum composite strip production line and the effective heating section length of the induction heating module are obtained. Divide the effective heating section length by the continuous operating speed of the production line to determine the effective heat preservation time of a specific strip section within the heating zone; A preset lower limit threshold for linear speed is set. When the continuous operating linear speed of the actual production line is lower than the preset lower limit threshold, the previously calculated value is automatically locked to maintain the integrity of the control logic.

3. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 1, characterized in that, The synchronous injection of high-frequency and low-frequency excitation signals into the continuously moving copper-aluminum composite strip specifically includes: Set the alternating frequency of the high-frequency electromagnetic excitation signal so that the corresponding electromagnetic skin depth is constrained within the outer surface layer of the main heating surface of the copper-aluminum composite strip. The alternating frequency of the low-frequency electromagnetic excitation signal is set so that the corresponding electromagnetic skin depth penetrates the outer surface layer of the main heating surface and extends to the metallurgical bonding interface layer inside the copper and aluminum of the copper-aluminum composite strip. The induced electromotive force signal fed back from the copper-aluminum composite strip is collected and subjected to high-speed analog-to-digital conversion to demodulate the high-frequency equivalent conductivity and low-frequency equivalent conductivity corresponding to the specific longitudinal position.

4. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 1, characterized in that, The step of extracting the interface micro-strain characteristic values ​​by performing a difference operation between the high-frequency equivalent conductivity and the low-frequency equivalent conductivity specifically includes: Obtain the skin depth weighting coefficients obtained from the calibration of a pre-acquired reference copper-aluminum composite strip without lattice distortion; The low-frequency equivalent conductivity and the high-frequency equivalent conductivity are subjected to a difference matrix operation. The high-frequency equivalent conductivity is used as a background reference and combined with the skin depth weighting coefficient to quantitatively remove the interference contribution of surface macroscopic deformation from the low-frequency equivalent conductivity, and the interface micro-strain characteristic value reflecting the micro-lattice distortion of the copper-aluminum interface is extracted.

5. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 4, characterized in that, After extracting the interface micro-strain characteristic values, the method further includes a boundary limiting filtering step: Determine whether the calculated interface micro-strain characteristic value is less than zero. If it is determined to be less than zero, force the interface micro-strain characteristic value at the current coordinate point to be reset to zero.

6. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 1, characterized in that, The atomic diffusion mapping model uses an empirical polynomial equation calibrated based on crystal defect diffusion kinetics to solve for the strain-induced diffusion enhancement factor, specifically including: The empirical polynomial equation includes a quadratic term, a linear term, and a constant term, wherein the constant term corresponds to the reference property correction value under the condition of no initial strain. The strain-induced diffusion enhancement factor is obtained by substituting the interface micro-strain characteristic values ​​into the empirical polynomial equation and mapping the calculation. When outputting the calculation results, a lower limit value for the strain-induced diffusion enhancement factor is forcibly limited to avoid unsolvable situations in subsequent thermodynamic inverse solution calculations.

7. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 1, characterized in that, The step of substituting the preset target metallurgical bonding interface intermetallic compound layer thickness, effective holding time, and strain-induced diffusion enhancement factor into the Arrhenius inverse kinetic model to obtain the target heating power command specifically includes: The embedded Arrhenius inverse dynamics model is invoked to solve inversely the optimal target temperature required to suppress or compensate for local strain at a specific longitudinal position. The initial temperature of the copper-aluminum composite strip before entering the multi-physics field interaction zone is obtained by real-time acquisition through an infrared temperature sensor. By combining the energy balance equation, the difference between the optimal target temperature and the initial temperature, and taking into account the average density and specific heat capacity of the copper-aluminum composite strip and the electrothermal conversion efficiency of the induction heating module, is converted into the target heating power command.

8. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 1, characterized in that, The calculation of the target heating frequency command and cooling gas flow rate command based on the interface micro-strain characteristic values ​​employs a segmented calculation strategy based on the interface strain tolerance threshold. When the characteristic value of the interface micro-strain is less than or equal to the preset interface strain tolerance threshold, the basic heating frequency and minimum cooling gas flow rate are output. When the micro-strain characteristic value of the interface is greater than the preset interface strain tolerance threshold, the target heating frequency command is solved piecewise using the preset frequency compensation gain coefficient, the cooling gas flow command is solved piecewise using the preset cooling gas flow compensation gain coefficient, and the solution result is limited to the safe working range of the actuator by the maximum value limiting term embedded in the function.

9. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 1, characterized in that, The step of performing a first-in-first-out data queue shift delay on the three-dimensional control command set specifically includes: The physical distance between the dual-frequency eddy current sensor module and the multi-physics field action area is read and divided by the continuous operating line speed of the production line to calculate the spatial delay time under ideal conditions. A first-in-first-out data queue is established, the received three-dimensional control command set is pushed into the end, and when the displacement pulse fed back by the speed encoder is received, the stepping pointer inside the data queue is driven to move forward synchronously. When it is determined that the continuous operating speed of the production line is lower than the set safety lower limit or the line stops, the data queue pause pointer shifts to enter the data hold state.

10. The method for controlling intermetallic compounds at the interface of copper-aluminum composite strips according to claim 9, characterized in that, Before the synchronous control of the driving induction heating module and the gas cooling module, a transient driving and feedforward execution compensation step is also included: Obtain the pre-calibrated hardware response compensation time, which is determined by taking the maximum value of the electrical response time of the inverter power supply and the mechanical action time of the servo proportional valve. The hardware response compensation time is converted into lead space steps by combining the real-time continuous operating line speed of the production line. When the copper-aluminum composite strip reaches the multiphysics field action zone by the corresponding advance step, the three-dimensional control instruction set is triggered to pop out from the end of the first-in-first-out data queue in advance, and drive control instructions are issued in parallel.

Images