PCB board nano copper ink sintering process temperature field dynamic compensation control method and system

Abstract

This invention relates to the field of printed circuit board manufacturing control, and provides a method and system for dynamic temperature field compensation control during the sintering process of nano-copper ink on PCB boards. The method includes: meshing the three-dimensional contour of the nano-copper ink layer, generating a heat capacity load distribution matrix based on the original PCB board design file, and mapping it to an initial energy field matrix; acquiring original multispectral thermal radiation signals, obtaining the true temperature field distribution through colorimetric thermometry, and generating a temperature field gradient vector based on the temperature rise difference between adjacent meshes; capturing the real-time thermal response rate, identifying the latent heat energy value of phase change based on the initial energy field matrix, and generating a phase change energy compensation matrix; generating an execution power command based on the temperature field gradient vector, the initial energy field matrix, the phase change energy compensation matrix, and the real-time thermal response rate, and converting it into an execution pulse modulation command through heat flux pulse modulation. This invention integrates multimodal features to achieve a collaborative control mechanism of dynamic energy compensation and heat flux pulse modulation during the nano-copper ink sintering process.

Description

Technical Field

[0001] This invention relates to the field of printed circuit manufacturing control, and in particular to a method and system for dynamic compensation control of temperature field during the sintering process of nano-copper ink on PCB boards. Background Technology

[0002] As PCB manufacturing processes rapidly shift from traditional subtractive etching to additive manufacturing based on nano-ink printing, the effective fusion and stable film formation of nano-conductive copper ink has become a critical factor in improving printed circuit performance. Ensuring that nano-copper particles fully overcome the energy barrier of physical state evolution to construct a continuous conductive film, while simultaneously identifying and responding in real-time to transient differences in local thermal fields caused by the complex morphology of the printed circuit, and effectively compensating for issues such as insufficient energy matching, limited dynamic feedback capabilities, and lag in manual control in existing technologies, is a crucial technical challenge that urgently needs to be addressed in achieving stable sintering below 200℃ and transitioning from single-point process verification to system-level controllable operation.

[0003] Chinese patent application CN111459210A discloses a method for controlling the temperature of PCB reflow soldering, comprising the following steps: S1, providing a soldering furnace with heat sources distributed on its inner wall, and providing a temperature profiler to measure the temperature change of point P0 at any time t over time; S2, providing a temperature profiler with a strip-shaped sensor array to sample along the path P0 at multiple times, obtaining the temperature curve of any position P on the path P0 within the reflow furnace at these times; S3, substituting the target temperature T0 into the above formula to obtain the value of P at the intersection of the curve and the target temperature T0, thus obtaining a constant value; S4, using the isothermal surface where the target temperature T0 is located within the soldering furnace as a boundary, calculating the absolute value of the heat on both sides of the boundary within the furnace cavity; S5, recording the inflection point time tg at each inflection point, and rapidly reducing the temperature of the heat source to the target temperature after reaching the inflection point time tg. This method can accurately control the temperature within the soldering furnace with minimal fluctuations, protecting the reliability of the PCB after soldering.

[0004] However, current technology still faces many challenges. In the additive sintering process of nano-conductive copper ink, traditional fixed-power scanning methods struggle to effectively identify the energy absorption characteristics of nanoparticles during their physical phase transition. When the heat source scans to specific areas with thick ink layers or dense wiring, the local temperature field often exhibits a significantly slower rate of temperature rise because nanoparticles need to absorb substantial latent heat energy to complete the phase transition. If the control system fails to detect this energy deficiency and adjust the power accordingly, the nano-copper particles will not fuse sufficiently, making it difficult to form a continuous and stable conductive film, thus causing circuit conductivity failure or connectivity defects. In such cases, simply increasing the heat source output intensity to ensure sintering depth can easily lead to short-term heat accumulation in areas with limited heat dissipation capacity of the substrate, causing the local temperature to exceed the substrate's heat resistance limit of 200°C. This can result in irreversible carbonization, thermal deformation, or even structural damage to the PCB substrate, directly causing overall PCB failure and reducing the consistency of finished products and the stability of system operation in the precision electronic additive manufacturing process. Summary of the Invention

[0005] To achieve the above objectives, this invention provides a method for dynamic compensation and control of the temperature field during the sintering process of nano-copper ink on PCB boards. The specific technical solution is as follows:

[0006] The collected three-dimensional contour information of the nano-copper ink layer is meshed to divide it into several mesh units. Combined with the original PCB board design file, the heat capacity load distribution matrix is ​​obtained, and the heat capacity load distribution matrix is ​​mapped to the initial energy field matrix.

[0007] The original multispectral thermal radiation signal of each grid unit during the sintering process is collected, and the real temperature field distribution is obtained through colorimetric thermometry logic. The temperature field gradient vector is generated based on the temperature rise difference between adjacent grids in the real temperature field distribution.

[0008] The real-time thermal response rate of each grid cell is captured in real time. Combined with the initial energy field matrix, the latent heat energy value of phase change generated due to fusion endothermic heat is identified, and a phase change energy compensation matrix is ​​generated based on the heat capacity load distribution matrix.

[0009] The execution power command is generated based on the temperature field gradient vector, the initial energy field matrix, the phase change energy compensation matrix, and the real-time thermal response rate. The execution power command is then converted into an execution pulse modulation command via the heat flow pulse modulation conversion logic.

[0010] Furthermore, the method for generating the initial matrix of the energy field includes:

[0011] The three-dimensional contour information of the nano-copper ink layer is collected, and the nano-copper ink layer is divided into several grid units based on the three-dimensional contour information. The average thickness feature value and morphological fluctuation deviation factor are calculated based on the measured height value of the sampling points in the grid unit, and the layer thickness distribution feature matrix is ​​encapsulated and constructed.

[0012] Based on the line vector topology information in the original PCB design file, the local line coverage of each grid cell is determined, and coupled calculation is performed with the layer thickness distribution feature matrix to obtain the heat capacity load distribution matrix;

[0013] A physical thermal energy-power saturation mapping relationship is established, and the thermal load factor in the thermal load distribution matrix is ​​mapped to the energy injection density of each grid cell to generate the initial energy field matrix.

[0014] Furthermore, the method for calculating the average thickness eigenvalue and the initial matrix of the energy field for the morphological undulation deviation factor includes:

[0015] Extract the measured height values ​​of each sampling point within the grid cell, calculate the sum of the measured height values ​​of all sampling points and the ratio of the total number of sampling points, and obtain the average thickness feature value corresponding to the grid cell;

[0016] Calculate the sum of squares of the differences between the measured height value and the average thickness characteristic value of each sampling point within the grid cell, divide the sum of squares by the total number of sampling points and perform a squaring operation to obtain the discrete mean square error corresponding to the grid cell;

[0017] Using the average thickness characteristic value as a scaling reference, the ratio of the discrete root mean square error to the average thickness characteristic value is calculated to obtain the morphological fluctuation deviation factor.

[0018] Furthermore, the method for generating the temperature field gradient vector includes:

[0019] The original multispectral thermal radiation signals of each grid unit during the sintering process are collected using a dual-band infrared detector. The emissivity interference in the original multispectral thermal radiation signals is eliminated by colorimetric temperature measurement logic in order to obtain the true temperature field distribution.

[0020] The real-time temperature values ​​of each grid cell and its surrounding adjacent grid cells in the real temperature field distribution are extracted. The local thermal evolution gradient values ​​of each grid cell are calculated through first-order difference operation in discrete space to generate the temperature field gradient vector.

[0021] Furthermore, the step of removing emissivity interference from the original multispectral thermal radiation signal includes:

[0022] Extract the first spectral radiance and the second spectral radiance corresponding to the original multispectral thermal radiation signal at the first detection wavelength and the second detection wavelength;

[0023] The temperature-scale numerator is obtained by multiplying Planck's second constant by the product of the difference between the reciprocals of the first and second detection wavelengths.

[0024] Calculate the natural logarithm of the ratio of the first spectral radiance to the second spectral radiance, and sum five terms based on the natural logarithm of the ratio of the first detection wavelength to the second detection wavelength to obtain the logarithmic denominator.

[0025] Divide the numerator of the temperature scale by the denominator of the logarithm to calculate the real-time temperature value of each grid cell after removing surface emissivity interference. The real temperature field distribution is then constructed from the real-time temperature values ​​of all grid cells.

