Aluminum-based copper-clad plate laminating forming method and related device

Abstract

The invention discloses an aluminum-based copper-clad plate laminating forming method and a related device. The method comprises the following steps: when an aluminum-based copper-clad plate enters a hot-pressing area, synchronously applying broadband acoustic excitation and dual-band thermal excitation; acoustic propagation response and thermal diffusion response of the aluminum-based copper-clad plate in the advancing process of the hot-pressing area are collected, and an acoustic-thermal synchronous response sequence is obtained; performing differential transformation and time-frequency domain analysis on the sound-heat synchronous response sequence and the defect-free reference spectrum, and extracting a multi-frequency sound velocity drift rate, a sound attenuation vector, a dual-band thermal attenuation slope and a thermal hysteresis ratio to obtain a sound-heat coupling feature set; and based on a correlation mode of acoustic features and thermal features in the acoustic-thermal coupling feature set, performing defect identification and grading evaluation in combination with a dynamically adjusted process tolerance threshold to obtain a defect grade map of spatial distribution. According to the scheme, hidden defects such as interface cavities, thickness standing waves and filler settlement can be detected in real time in the film covering forming process, extra detection time is avoided, and the production efficiency and the product quality consistency are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of film detection of aluminum-based copper-clad laminates, and in particular to a film forming method for aluminum-based copper-clad laminates and related devices. Background Art

[0002] Aluminum Base Copper-Clad Laminate (CCL) is a multi-layer composite substrate with aluminum plate as the metal heat dissipation base, an insulating dielectric layer laid on top, and copper conductors. Compared with traditional glass fiber epoxy systems, aluminum-based CCL leverages the high thermal conductivity and excellent mechanical strength of aluminum to quickly draw out the heat generated by power devices or high-brightness LEDs and disperse it to the heat sink, while maintaining the electrical integrity and dimensional stability of the circuit layer. With the increasing demand for high power density and long-term reliability in fields such as lighting, power supply, and high-speed communications, the industry typically covers the copper surface with a functional film (such as white solder mask, optical reflective film, or chemical-resistant protective film) to achieve a balance between optical efficiency, insulation strength, and environmental tolerance.

[0003] Current aluminum-based copper-clad laminate (CCCL) lamination production mostly utilizes a roll-to-roll or sheet-to-sheet hot pressing / UV curing process. In this process, after roughening and plasma activation, the copper surface is bonded to a liquid or prepreg film containing ceramic fillers, which undergoes a series of degassing, pressurization, heating, curing, and cooling steps within tens of seconds. Due to the high temperature and pressure conditions and multi-material interface reactions during the lamination process, defects such as interface voids, uneven filler distribution, and thickness standing waves can easily occur, impacting product performance. Therefore, lamination quality testing is essential. Currently, manufacturers generally perform offline testing after lamination is completed, including visual AOI inspection, dielectric withstand voltage testing, and cross-section analysis. This post-lamination testing method requires additional dedicated testing time after the lamination process, significantly reducing production efficiency. On high-speed production lines, each additional offline testing step directly impacts overall throughput. Especially in high-volume production scenarios, this additional testing time can cumulatively become a critical factor impacting production cycle time. Summary of the Invention

[0004] The main purpose of the present invention is to solve the technical problem that the quality inspection of the existing aluminum-based copper-clad laminate film molding requires additional special inspection time after the completion of the film coating process.

[0005] A first aspect of the present invention provides a method for laminating and forming an aluminum-based copper-clad laminate, the method comprising: Broadband acoustic excitation and dual-band thermal excitation are applied simultaneously when the aluminum-based copper-clad laminate enters the hot pressing zone; Acquiring an acoustic propagation response and a thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone to obtain an acoustic-thermal synchronous response sequence; Performing differential transformation and time-frequency domain analysis on the acoustic-thermal synchronous response sequence and the defect-free reference spectrum, extracting the multi-frequency sound velocity drift rate, acoustic attenuation vector, dual-band thermal attenuation slope, and thermal hysteresis ratio, and obtaining an acoustic-thermal coupling feature set; Based on the correlation pattern of acoustic features and thermal features in the acoustic-thermal coupling feature set, defect identification and grading evaluation are performed in combination with a dynamically adjusted process tolerance threshold to obtain a spatially distributed defect grade map.