[0026] Furthermore, the method for calculating the local thermal evolution gradient value of each grid cell includes:

[0027] Obtain the real-time temperature values ​​of the grid cells and their adjacent grid cells in the horizontal and vertical reference directions of the PCB board;

[0028] Based on the real-time temperature values ​​of the current grid cell in the positive and negative directions adjacent to the horizontal reference, the horizontal spatial temperature difference component is calculated; the horizontal spatial temperature difference component is divided by twice the horizontal reference step size to obtain the horizontal partial derivative; the positive and negative adjacent grid cells in the horizontal reference are respectively the positions offset by one unit step length along the horizontal axis in the positive or negative direction;

[0029] Based on the real-time temperature values ​​of the current grid cell in the positive and negative directions of the vertical reference direction, the vertical spatial temperature difference component is calculated; the vertical spatial temperature difference component is divided by twice the vertical reference direction step size to obtain the vertical partial derivative; the positive and negative adjacent grid cells in the vertical reference direction are respectively the positions offset by one unit step length along the vertical axis in the positive or negative direction.

[0030] The square root operation is performed on the sum of the squares of the horizontal and vertical partial derivatives to calculate the local thermal evolution gradient value of each grid cell.

[0031] Furthermore, the method for generating the phase transition energy compensation matrix includes:

[0032] The temperature change rate of each grid cell along the time axis in the real temperature field distribution is calculated to obtain the real-time thermal response rate and identify the critical triggering time for the fusion state transition of the nano-copper ink conductive medium.

[0033] The latent heat absorption intensity is identified based on the initial energy field matrix and real-time thermal response rate, and dynamic calibration is performed in combination with the preset substrate kinetic energy dissipation coefficient to obtain the phase change latent heat energy value of each grid cell.

[0034] The transient compensation components of each grid cell are calculated based on the latent heat energy value of phase change and the heat capacity load distribution matrix, and then encapsulated to generate a phase change energy compensation matrix.

[0035] Furthermore, the method for generating the pulse modulation instruction includes:

[0036] Extract the local thermal evolution gradient values ​​of each grid cell in the temperature field gradient vector, and construct negative feedback suppression logic for the heat flow distribution to generate a heat loss suppression matrix.

[0037] The initial energy field matrix and the phase change energy compensation matrix are superimposed in the same position to establish the heat flux benchmark to be executed. The initial value of the heat energy injection intensity is obtained by combining the heat loss suppression matrix. The dynamic weight adjustment is performed according to the real-time heat response rate to obtain the real-time correction amount. The real-time correction amount is subtracted from the initial value of the heat energy injection intensity to generate the local power command, and then encapsulated as the execution power command.

[0038] The heat flux pulse modulation conversion logic is executed. When the real-time temperature value is detected to reach the physical heat resistance threshold of 200℃, the power command is mapped to the pulse modulation command corresponding to the heat source generation component of the corresponding grid cell.

[0039] Furthermore, the method for constructing the negative feedback suppression logic includes: using the natural constant Using the base as the base, the product of the preset substrate thermal sensitivity constant and the local thermal temperature evolution gradient value is divided by the negative value of the preset critical heat flux threshold as the exponent to construct an exponential decay function, and the local heat loss suppression factor of each grid cell is calculated and generated; by traversing the grid cells and encapsulating the spatial features of each local heat loss suppression factor, a heat loss suppression matrix is ​​constructed.

[0040] A dynamic temperature field compensation control system for the sintering process of nano-copper ink on PCB boards is used to implement the above-mentioned dynamic temperature field compensation control method for the sintering process of nano-copper ink on PCB boards. The system includes an initial energy field module, a temperature field sensing module, a latent heat identification module, and a collaborative control module.

[0041] The initial energy field module is used to perform meshing processing on the collected three-dimensional contour information of the nano-copper ink layer to divide it into several mesh units, and combine it with the original design file of the PCB board to obtain the heat capacity load distribution matrix, and map the heat capacity load distribution matrix into the initial energy field matrix.

[0042] The temperature field sensing module is used to collect the original multispectral thermal radiation signals of each grid unit during the sintering process, obtain the real temperature field distribution through colorimetric thermometry logic, and generate a temperature field gradient vector based on the temperature rise difference between adjacent grids in the real temperature field distribution.

[0043] The latent heat identification module is used to capture the real-time thermal response rate of each grid cell, identify the latent heat energy value of the phase change due to fusion endothermic heat generation in combination with the initial energy field matrix, and encapsulate and generate a phase change energy compensation matrix according to the heat capacity load distribution matrix.

[0044] The collaborative control module is used to generate an execution power command based on the temperature field gradient vector, the initial energy field matrix, the phase change energy compensation matrix, and the real-time thermal response rate, and to convert the execution power command into an execution pulse modulation command through the heat flow pulse modulation conversion logic.

[0045] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0046] This invention establishes a physical thermal energy-power saturation mapping logic with thermal oversaturation protection characteristics by matching and fusing the measured three-dimensional contour features of the nano-copper ink layer with the vector topology information of the printed circuit in the spatiotemporal dimension. This enables adaptive matching between the energy supply intensity and the local non-uniform heat capacity requirements, eliminates the initial thermal field imbalance caused by ink layer thickness fluctuations or wiring density, and constructs a physical-level active defense barrier against the 200°C substrate damage red line from the source.

[0047] This invention eliminates the interference of dynamic surface emissivity drift caused by the metallization phase transition of nano-copper ink by capturing dual-band multispectral thermal radiation signals and executing colorimetric thermometry logic. It achieves orthogonal decoupling between the true temperature field intensity and surface optical noise. Furthermore, by quantifying the spatial temperature difference components between adjacent grids, it establishes the temperature field gradient vector, avoiding misjudgment of energy regulation caused by false temperature rise signals. This provides a spatial evolution criterion for intercepting excessive longitudinal penetration of thermal energy into the substrate.

[0048] This invention identifies the latent heat energy of phase change during the fusion process of nanoparticles by capturing the deviation between the real-time thermal response rate and the thermal equilibrium state of each grid unit, and uses the thermal load distribution matrix to correct the transient energy compensation amplitude, thereby enabling the copper nanoparticles to cross the energy barrier of the physical evolution from discrete particles to continuous conductive film within a preset temperature range. This eliminates the connectivity defects of the conductive network caused by the fusion kinetic lag, and simultaneously intercepts the penetration and accumulation of heat energy into the deep layers of the substrate.

[0049] This invention generates pulse modulation commands with physical boundary confinement adjustment characteristics by weighted coupling mapping of phase change energy increment, heat spillover gradient constraint and medium thermal response inertia. This achieves synchronous synergy between high-energy heat flow supply and heat flow diffusion interception. While assisting nano-copper particles to transiently cross the energy barrier of physical state evolution, it uses the pulse gap to periodically suppress the thermal accumulation intensity of the substrate, thus solving the technical paradox of substrate thermally sensitive damage and local sintering unevenness under complex circuit topologies. Attached Figure Description

[0050] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0051] Figure 1 This is a flowchart illustrating the principle of the dynamic temperature field compensation and control method for the PCB board nano-copper ink sintering process of the present invention.

[0052] Figure 2 This is a schematic diagram illustrating the principle of generating partial circuit coverage based on the original PCB design file according to the present invention;

[0053] Figure 3 This is a schematic diagram illustrating the principle of generating local thermal temperature evolution gradient values ​​for quantifying the intensity of horizontal thermal energy diffusion according to the present invention.

[0054] Figure 4 This is a functional block diagram of the dynamic compensation control system for the temperature field during the sintering process of nano-copper ink on PCB boards according to the present invention. Detailed Implementation

[0055] The technical solutions of 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.

[0056] Example 1:

[0057] Please see Figure 1 As shown, this embodiment provides a method for dynamic compensation and control of the temperature field during the sintering process of nano-copper ink on a PCB board, including:

[0058] Step S1000: Process the collected three-dimensional contour information of the nano-copper ink layer. Perform meshing to divide into several mesh cells Combined with the original PCB design files To determine each grid cell Energy injection density To generate the initial matrix of the energy field .

[0059] Specifically, this step aims to transform the original design files of the PCB board to be sintered. and the three-dimensional contour information of the printed nano-copper ink layer By performing spatiotemporal matching and fusion, and mapping the nonlinear morphological differences of printed circuits to quantifiable differences in heat absorption capacity through physical modeling, the initial thermal field imbalance caused by ink layer thickness fluctuations or wiring density is eliminated at the energy injection source. This provides an initial energy field matrix with physical a priori characteristics for subsequent dynamic compensation. .

[0060] Further, step S1000 includes:

[0061] Step S1100: Acquire the three-dimensional contour information of the nano-copper ink layer. Based on the three-dimensional contour information of the nano-copper ink layer Divided into several grid units Based on grid cells The average thickness characteristic value is calculated from the measured height values ​​of the internal sampling points. and morphological fluctuation deviation factor The layer thickness distribution feature matrix is ​​obtained by encapsulation and construction. .