[0006] Preferably, the simultaneous application of broadband acoustic excitation and dual-band thermal excitation when the aluminum-based copper-clad laminate enters the hot pressing zone includes: According to the acoustic impedance ratio of the aluminum substrate to the copper foil in the aluminum-based copper-clad laminate, a 2-8 MHz linear frequency modulation acoustic pulse is divided into a 2-4 MHz low frequency band and a 6-8 MHz high frequency band, the low frequency band is amplitude-enhanced to detect thickness standing wave defects, and the high frequency band is amplitude-enhanced to detect interface void defects, thereby obtaining a defect-selective broadband acoustic excitation signal; According to the thickness of the insulating dielectric layer and the filler content in the aluminum-based copper-clad laminate, the 3-5 μm band infrared pulse is pulse-width modulated to detect the shallow filler distribution, and the 8-12 μm band infrared pulse is intensity modulated to detect the deep filler distribution, thereby obtaining a dual-band thermal excitation signal for layered detection; The defect-selective broadband acoustic excitation signal and the layered detection dual-band thermal excitation signal are applied synchronously in the resin gelation window to ensure that the excitation propagation path is consistent along the acoustic and thermal conduction channel between the copper surface and the aluminum surface.

[0007] Preferably, according to the acoustic impedance ratio of the aluminum substrate to the copper foil in the aluminum-based copper clad laminate, the 2-8 MHz linear frequency modulation acoustic pulse is divided into a 2-4 MHz low frequency band and a 6-8 MHz high frequency band, the low frequency band is amplitude enhanced to detect thickness standing wave defects, and the high frequency band is amplitude enhanced to detect interface void defects, to obtain a defect-selective broadband acoustic excitation signal, including: According to the thickness specification parameters and material model of the aluminum-based copper-clad laminate, the corresponding aluminum substrate acoustic impedance value and copper foil acoustic impedance value are obtained from a preset acoustic impedance database, and a ratio operation is performed to obtain a standard acoustic impedance ratio; According to the ratio range of the standard acoustic impedance ratio, the thickness standing wave sensitivity weight coefficient is set for the 2-4 MHz low frequency band, and the interface void sensitivity weight coefficient is set for the 6-8 MHz high frequency band. After amplitude modulation processing, they are respectively synthesized into defect-selective broadband acoustic excitation signals.

[0008] Preferably, the collecting of the acoustic propagation response and thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone to obtain an acoustic-thermal synchronous response sequence comprises: performing path separation processing on the direct wave signal, the interface reflection wave signal, and the mode conversion wave signal in the acoustic propagation response according to the traveling speed of the aluminum-based copper-clad laminate and the length of the hot pressing zone to obtain multipath acoustic response data; According to the temperature distribution of the aluminum-based copper-clad laminate in the hot pressing zone, temperature compensation processing is performed on the 3-5 μm band signal and the 8-12 μm band signal in the thermal diffusion response to obtain corrected dual-band thermal response data; Performing spatiotemporal calibration processing on the multipath acoustic response data and the corrected dual-band thermal response data according to the longitudinal coordinates of the panel surface, establishing a correspondence between the response signal and the spatial position, and obtaining spatiotemporally correlated acoustic-thermal response data; The temporally and spatially correlated acoustic-thermal response data are serialized and arranged according to a uniform sampling time interval to obtain an acoustic-thermal synchronous response sequence.

[0009] Preferably, the acoustic-thermal synchronous response sequence and the defect-free reference spectrum are subjected to differential transformation and time-frequency domain analysis to extract the multi-frequency sound velocity drift rate, the acoustic attenuation vector, the dual-band thermal attenuation slope and the thermal hysteresis ratio to obtain the acoustic-thermal coupling feature set, including: The acoustic-thermal synchronous response sequence is divided into a flow state response section, a gelation state response section, and a solidification state response section according to the resin gelation window, and differential transformation processing is performed on each of them with the defect-free reference spectrum of the corresponding state to obtain a staged acoustic difference spectrum and a staged thermal difference spectrum; Performing interface separation and analysis on the phased acoustic difference spectrum according to its frequency characteristics, extracting the acoustic velocity drift trends at the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface, and obtaining the dual-interface multi-frequency acoustic velocity drift rate; According to the variation pattern of the dual-interface multi-frequency acoustic velocity drift rate at different gelation window stages, the defect evolution trajectory is analyzed and processed in combination with the interface void formation mechanism to obtain the acoustic attenuation vector for void development prediction; A hierarchical identification process is performed on the thermal diffusion anomaly caused by filler sedimentation in the staged thermal differential spectrum. By comparing the attenuation differences between the 3-5μm band signal and the 8-12μm band signal at each gelation window stage, a dual-band thermal attenuation slope representing the filler gradient and a thermal hysteresis ratio representing the sedimentation dynamics are obtained. The dual-interface multi-frequency acoustic velocity drift rate, the acoustic attenuation vector for cavity development prediction, the dual-band thermal attenuation slope characterized by filler gradient, and the thermal hysteresis ratio characterized by sedimentation dynamics are subjected to multi-dimensional correlation fusion processing to obtain the acoustic-thermal coupling feature set.