[0062] Specifically, this step aims to extract the three-dimensional contour information of the nano-copper ink layer with microscopic undulations formed after the printing process. As a physical sensing object, the longitudinal deposition depth of the ink layer surface is mapped to the average thickness feature value within a discrete grid by utilizing the mapping relationship between geometric shape and thermophysical response. Furthermore, the microscopic peak and valley characteristics that characterize surface roughness and thermal stress concentration potential are mapped into a layer thickness distribution characteristic matrix. Morphological fluctuation deviation factor Digital decoupling of morphological geometric features and microscopic topological perturbation factors is achieved at the source of perception.

[0063] In the specific implementation process, during the static benchmark extraction stage before the sintering process starts, a non-contact laser triangulation sensor is used to perform a full-path scan of the nano-copper ink layer on the surface of the PCB board to be sintered, thereby obtaining the three-dimensional contour information of the nano-copper ink layer that characterizes the spatial dimensional distribution of the ink layer. Subsequently, the three-dimensional contour information of the nano-copper ink layer was... The coordinates are aligned coaxially with the physical topology coordinate system of the PCB board surface, and discretized along the horizontal and vertical axes of the PCB board surface, dividing it into several preset sizes. Grid cells .in, This indicates the step length of the grid cell along the horizontal reference direction of the PCB board, i.e., the step size in the horizontal reference direction; This indicates the step width of the grid cell along the vertical reference direction of the PCB board, i.e., the step size in the vertical reference direction; and These represent the row and column indices of the grid cells in the digital spatial matrix, respectively, and are used to lock the physical spatial coordinates of the area to be sintered on the PCB board surface.

[0064] Based on the accuracy requirements of printed circuits, the three-dimensional contour information of the nano-copper ink layer is... Spatial coordinate data projected onto specific grid cells In the middle. For each grid cell. By statistically processing the measured height values ​​of all discrete sampling points collected within the grid area, the average stacked thickness of the nano-copper ink within the grid area, i.e., the average thickness characteristic value, is calculated. This is used to characterize the fundamental thermal potential of the grid region. The average thickness characteristic value within each grid cell... Logically, the result is equal to the sum of the measured height values ​​of all sampling points within the grid divided by the total number of sampling points. The total number of sampling points is determined by the product of the sensor's sampling frequency and scanning speed.

[0065] To further identify the risk of localized overheating failure caused by microscopic unevenness on the surface of the nano-copper ink due to printing process fluctuations, this step simultaneously extracts data from the grid cells. Internal morphological fluctuation deviation factor The morphological fluctuation deviation factor This is used to quantitatively characterize the microscopic topological roughness of the ink layer surface within a single grid cell, which is achieved by calculating the grid cells. The measured height values ​​of each sampling point relative to the average thickness characteristic value The discrete mean square error, and utilize the combined average thickness eigenvalues This is obtained by scaling and normalizing the data as a scaling reference. The morphological fluctuation deviation factor is... In computational logic, it equals the measured height value and the average thickness characteristic value of each sampling point within the grid cell. The square root of the average of the sum of the squares of the differences is then divided by the average thickness characteristic value. The morphological fluctuation deviation factor The physical significance lies in its ability to quantify the microscopic tip density of the surface of nano-copper ink. As this value increases, it indicates that the corresponding mesh unit... The region is more prone to tip thermal effect during sintering, which can induce local temperature rise exceeding the 200°C substrate damage threshold before the macroscopic input power reaches the threshold.

[0066] Finally, by traversing all the grid cells on the PCB board surface... , each grid cell Average thickness characteristic value and morphological fluctuation deviation factor Spatial feature encapsulation is performed to construct a layer thickness distribution feature matrix covering the entire PCB board sintering area. The layer thickness distribution feature matrix As a physical digital twin of the entire circuit board, it can simultaneously reflect the macroscopic heat capacity distribution and the microscopic thermal failure risk distribution, providing global navigation guidance for the energy allocation of subsequent actuators.

[0067] Step S1200, based on the original PCB design file The line vector topology information determines each grid cell. Local line coverage Combined with the layer thickness distribution feature matrix Coupled calculations are performed to obtain the heat capacity load distribution matrix. .

[0068] Specifically, this step aims to extract the original PCB design files. The logical topology information and the layer thickness distribution feature matrix from step S1100 The physical morphology of the structure serves as a carrier for thermodynamic characterization, utilizing nano-copper ink conductive media in each grid unit. The distribution ratio within the area, i.e., the local line coverage rate. For the average thickness characteristic value Perform spatial weighted mapping to divide each grid cell The potential thermal inertia characteristics are mapped to a quantifiable heat capacity load factor. And at the physical level, by introducing a morphological fluctuation deviation factor Nonlinear correction is applied to the energy absorption deviation to enable advance prediction of the thermal response characteristics of areas with different wiring densities before energy injection.

[0069] In the specific implementation process, the layer thickness distribution feature matrix characterizing the actual stacking state of the nano-copper-ink conductive medium is obtained. Then, the original PCB design files are parsed simultaneously. This step extracts the original PCB design files. The line vector topology information is obtained and projected onto the grid cells. In the spatial discrete coordinate system, the printed line pairs in the grid space are analyzed from the line vector topology information. Physical space allocation, calculating each grid cell The actual physical area covered by the printed circuitry and the grid cell The ratio of the total area determines the local line coverage rate. Used for quantizing grid cells Total volumetric heat capacity of the internal nano-copper ink conductive medium.

[0070] Further, please refer to Figure 2 As shown, Figure 2 This is a schematic diagram illustrating the principle of partial circuit overlay generation based on the original PCB design file according to the present invention. Figure 2 As shown, extract the original PCB design file. The line vector topology information is projected onto the grid cells. In the spatial discrete coordinate system, the red squares represent individual grid cells after spatial discretization. The printed lines are plotted in the grid space by analyzing the line vector topology information. The physical footprint is determined, and each grid cell is calculated. The ratio of the actual physical area covered by the printed circuitry to the total area of ​​the grid cell is used to determine the local circuitry coverage. .

[0071] Subsequently, in order to quantize the grid cells The heat absorption capacity of each locked local area during transient energy injection is fully considered in this step, taking into account the heat absorption capacity of the nano-copper ink conductive medium and the physical constraints of heat dissipation from the substrate to the bottom layer. This is achieved by using the average thickness characteristic value. Local line coverage and morphological fluctuation deviation factor Nonlinear coupling is performed, taking into account the inherent thermal conductivity of the substrate. The reciprocal relationship is used to generate the heat load factor. The inherent thermal conductivity of the substrate is... These are preset physical constants used to characterize the rate at which a PCB substrate, such as FR-4 or polyimide, transfers heat to the non-heated side per unit temperature rise. The heat capacity load factor... Logically, it equals the average thickness characteristic value. and local line coverage The product of the two factors, divided by the initial thermal conductivity of the substrate. And introduce a morphological fluctuation deviation factor. The exponential gain term is corrected. This exponential gain term serves as a heat concentration correction coefficient, characterizing the tip endothermic gain induced by microscopic topological fluctuations.

[0072] By traversing all grid cells of the PCB board The calculated individual grid cells Heat load factor Spatial encapsulation is performed to ultimately generate a heat capacity load distribution matrix covering the entire PCB board area to be sintered. This transforms complex circuit topologies into directly callable thermophysical response benchmarks.

[0073] Step S1300: Establish the physical thermal energy-power saturation mapping relationship and convert the heat capacity load distribution matrix. Heat load factor Mapped to each grid cell Energy injection density To generate the initial matrix of the energy field .

[0074] Specifically, this step aims to transform the heat capacity load distribution matrix, which characterizes the printed circuit topology and measured morphological features from step S1200, into a matrix representing the heat capacity load distribution. As a physical benchmark for energy space allocation, the physical thermal energy-power saturation mapping logic of thermal energy injection and material phase change absorption is used to classify discrete grid cells. Internal heat load factor Mapped to an initial matrix of directly controllable energy field The power weighting signal in the signal enables adaptive matching between energy supply intensity and local heat capacity demand before the sintering energy flow contacts the printed circuit, providing an energy distribution benchmark with substrate overheat protection properties for subsequent steps.

[0075] In the specific implementation process, a heat capacity load distribution matrix characterizing the physical properties of heat absorption of the entire PCB circuit was established. Next, this step enters the stage of generating the initial energy field ratio, establishing a physical thermal-power saturation mapping logic with thermal oversaturation protection characteristics, and configuring each grid cell... Heat load factor This is converted into driving parameters for the actuator to generate the initial matrix of the energy field. The specific execution logic of the physical thermal energy-power saturation mapping logic is as follows:

[0076] The initial matrix of the energy field Each energy injection density in All are determined by the corresponding heat capacity load factor Obtained through logarithmic nonlinear function mapping, the physical thermal energy-power saturation mapping logic is logically equal to a preset static power reference constant. Multiply by grid cells Internal heat load factor The natural logarithm of the sum of the preset zero-value offset correction coefficients. Wherein, the preset static power reference constant... The value of is preset based on the melting point range of the nano-copper ink conductive medium and the travel speed of the conveying device to establish the basic level of the full-field energy injection; the value of the preset zero-value offset correction coefficient is determined based on the background thermal radiation threshold of the PCB board material to ensure that a basic controlled heat flux is maintained in the grid area where the printed circuit coverage approaches zero.

[0077] The physical logic behind using a logarithmic function as the core mapping in this step lies in leveraging the mathematical property that the slope of the logarithmic function decreases as the independent variable increases, thereby achieving saturated control of the energy injection intensity. When the thermal load factor in a certain region increases due to the increased thickness of the nano-copper-ink conductive medium... As the amplitude increases, the mapped energy injection density The growth rate will slow down accordingly. This mapping logic, by establishing a nonlinear growth boundary of energy flux density at the energy injection source, can forcibly constrain the rate of thermal energy accumulation while meeting the energy required for deep crystallization of thick media. Thus, during the dynamic scanning of the sintering heat source, a physical-level active defense barrier against the 200°C substrate damage red line is established.

[0078] Step S2000: Collect data from each grid unit during the sintering process. The original multispectral thermal radiation signal The true temperature field distribution is obtained through colorimetric thermometry logic. Based on the actual temperature field distribution The temperature rise difference between adjacent grids is used to generate the temperature field gradient vector. .

[0079] Specifically, this step aims to extract the original multispectral thermal radiation signal of a nano-copper-ink conductive medium undergoing transient evolution and dynamic drift in radiation characteristics within a physically controlled field. As a real-time monitoring object, the emissivity fluctuations caused by the phase transition metallization process within the current scanning field of view are mapped to the true temperature field distribution. The calibration temperature signal is used to map the microscopic heat flow distribution, which characterizes the tendency of heat energy to penetrate longitudinally into the substrate, into a temperature field gradient vector. The spatial difference signal in the data achieves orthogonal decoupling between the temperature field intensity signal and the thermal diffusion evolution characteristics during the information acquisition stage.

[0080] Further, step S2000 includes:

[0081] Step S2100: Use a dual-band infrared detector to collect data from each grid cell during the sintering process. The original multispectral thermal radiation signal The original multispectral thermal radiation signal is eliminated by colorimetric temperature measurement logic. Emissivity interference in the data to obtain the true temperature field distribution .

[0082] Specifically, this step aims to use the nano-copper-ink conductive medium, which is undergoing transient evolution and dynamically shifting radiation characteristics within the physical sintering field of view, as the detection object. Utilizing the energy conservation differences in radiation energy across different infrared bands, the current grid cell... The dynamic fluctuations in surface emissivity caused by the intrinsic metallization and crystallization process are mapped to the original multispectral thermal radiation signal. The background interference is included, and the net thermal energy intensity after removing the interference is mapped to the true temperature field distribution that reflects the true state of the physical field. At the data acquisition end, orthogonal decoupling of physical thermal information and surface optical interference information is achieved.

[0083] In the specific implementation process, during the dynamic scanning of the sintering area on the PCB board surface by the sintering heat source, this step uses an integrated dual-band infrared detector to simultaneously capture each grid cell. The corresponding original multispectral thermal radiation signal The grid cell Maintaining consistency with physical dimensions and spatial coordinates in step S1100, and strictly adhering to the spatial discrete coordinate system established in step S1200, the instantaneously acquired thermal radiation signal and the heat capacity load distribution matrix characterizing the thermal features of the line morphology generated in the preprocessing stage of step S1000 are integrated. Perform spatiotemporal alignment mapping.

[0084] Considering that the conductive nano-copper ink undergoes a phase transition from discrete particle slurry to a continuous metal film during sintering, the formation of its surface metallic luster can cause dynamic fluctuations in surface emissivity due to the metallization process, resulting in false temperature rise signals in the temperature measurement feedback. Therefore, this step employs a specifically calibrated and improved colorimetric thermometry logic, and performs band-sensitive calibration for the metallization phase transition range of nano-copper ink between 150℃ and 200℃. The colorimetric thermometry logic measures the original multispectral thermal radiation signal... The process involves using two preset narrowband detection wavelengths, namely the first detection wavelength. With the second detection wavelength The single-valued function relationship between the spectral radiance ratio and temperature is used to eliminate interference from unknown and fluctuating surface emissivity in order to generate each mesh cell. Corresponding real-time temperature value Composed of all grid cells Real-time temperature value Together they constitute the true temperature field distribution of the PCB circuit area. Wherein, the first detection wavelength Second detection wavelength All measurements were obtained by pre-detecting the main radiation energy spectrum envelope of nano-copper ink during the phase transition process from 150℃ to 200℃ using a spectrometer. Specifically, the center wavelength corresponding to the frequency band of the intense metallization reaction of nano-copper ink was selected to ensure that the detector can generate a significant voltage gain response to the subtle changes in radiative flux within this temperature range, thereby minimizing the masking of the effective thermal signal by background noise.

[0085] The real-time temperature value In terms of computational logic, based on the principle of dual-band spectral energy partitioning, Planck's second constant is first calculated. The product of the reciprocal differences between the two preset narrowband detection wavelengths yields the temperature-scale numerator term characterizing the energy level distribution; subsequently, the first spectral radiance is calculated. With the second spectral radiance The natural logarithm of the ratio, plus the value based on the first detection wavelength. Second detection wavelength Five times the natural logarithm of the ratio yields the logarithmic denominator, representing the shift in radiation intensity; finally, the numerator of the temperature scale is divided by the logarithmic denominator to obtain the real-time temperature value. Among them, Planck's second constant The value is set as The first spectral radiance With the second spectral radiance respectively at the first detection wavelength With the second detection wavelength The captured net spectral radiance is derived from the original multispectral thermal radiation signal. Real-time sampling.

[0086] Step S2200: Extract the true temperature field distribution. Each grid cell The real-time temperature values ​​of the grid cell and its surrounding adjacent grid cells are used to calculate the temperature of each grid cell through a first-order difference operation in discrete space. Local thermal temperature evolution gradient value To generate temperature field gradient vector .

[0087] Specifically, this step aims to obtain the true temperature field distribution from step S2100, reflecting the transient thermal energy evolution state of the nano-copper graphite conductive medium, from the physically controlled field. As a dynamic monitoring object, the discrete spatial difference algorithm is used to analyze each grid cell. The temperature rise difference between itself and its neighborhood is mapped to a temperature field gradient vector characterizing the intensity of horizontal thermal energy diffusion. By locking the spatial change singularity in the temperature field evolution process, the physical quantification of the risk of local heat accumulation and the trend of heat leakage to the substrate longitudinally can be achieved on a millisecond scale.

[0088] The grid unit Strictly inherit and align the mapping from step S1200 based on the original PCB design file. The established spatial discrete coordinate system ensures that the heat diffusion data acquired during the dynamic scanning process can be correlated with the heat capacity load distribution matrix extracted in the static preprocessing stage of step S1200. Achieve physical-level spatiotemporal alignment mapping.

[0089] In the specific implementation process, the real temperature field distribution of the entire field is obtained digitally. Next, this step analyzes the heat diffusion dynamics within a controlled area on the PCB surface in real time by constructing spatial feature gradient logic. The specific construction logic of the spatial feature gradient logic is as follows: for each grid cell in the spatial discrete coordinate system... By extracting the real-time temperature values ​​of the grid cell and its surrounding adjacent grid cells, a first-order difference operation in the discrete space is performed to generate each grid cell. The corresponding local thermal-temperature evolution gradient value characterizing the intensity of local temperature rise heterogeneity .

[0090] The local thermal evolution gradient value In terms of physical logic, the mesh cells are quantized. The transient heterogeneity intensity of the internal thermal energy flow distribution is calculated using the following logic: the local thermal temperature evolution gradient value. Equal to the grid cell The arithmetic square root of the sum of the squares of the rates of temperature change in the horizontal and vertical directions. This is used to calculate the modulus of the local thermal evolution gradient. First, the difference between the real-time temperature values ​​of adjacent grid cells in the horizontal direction is obtained, i.e., the temperature values ​​of adjacent grid cells in the positive horizontal direction. Temperature values ​​of horizontally negative adjacent grids The difference between the two values ​​is used to obtain the horizontal spatial temperature difference component; the horizontal spatial temperature difference component is then divided by twice the horizontal reference step size. This yields the horizontal partial derivative, which characterizes the tendency for horizontal heat diffusion. Similarly, the difference between the real-time temperature values ​​of adjacent grid cells in the vertical direction is obtained, i.e., the temperature values ​​of adjacent grid cells in the positive vertical direction. Temperature values ​​of adjacent grids in the vertical negative direction The difference between the values ​​is used to obtain the vertical spatial temperature difference component; the vertical spatial temperature difference component is then divided by twice the vertical reference direction step size. This yields the vertical partial derivatives characterizing the tendency for vertical heat diffusion. Finally, the squares of the two axial partial derivatives are summed and a square root is taken to obtain the current mesh element. Local thermal temperature evolution gradient value Wherein, the temperature value of the horizontally adjacent grid in the positive direction. Indicates the current grid cell Real-time temperature value offset by one unit step length along the positive horizontal axis; temperature value of the horizontally adjacent negative grid. Indicates the current grid cell The real-time temperature value offset by one unit step length along the negative horizontal axis; the temperature value of the adjacent grid in the positive vertical direction. Indicates the current grid cell Real-time temperature value offset by one unit step length along the positive vertical axis; temperature value of the adjacent grid in the negative vertical direction. Indicates the current grid cell The real-time temperature value at a point offset by one unit step length along the negative vertical axis.