[0010] Preferably, the defect evolution trajectory analysis is performed based on the variation pattern of the dual-interface multi-frequency acoustic velocity drift rate at different gelation window stages in combination with the interface void formation mechanism to obtain the acoustic attenuation vector for void development prediction, including: The evolution trajectory of the dual-interface multi-frequency sound velocity drift rate is reconstructed according to the three gelation window stages of the flow state, gelation state, and solidification state, and the difference in the sound velocity change trend of the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface is analyzed to obtain the interface acoustic evolution trajectory data; According to the abnormal change points and amplitudes of the sound velocity in the interface acoustic evolution trajectory data, combined with the resin degassing process and the interface infiltration mechanism, the cavity initiation moment, expansion path and final size are predicted and calculated to obtain the acoustic attenuation vector for cavity development prediction.

[0011] Preferably, the defect identification and grading evaluation based on the correlation pattern of acoustic features and thermal features in the acoustic-thermal coupling feature set is combined with a dynamically adjusted process tolerance threshold to obtain a spatially distributed defect grade map, including: Based on the interface difference pattern of the dual-interface multi-frequency acoustic velocity drift rate in the acoustic-thermal coupling feature set and the interlayer difference pattern of the dual-band thermal attenuation slope characterized by the filler gradient, acoustic-thermal feature cross-matching processing is performed to identify three defect types: interface voids, thickness standing waves, and filler settlement, and obtain defect type identification results; According to the real-time temperature, pressure and travel speed of the aluminum-based copper-clad laminate in the hot pressing zone, a process state compensation adjustment process is performed on the preset reference tolerance threshold to obtain a dynamically adjusted process tolerance threshold; The acoustic attenuation vector amplitude predicted by the cavity development at each defect location in the defect type identification result and the thermal hysteresis ratio value characterized by the sedimentation dynamics are quantitatively compared with the dynamically adjusted process tolerance threshold, and the defect degree is divided into three levels: mild, moderate, and severe to obtain a defect grading assessment result; The defect grading assessment results are spatially mapped according to the longitudinal coordinates of the panel surface, and connected domain analysis is performed on similar defects at adjacent positions to obtain a spatially distributed defect grade map.

[0012] Preferably, the acoustic-thermal feature cross-matching process is performed based on the interface difference mode of the dual-interface multi-frequency acoustic velocity drift rate and the interlayer difference mode of the dual-band thermal attenuation slope characterized by the filler gradient in the acoustic-thermal coupling feature set to identify three defect types: interface voids, thickness standing waves, and filler sedimentation, and obtain defect type identification results, including: The multi-frequency sound velocity drift rate of the dual interface in the acoustic-thermal coupling feature set is subjected to interface difference pattern recognition processing to extract the sound velocity drift difference characteristics of the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface; the dual-band thermal attenuation slope characterized by the filler gradient in the acoustic-thermal coupling feature set is subjected to interlayer difference pattern recognition processing to extract the thermal attenuation difference characteristics of the 3-5μm band and the 8-12μm band; According to the coupled combination mode of the acoustic velocity drift difference feature and the thermal attenuation difference feature, pattern matching processing is performed with the preset interface void feature template, thickness standing wave feature template and filler sedimentation feature template to determine the dominant defect type at each position and obtain the defect type identification result.