[0091] By traversing all grid cells in the spatial discrete coordinate system The calculated local thermal evolution gradient values ​​will be used to... Spatial encapsulation is performed to form a temperature field gradient vector describing the heating uniformity of the entire sintering region. .

[0092] Further, please refer to Figure 3 As shown, Figure 3 This is a schematic diagram illustrating the principle of generating local thermal-temperature evolution gradient values ​​for quantifying the intensity of horizontal thermal energy diffusion according to the present invention. Figure 3 As shown, based on the actual temperature field distribution For grid cells The gradient calculation is performed using a discrete spatial coordinate system. The red square within the purple box in the figure represents the central grid cell to be calculated, which is strictly aligned with the discrete spatial coordinate system. Using this central grid cell as a reference, the blue and orange squares represent the horizontally adjacent negative and positive grid cells, respectively. The displacement between them is determined by the horizontal reference direction indicated by the red double arrows, with a step size... Characterization; the yellow and green squares represent the vertically adjacent positive and negative grids, respectively, and the displacement between them is determined by the vertical reference direction step size indicated by the green double arrow. The grid cell is calculated by extracting the real-time temperature values ​​of adjacent grids within the red box and performing a first-order difference operation. The corresponding temperature field gradient vector characterizing the intensity of local thermal evolution and the trend of heat diffusion .

[0093] Step S3000: Real-time capture of each grid cell Real-time thermal response rate Combined with the initial matrix of the energy field Identify the latent heat energy value of phase change due to fusion endothermic reaction. And based on the heat load distribution matrix Encapsulation generates phase transition energy compensation matrix .

[0094] Specifically, this step aims to integrate each grid cell in the sintering field of view. Using nano-copper-ink conductive media undergoing dynamic fusion and exhibiting a non-equilibrium thermal response as an energy analysis carrier, the suppression characteristics of the sensible temperature increment induced by the phase transition endothermic reaction of nano-copper particles are mapped to the phase transition missing energy value. The potential, invisible latent heat absorption load is mapped to the phase change energy compensation matrix of the actuator. This achieves characteristic decoupling of sensible heat rise and latent heat dissipation at the physical evolution level.

[0095] Further, step S3000 includes:

[0096] Step S3100: Calculate the actual temperature field distribution. Each grid cell The rate of temperature change along the time axis yields the real-time thermal response rate. To identify the critical triggering moment when the conductive medium of nano-copper ink undergoes a fusion state transition.

[0097] Specifically, this step aims to monitor the nano-copper-ink conductive medium, which is in a dynamic thermodynamic evolution state during the sintering process, and to obtain the actual temperature field distribution from step S2100. Each grid cell The temperature rise stagnation induced by the internal consumption of latent heat for molecular chain recombination is mapped to the real-time thermal response rate. By capturing the isothermal plateau characteristics of the temperature rise curve within the recrystallization threshold of nanomaterials that deviate from the preset thermal equilibrium state, the critical triggering moment for the transformation of nanoparticles from a spatially discrete distribution to a continuous fused state can be identified at the physical level.

[0098] In the specific implementation process, during the energy injection of the sintering heat source onto the printed circuit, this step tracks the actual temperature field distribution in real time. Each grid cell Corresponding real-time temperature value The rate of temperature change over time is dynamically monitored. When the nano-copper-ink conductive medium is heated to the temperature range corresponding to the recrystallization threshold of the nanomaterial, the fusion between nanoparticles is accompanied by the absorption of a large amount of latent heat of phase change. At the physical level, this process manifests as the injection of thermal energy being converted into the internal energy of the material rather than as an increase in sensible temperature, resulting in a higher real-time thermal response rate. The temperature rise slope deviates from the preset thermal equilibrium state and produces a significant phase transition isothermal plateau.

[0099] Each grid cell Corresponding real-time thermal response rate In terms of computational logic, the first step is to extract the mesh cells. The sensible heat increment component is obtained by dividing the current real-time temperature value and the previous accurate real-time temperature value; then, the sensible heat increment component is divided by the preset sampling period step size. Finally, the characterization of the mesh cell is obtained. Real-time thermal response rate of transient thermal response characteristics Wherein, the sampling period step size The value is preset based on the scanning speed of the sintering heat source and the controlled field thermal relaxation time, used to define the time resolution for thermal response feature extraction; the real-time thermal response rate Indicates the current sampling time in the grid cell The transient rate at which thermal energy is converted into sensible temperature is used to quantify the evolution depth of melting and crystallization of nano-copper ink conductive media.

[0100] By real-time monitoring of the real-time thermal response rate The degree of deviation within the recrystallization threshold of nanomaterials, when the real-time thermal response rate When the value decreases exponentially and approaches zero, the grid cell is considered to be... The critical trigger point for entering the fusion state transition is used as the starting point for initiating the subsequent phase transition energy compensation matrix. Decision-making criteria.

[0101] Step S3200, based on the initial matrix of the energy field and real-time thermal response rate Identify the latent heat absorption intensity and combine it with the preset substrate kinetic energy dissipation coefficient. Perform dynamic calibration to obtain the values ​​of each grid cell. latent heat energy value of phase change .

[0102] Specifically, this step aims to analyze the non-equilibrium endothermic process caused by the melting and crystallization of the nano-copper ink conductive medium in a physically controlled field, using the initial energy field matrix from step S1300 as the analytical object. and the real-time thermal response rate measured in step S3100 By comparing the printed circuit temperature rise energy consumption and the environmental heat dissipation loss due to substrate conduction, the temperature rise response hysteresis state is quantified into the latent heat energy value of phase change. The physical determination of heat energy destination is realized at the data processing end, providing a quantitative energy deficit criterion for determining the evolution depth of the metallization reaction.

[0103] In practice, during the sintering process, this step involves capturing the real-time thermal response rate. The deviation from the theoretical thermal equilibrium state is used to identify the latent heat absorption intensity within the nano-copper-ink conductive medium. This is done for each grid cell. Extract its corresponding real-time thermal response rate and compare it with the initial matrix of the energy field. The corresponding energy injection density Perform a peer comparison.

[0104] Considering the uneven heat dissipation rate caused by differences in printed circuit layout and cross-sectional structure distribution, this step introduces a preset substrate kinetic energy dissipation coefficient. Dynamic calibration of the dynamic energy balance relationship is performed to eliminate temperature rise lag interference caused by non-phase change factors. The preset substrate kinetic energy dissipation coefficient is mentioned above. Its value is obtained by analyzing the heat dissipation effect at the edge of the circuit and the longitudinal thermal conductivity of the FR-4 / polyimide substrate, and is used to dynamically offset the energy identification error caused by environmental heat dissipation.

[0105] Each grid cell Corresponding latent heat energy value of phase change In terms of computational logic, the first step is to calculate the preset energy conversion coefficient and energy injection density. The product of these terms yields the characterization of the mesh cell. The theoretical total heat flux component received at the current moment; and the preset material specific heat characteristic constant and substrate kinetic energy dissipation coefficient are subtracted from this theoretical total heat flux component. The sum of the real-time thermal response rates The product is used to obtain the transient residual thermal energy intensity after removing the apparent temperature rise and heat dissipation, and finally multiplied by the preset transient response gain operator. The latent heat energy value of phase change was obtained. .

[0106] The preset energy conversion coefficient is obtained through experimental calibration of heat source power and surface absorptivity, and is used to map the electronically controlled power value into a physical heat flux component; the preset material specific heat characteristic constant is preset based on the solid content and volumetric specific heat capacity of the nano-copper ink paste, and is used to quantify the intrinsic thermal response capability of the material in a non-phase change state; the preset transient response gain operator... The value is nonlinearly mapped according to different stages of metallization of nano-copper graphite, such as drying, fusion, and densification, to correct the gain of the energy consumption slope at different reaction depths; the latent heat energy value of phase change Used for quantizing grid cells The net heat energy absorbed by the copper nanoparticles during the transition from particulate state to dense conductive film; the larger the value, the more intense the current metallization reaction.

[0107] Step S3300, based on the latent heat energy value of phase change and heat load distribution matrix Calculate each grid cell Transient compensation component And encapsulate and generate a phase transition energy compensation matrix. .

[0108] Specifically, this step aims to map the energy-starved state exhibited during the metallization process of nano-copper ink conductive dielectric due to the unique thermal energy absorption characteristics of nanoparticles during melting into a characterizing grid cell. Transient compensation component of power supply amplitude Using the heat capacity load distribution matrix from step S1200 For grid cells latent heat energy value of phase change at the location After performing weighted corrections based on physical properties, the phase transition energy compensation matrix characterizing the intensity of the full-field energy flux supplementation is calculated and encapsulated. By implementing the directional injection of transient high-energy heat flux, copper nanoparticles are assisted in overcoming the energy barrier of the physical state evolution from discrete particles to continuous conductive film within a preset steady-state heating temperature range below 200℃, thereby eliminating the conductivity network connectivity defects caused by the sluggish fusion dynamics inside the nanoparticles at the physical source.

[0109] In the specific implementation process, after identifying and locking the energy gap required for metallization during the fusion process of copper nanoparticles, this step determines the phase transition energy compensation matrix. The quantitative distribution of energy flow control compensation command established in this step is not the traditional linear power superposition, but a targeted instantaneous energy supply based on the latent heat absorption characteristics of nanoparticle recrystallization. It uses instantaneous high-energy heat flow to make the conductive medium cross the melting energy barrier and then quickly fall back to the preset steady-state heating temperature zone to protect the PCB substrate from deformation damage caused by long-term thermal shock.

[0110] For each grid cell This step calls the heat load distribution matrix. Heat load factor This is used to ensure that thick layers of printed circuits with large cross-sectional dimensions receive penetrating thermal support. The grid unit... Corresponding transient compensation components In terms of computational logic, the preset compensation adjustment factor is first calculated. With heat load factor The product of the two components, plus the preset intensity reference component. The energy regulation slope characterizing the thermal response at that point is obtained; subsequently, the energy regulation slope and the latent heat energy value of the phase change are... Multiplying these yields a dynamic compensation increment characterizing the local state transition; a preset sustaining power bias is then added to this dynamic compensation increment to determine the grid cell. The transient compensation component required at the current moment .

[0111] Wherein, the preset compensation adjustment factor and preset strength reference components The values ​​are all obtained based on the gradient thermal conduction calibration curve of the thermal diffusivity of conductive ink and the cross-sectional dimensions of printed traces, and are used to map the energy consumption gap to a matching power injection slope; the preset maintenance power bias value is based on the thermal saturation threshold of the PCB substrate in the critical temperature range, and is used to ensure the continuity of energy flow output during the switching of transient energy flow control compensation commands, and to prevent local crystallization stress failure caused by transient energy interruption.

[0112] Finally, by traversing all grid cells in the discrete coordinate system of space... All transient compensation components obtained from the calculation Spatial encapsulation is performed to construct a phase transition energy compensation matrix that describes the thermal intensification requirements of the complete sintering region. .

[0113] Step S4000, based on the temperature field gradient vector Initial matrix of energy field Phase transition energy compensation matrix and real-time thermal response rate Generate execution power instructions The power command will be executed via heat flux pulse modulation conversion logic. It is converted into an execution pulse modulation instruction.

[0114] Specifically, this step aims to use the nano-copper-ink conductive medium undergoing the dynamic evolution of sintering as a physical control object, and to control the phase transition energy compensation matrix from step S3300. The transient energy increment required for the phase transformation of the characterized copper nanoparticles is determined by the temperature field gradient vector from step S2200. The risk boundary of the PCB substrate undergoing unexpected temperature rise deformation is characterized by the initial energy field matrix from step S1300. The printed trace preset heat flux density and the real-time thermal response rate from step S3100 are characterized. The sensible heat rise efficiency is nonlinearly decoupled and weighted to the instantaneous response rate of energy injection to generate an execution pulse modulation command that can take into account both the phase change fusion efficiency and the thermal safety boundary of the substrate. This enables the synchronous synergy of high-energy heat flow supply and spatial gradient suppression at the physical execution end, thereby driving the copper nanoparticles to cross the fusion energy barrier and suppressing the penetration of thermal energy into the deep layers of the substrate.

[0115] Further, step S4000 includes:

[0116] Step S4100: Extract the temperature field gradient vector. Each grid cell Local thermal temperature evolution gradient value A negative feedback suppression logic targeting the heat flow distribution is constructed to generate a heat loss suppression matrix. .

[0117] Specifically, this step aims to transform the temperature field gradient vector from step S2200... The quantified potential for heat flux diffusion from the printed circuit lines to the PCB substrate in both the longitudinal and lateral directions is used as a physical constraint. Negative feedback suppression logic is then employed to suppress this potential in each grid cell. Local thermal evolution gradient values ​​of surrounding adjacent grids Mapped to a heat loss suppression matrix used to adjust the energy injection amplitude. By implementing instantaneous suppression at the thermal instability coordinate point, the risk of excessive longitudinal diffusion of heat into the substrate is intercepted when the heat source generating device executes power commands.

[0118] In the specific implementation process, during the dynamic feedback regulation of the heat source generating device, this step calls the temperature field gradient vector. Extract each grid cell in real time The local temperature field evolution characteristics of the surrounding and adjacent grids, i.e., the local thermal temperature evolution gradient values. For each grid cell defined by the spatial discrete coordinate system. This step involves establishing negative feedback suppression logic to address the uneven distribution of heat flow, thereby reducing the local temperature evolution gradient value. Mapped to a local heat loss suppression factor for suppressing local energy flux density. This allows for the construction of a dynamic defense boundary against the risk of thermal damage to the PCB substrate at the physical execution end.

[0119] The specific execution logic of the negative feedback suppression logic is as follows: Each grid cell Corresponding local heat loss suppression factor In computational logic, it is equal to the natural constant. Using the preset substrate thermal sensitivity constant and local thermal evolution gradient value as the base, The product divided by the preset critical heat flux threshold The negative value of is a power function of exponent. When the grid cell Local thermal evolution gradient value Exceeding the critical heat flux threshold When the absolute value of the negative power function of the exponent increases, it causes the grid cell to... Local heat loss inhibition factor It rapidly approaches 0 from 1. Finally, by traversing the entire grid cells... All calculated local heat loss suppression factors Spatial features are encapsulated to construct a thermal loss suppression matrix describing the full-field safety boundary constraints of the controlled region. The preset substrate thermal sensitivity constant is obtained by thermodynamic calibration based on the glass transition temperature and intrinsic thermal diffusivity of the PCB substrate, and is used to define the feedback strength of gradient fluctuations on power suppression; the preset critical heat flux threshold... The value defines the upper limit of the maximum spatial temperature difference that a PCB substrate can withstand without undergoing physical property deformation, and is determined by the physical heat resistance limit of the PCB substrate material.

[0120] Step S4200: Initialize the energy field matrix. Phase transition energy compensation matrix In-situ superposition is used to establish the heat flux baseline to be executed, combined with the heat loss suppression matrix. The initial value of the heat energy input intensity is obtained, based on the real-time thermal response rate. Dynamic weight adjustment is performed to obtain a real-time correction amount, and the real-time correction amount is subtracted from the initial value of thermal energy delivery intensity to generate a local power command. And encapsulate it into an execution power command. .

[0121] Specifically, this step aims to use the nano-copper-ink conductive medium, which is in the critical state of recrystallization evolution of nanomaterials during the sintering process and exhibits significant nonlinear thermal response, as a controlled physical field object. The initial matrix of the energy field characterizing the preset heat flux density of the printed traces from step S1300 is used. The phase transition energy compensation matrix from step S3300 characterizes the compensation demand caused by the endothermic melting of nanoparticles. And the heat loss suppression matrix characterizing the space safety constraint terms from step S4100 Physical correlation and real-time weight allocation are performed using the real-time thermal response rate, which characterizes the thermal conduction inertia of the controlled medium, derived from step S3100. Implement smoothing calibration to address thermal response hysteresis, and generate execution power commands with nonlinear feedforward characteristics in the central control computing unit. This provides a deterministic execution criterion for driving the laser heat source output component to overcome the physical energy barrier of the transition from discrete particles to a continuous conductive film and to avoid thermal damage to the substrate.

[0122] In the specific implementation process, after obtaining the energy compensation requirements generated by the latent heat absorbed during the particle fusion of the nano-copper ink conductive medium and the constraints on the lateral and longitudinal heat diffusion at the PCB substrate level, this step executes the power command. Multidimensional feature synthesis. This step no longer employs linear error adjustment, but instead constructs an energy flow-controlled logic that couples the physical morphology of the line with the progress of its physical property evolution.

[0123] The specific execution logic of the energy flow control logic is as follows: First, extract the initial matrix of the energy field. Energy injection density in Combine it with the phase transition energy compensation matrix Transient compensation components in In-situ superposition is performed to construct a baseline of the heat flux required for the current physical node to complete the fusion of the dense conductive layer. Then, the heat loss suppression matrix is ​​invoked. Local heat loss suppression factor Multiplicative suppression is applied to the heat flux reference to be executed to obtain an initial value of the heat energy injection intensity after calibration against the substrate safety boundary. This dynamically limits the heat injection component that may cause unexpected temperature rise in the non-wiring area of ​​the substrate based on the spatial steepness of the local heat flux diffusion to the substrate.

[0124] Furthermore, to eliminate the transient response delay caused by the conduction characteristics between the temperature measuring device and the controlled medium, and to prevent drastic fluctuations in the thermal field amplitude caused by instantaneous energy flow jumps, this step introduces a preset thermal inertia damping coefficient to affect the real-time thermal response rate. Dynamic weight adjustment is performed to obtain a real-time correction amount characterizing the kinetic energy of the temperature rise in the controlled medium. Then, by subtracting the real-time correction amount from the initial value of the thermal energy input intensity, the values ​​of each grid cell are calculated. Corresponding local power command Finally, traverse all grid cells in the entire field. All grid cells in the field Local power command Spatial encapsulation is performed to ultimately form the execution power command covering the entire sintering area. The preset thermal inertia damping coefficient is obtained by experimentally calibrating the thermal relaxation time of the controlled field to the pulsed energy injection, and is used to quantify and mitigate the hysteresis effect of the temperature field response.

[0125] For example, consider sintering a nano-copper ink trace with uneven width and thickness distribution on a thermosensitive PCB substrate: when the laser heat source output component scans to the pad area with a thicker ink layer, according to the phase transition energy compensation matrix... The nano-copper particles in this region were identified as being in a critical transition phase from solid to molten, requiring additional energy to overcome the latent heat of phase transition. A moderate increase in the instantaneous heat flux output of this region ensures efficient heat transfer to the ink layer, guaranteeing that particles closer to the substrate also participate in sintering, thus forming a dense and continuous metal structure. Simultaneously, if the central control computing unit uses a heat loss suppression matrix... If the intensity of heat conduction from this area towards the PCB substrate is detected to be close to a preset safety threshold, then a local heat loss suppression factor is used. The output power of the heat source is proportionally suppressed to prevent heat from penetrating and accumulating inside the PCB substrate, thus preventing the PCB substrate from overheating. Simultaneously, a real-time thermal response rate is introduced. As a real-time correction, this step anticipates and mitigates potential temperature jumps caused by the physical conduction inertia of the controlled medium, ensuring the stability of the temperature field during energy switching. Specifically, this step dynamically distributes the energy flow output intensity so that when the nano-copper particles obtain the transient energy required to trigger phase evolution, the thermal state of the PCB substrate is simultaneously limited within the physical deformation threshold. This achieves a high-density metallization sintering effect without exceeding the 200℃ heat resistance limit of the PCB substrate, effectively balancing the difficult-to-satisfy process constraints of fully sintering thick ink areas and ensuring the thermal safety of the PCB substrate.

[0126] Step S4300: Execute the heat flux pulse modulation conversion logic when the real-time temperature value is detected. When the physical heat resistance threshold of 200°C is reached, a power command will be executed. Mapped to corresponding grid cells The pulse modulation command corresponding to the heat source generation component at the execution end.

[0127] Specifically, this step aims to treat the nano-copper ink conductive medium during the sintering process as a controlled object, utilizing the difference in thermal response speed between the nano-copper ink conductive medium and the PCB substrate to pulse-modulate the continuous power output of the heat source. The power command executed in step S4200 is then converted using a safety-limited transition function. The mapping is performed using pulse modulation commands with periodic on / off characteristics, causing the heat source to alternate between short-duration high-power operation and intermittent cooling. Simultaneously, real-time temperature values ​​from step S2100 are introduced. As a safety constraint, when monitoring the corresponding grid cell When the temperature change trend is close to the upper limit of the allowable heat resistance of the PCB substrate, the overall heat accumulation intensity of the PCB substrate layer is limited to the preset steady-state heating temperature range of 200℃ by reducing the pulse duty cycle, so as to provide a physical drive output with deterministic safety constraints for the heat source output component.

[0128] In the specific implementation process, upon receiving the power execution command... Next, this step targets the mesh cells. In the region where the energy to be executed is close to the thermal deformation threshold of the PCB substrate, the static continuous energy flow output is converted into a pulse flow with periodic on and off by activating the thermal pulse modulation conversion logic, thereby performing millisecond-level time-division adjustment control of the energy input and implementing unbalanced energy delivery.

[0129] The specific execution logic of the heat flux pulse modulation conversion is as follows: Real-time monitoring of each grid cell. Real-time temperature value at the location And use it as a dynamic variable of the preset safety limit transformation function. When the real-time temperature value is monitored... When the physical heat resistance threshold of 200°C is reached, a power command will be executed. Mapped to corresponding grid cells The execution end heat source generating component corresponds to the pulse modulation command. The mapping logic adjusts the effective work duration of the heat source within a single pulse modulation cycle by calculating the ratio of the heat source on-time to the total cycle duration within a single pulse modulation cycle, i.e., the pulse duty cycle, thereby achieving physical boundary constraint adjustment of the injected total heat flux. The core logic of the preset safety constraint conversion function lies in the fact that the response rate of the nano-copper ink conductive medium to thermal energy is much faster than the thermal conduction and diffusion process of the PCB substrate. By reasonably setting the pulse duration and pulse interval, the nanoparticles complete the instantaneous energy density required for melting and recrystallization during the pulse peak stage, while simultaneously providing the necessary heat dissipation time for the PCB substrate layer during the pulse gap stage to prevent continuous heat accumulation.

[0130] Finally, by traversing all grid cells in the spatial discrete coordinate system... The corresponding pulse modulation command is generated, and after being uniformly packaged, it is sent to the heat source driver hardware. The driver execution end heat source generating component outputs a pulse modulation energy field with peak power maintenance and macroscopic temperature rise suppression characteristics to complete the controlled sintering process.

[0131] Example 2:

[0132] This embodiment, based on Embodiment 1, provides a dynamic temperature field compensation and control system for the sintering process of nano-copper ink on a PCB board, such as... Figure 4As shown, the system includes an initial energy field module, a temperature field sensing module, a latent heat identification module, and a collaborative control module;

[0133] The initial energy field module is used to process the acquired three-dimensional contour information of the nano-copper ink layer. Perform meshing to divide into several mesh cells Combined with the original PCB design files Obtain the heat load distribution matrix The heat load distribution matrix Mapped to the initial matrix of the energy field .

[0134] The temperature field sensing module is used to collect data from each grid unit during the sintering process. The original multispectral thermal radiation signal The true temperature field distribution is obtained through colorimetric thermometry logic. Based on the actual temperature field distribution The temperature rise difference between adjacent grids is used to generate the temperature field gradient vector. .

[0135] The latent heat identification module is used to capture the real-time heat of each grid cell. Real-time thermal response rate Combined with the initial matrix of the energy field Identify the latent heat energy value of phase change due to fusion endothermic reaction. And based on the heat load distribution matrix Encapsulation generates phase transition energy compensation matrix .

[0136] The collaborative control module is used to determine the temperature field gradient vector. Initial matrix of energy field Phase transition energy compensation matrix and real-time thermal response rate Generate execution power instructions The power command will be executed via heat flux pulse modulation conversion logic. It is converted into an execution pulse modulation instruction.

[0137] The parts of the technical solutions provided in the embodiments of this application that are consistent with the implementation principles of corresponding technical solutions in the prior art have not been described in detail to avoid excessive elaboration.

[0138] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the invention. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for dynamic compensation and control of the temperature field during the sintering process of nano-copper ink on PCB boards, characterized in that, include: The collected three-dimensional contour information of the nano-copper ink layer is meshed to divide it into several mesh units. Combined with the original PCB board design file, the heat capacity load distribution matrix is ​​obtained, and the heat capacity load distribution matrix is ​​mapped to the initial energy field matrix. The original multispectral thermal radiation signal of each grid unit during the sintering process is collected, and the real temperature field distribution is obtained through colorimetric thermometry logic. The temperature field gradient vector is generated based on the temperature rise difference between adjacent grids in the real temperature field distribution. The real-time thermal response rate of each grid cell is captured in real time. Combined with the initial energy field matrix, the latent heat energy value of phase change generated due to fusion endothermic heat is identified, and a phase change energy compensation matrix is ​​generated based on the heat capacity load distribution matrix. The execution power command is generated based on the temperature field gradient vector, the initial energy field matrix, the phase change energy compensation matrix, and the real-time thermal response rate. The execution power command is then converted into an execution pulse modulation command via the heat flow pulse modulation conversion logic.

2. The method for dynamic compensation and control of temperature field during the sintering process of nano-copper ink on a PCB board according to claim 1, characterized in that, The method for generating the initial matrix of the energy field includes: The three-dimensional contour information of the nano-copper ink layer is collected, and the nano-copper ink layer is divided into several grid units based on the three-dimensional contour information. The average thickness feature value and morphological fluctuation deviation factor are calculated based on the measured height value of the sampling points in the grid unit, and the layer thickness distribution feature matrix is ​​encapsulated and constructed. Based on the line vector topology information in the original PCB design file, the local line coverage of each grid cell is determined, and coupled calculation is performed with the layer thickness distribution feature matrix to obtain the heat capacity load distribution matrix; A physical thermal energy-power saturation mapping relationship is established, and the thermal load factor in the thermal load distribution matrix is ​​mapped to the energy injection density of each grid cell to generate the initial energy field matrix.

3. The method for dynamic compensation and control of temperature field during the sintering process of nano-copper ink on a PCB board according to claim 2, characterized in that, The method for calculating the average thickness eigenvalues ​​and the initial matrix of the energy field for the morphological undulation deviation factor includes: Extract the measured height values ​​of each sampling point within the grid cell, calculate the sum of the measured height values ​​of all sampling points and the ratio of the total number of sampling points, and obtain the average thickness feature value corresponding to the grid cell; Calculate the sum of squares of the differences between the measured height value and the average thickness characteristic value of each sampling point within the grid cell, divide the sum of squares by the total number of sampling points and perform a squaring operation to obtain the discrete mean square error corresponding to the grid cell; Using the average thickness characteristic value as a scaling reference, the ratio of the discrete root mean square error to the average thickness characteristic value is calculated to obtain the morphological fluctuation deviation factor.

4. The method for dynamic temperature field compensation and control during the sintering process of nano-copper ink on a PCB board according to claim 1, characterized in that, The method for generating the temperature field gradient vector includes: The original multispectral thermal radiation signals of each grid unit during the sintering process are collected using a dual-band infrared detector. The emissivity interference in the original multispectral thermal radiation signals is eliminated by colorimetric temperature measurement logic in order to obtain the true temperature field distribution. The real-time temperature values ​​of each grid cell and its surrounding adjacent grid cells in the real temperature field distribution are extracted. The local thermal evolution gradient values ​​of each grid cell are calculated through first-order difference operation in discrete space to generate the temperature field gradient vector.

5. The method for dynamic temperature field compensation and control during the sintering process of nano-copper ink on a PCB board according to claim 4, characterized in that, The step of removing emissivity interference from the original multispectral thermal radiation signal includes: Extract the first spectral radiance and the second spectral radiance corresponding to the original multispectral thermal radiation signal at the first detection wavelength and the second detection wavelength; The temperature-scale numerator is obtained by multiplying Planck's second constant by the product of the difference between the reciprocals of the first and second detection wavelengths. Calculate the natural logarithm of the ratio of the first spectral radiance to the second spectral radiance, and sum five terms based on the natural logarithm of the ratio of the first detection wavelength to the second detection wavelength to obtain the logarithmic denominator. Divide the numerator of the temperature scale by the denominator of the logarithm to calculate the real-time temperature value of each grid cell after removing surface emissivity interference. The real temperature field distribution is then constructed from the real-time temperature values ​​of all grid cells.

6. The method for dynamic compensation and control of temperature field during the sintering process of nano-copper ink on a PCB board according to claim 4, characterized in that, The calculation method for the local thermal evolution gradient value of each grid cell includes: Obtain the real-time temperature values ​​of the grid cells and their adjacent grid cells in the horizontal and vertical reference directions of the PCB board; Based on the real-time temperature values ​​of the current grid cell in the positive and negative directions adjacent to the horizontal reference, the horizontal spatial temperature difference component is calculated; the horizontal spatial temperature difference component is divided by twice the horizontal reference step size to obtain the horizontal partial derivative; the positive and negative adjacent grid cells in the horizontal reference are respectively the positions offset by one unit step length along the horizontal axis in the positive or negative direction; Based on the real-time temperature values ​​of the current grid cell in the positive and negative directions of the vertical reference direction, the vertical spatial temperature difference component is calculated; the vertical spatial temperature difference component is divided by twice the vertical reference direction step size to obtain the vertical partial derivative; the positive and negative adjacent grid cells in the vertical reference direction are respectively the positions offset by one unit step length along the vertical axis in the positive or negative direction. The square root operation is performed on the sum of the squares of the horizontal and vertical partial derivatives to calculate the local thermal evolution gradient value of each grid cell.

7. The method for dynamic temperature field compensation and control during the sintering process of nano-copper ink on a PCB board according to claim 1, characterized in that, The method for generating the phase transition energy compensation matrix includes: The temperature change rate of each grid cell along the time axis in the real temperature field distribution is calculated to obtain the real-time thermal response rate and identify the critical triggering time for the fusion state transition of the nano-copper ink conductive medium. The latent heat absorption intensity is identified based on the initial energy field matrix and real-time thermal response rate, and dynamic calibration is performed in combination with the preset substrate kinetic energy dissipation coefficient to obtain the phase change latent heat energy value of each grid cell. The transient compensation components of each grid cell are calculated based on the latent heat energy value of phase change and the heat capacity load distribution matrix, and then encapsulated to generate a phase change energy compensation matrix.

8. The method for dynamic compensation and control of temperature field during the sintering process of nano-copper ink on a PCB board according to claim 1, characterized in that, The method for generating the pulse modulation instruction includes: Extract the local thermal evolution gradient values ​​of each grid cell in the temperature field gradient vector, and construct negative feedback suppression logic for the heat flow distribution to generate a heat loss suppression matrix. The initial energy field matrix and the phase change energy compensation matrix are superimposed in the same position to establish the heat flux benchmark to be executed. The initial value of the heat energy injection intensity is obtained by combining the heat loss suppression matrix. The dynamic weight adjustment is performed according to the real-time heat response rate to obtain the real-time correction amount. The real-time correction amount is subtracted from the initial value of the heat energy injection intensity to generate the local power command, and then encapsulated as the execution power command. The heat flux pulse modulation conversion logic is executed. When the real-time temperature value is detected to reach the physical heat resistance threshold of 200℃, the power command is mapped to the pulse modulation command corresponding to the heat source generation component of the corresponding grid cell.

9. The method for dynamic compensation and control of temperature field during the sintering process of nano-copper ink on a PCB board according to claim 8, characterized in that, The method for constructing the negative feedback suppression logic includes: using the natural constant Using the base as the base, the product of the preset substrate thermal sensitivity constant and the local thermal temperature evolution gradient value is divided by the negative value of the preset critical heat flux threshold as the exponent to construct an exponential decay function, and the local heat loss suppression factor of each grid cell is calculated and generated; by traversing the grid cells and encapsulating the spatial features of each local heat loss suppression factor, a heat loss suppression matrix is ​​constructed.

10. A dynamic temperature field compensation control system for the sintering process of nano-copper ink on a PCB board, used to implement the dynamic temperature field compensation control method for the sintering process of nano-copper ink on a PCB board as described in any one of claims 1-9, characterized in that, The system includes an initial energy field module, a temperature field sensing module, a latent heat identification module, and a collaborative control module. The initial energy field module is used to perform meshing processing on the collected three-dimensional contour information of the nano-copper ink layer to divide it into several mesh units, and combine it with the original design file of the PCB board to obtain the heat capacity load distribution matrix, and map the heat capacity load distribution matrix into the initial energy field matrix. The temperature field sensing module is used to collect the original multispectral thermal radiation signals of each grid unit during the sintering process, obtain the real temperature field distribution through colorimetric thermometry logic, and generate a temperature field gradient vector based on the temperature rise difference between adjacent grids in the real temperature field distribution. The latent heat identification module is used to capture the real-time thermal response rate of each grid cell, identify the latent heat energy value of the phase change due to fusion endothermic heat generation in combination with the initial energy field matrix, and encapsulate and generate a phase change energy compensation matrix according to the heat capacity load distribution matrix. The collaborative control module is used to generate an execution power command based on the temperature field gradient vector, the initial energy field matrix, the phase change energy compensation matrix, and the real-time thermal response rate, and to convert the execution power command into an execution pulse modulation command through the heat flow pulse modulation conversion logic.

Images