[0013] A second aspect of the present invention provides a film forming device for an aluminum-based copper-clad laminate, the film forming device for an aluminum-based copper-clad laminate comprising: An excitation application module is used to simultaneously apply broadband acoustic excitation and dual-band thermal excitation when the aluminum-based copper-clad laminate enters the hot pressing zone; a response acquisition module, configured to acquire the acoustic propagation response and thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone, and obtain an acoustic-thermal synchronous response sequence; a feature extraction module for performing differential transformation and time-frequency domain analysis on the acoustic-thermal synchronous response sequence and the defect-free reference spectrum, extracting the multi-frequency sound velocity drift rate, the acoustic attenuation vector, the dual-band thermal attenuation slope, and the thermal hysteresis ratio to obtain an acoustic-thermal coupling feature set; The defect recognition module is used to perform defect recognition and grade evaluation based on the correlation pattern of acoustic features and thermal features in the acoustic-thermal coupling feature set and in combination with the dynamically adjusted process tolerance threshold to obtain a spatially distributed defect grade map.

[0014] The third aspect of the present invention provides an aluminum-based copper-clad laminate film forming device, comprising: a memory and at least one processor, wherein instructions are stored in the memory, and the memory and the at least one processor are interconnected through lines; the at least one processor calls the instructions in the memory so that the aluminum-based copper-clad laminate film forming device performs the steps of the above-mentioned aluminum-based copper-clad laminate film forming method.

[0015] The present invention applies broadband acoustic excitation and dual-band thermal excitation simultaneously when the aluminum-based copper-clad laminate enters the hot pressing zone, and utilizes the acoustic and thermal conductivity of the aluminum substrate and copper foil as signal propagation media, thereby completely embedding the detection process into the process flow of the lamination molding. When the plate moves in the hot pressing zone, the system collects the acoustic propagation response and thermal diffusion response in real time to form an acoustic-thermal synchronous response sequence. This process is carried out simultaneously with the lamination molding and does not require additional detection time. By performing differential transformation and time-frequency domain analysis on the collected response sequence and the pre-stored defect-free reference spectrum, key characteristic parameters reflecting the changes in the internal state of the material can be extracted, including multi-frequency sound velocity drift rate, sound attenuation vector, dual-band thermal attenuation slope and thermal hysteresis ratio. The acoustic-thermal coupling feature set composed of these parameters can sensitively reflect the presence of defects such as interface voids, uneven filler distribution, and thickness standing waves.

[0016] Based on the correlation pattern analysis of acoustic and thermal features in the acoustic-thermal coupling feature set, combined with the tolerance threshold dynamically adjusted according to real-time process parameters, the system can accurately identify different types of defects and conduct graded assessments, ultimately generating a spatially distributed defect grade map. Since the entire detection and analysis process completely overlaps with the lamination molding process in time, it not only avoids the additional time overhead required for traditional post-detection, but also enables timely detection of quality problems during the molding process, fundamentally solving the problems of long detection lags and low production efficiency. This method of integrating physical excitation, response acquisition, feature extraction, and defect identification into the molding process achieves a seamless integration of quality inspection and production processes, significantly improving the overall efficiency and quality control level of aluminum-based copper clad laminate lamination molding. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the structures shown in these drawings without paying any creative work.

[0018] Figure 1 Schematic diagram of an embodiment of a film forming method for an aluminum-based copper-clad laminate according to an embodiment of the present invention; Figure 2 Schematic diagram of an embodiment of a film forming device for an aluminum-based copper-clad laminate according to an embodiment of the present invention; Figure 3 Schematic diagram of an embodiment of the film forming equipment of the aluminum-based copper-clad laminate in an embodiment of the present invention.

[0019] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings and in conjunction with the embodiments. DETAILED DESCRIPTION

[0020] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0021] It should be noted that if the embodiments of the present invention involve directional indications (such as up, down, left, right, front, back, etc.), the directional indications are only used to explain the relative position relationship, movement status, etc. between the various components under a certain specific posture. If the specific posture changes, the directional indications will also change accordingly.

[0022] In addition, the descriptions of "first", "second", etc. in the present invention are only for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In addition, "and / or" in the full text includes three solutions. Taking A and / or B as an example, it includes technical solution A, technical solution B, and technical solution that satisfies both A and B. In addition, the technical solutions between the various embodiments can be combined with each other, and must be based on the ability of ordinary technicians in this field to implement. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by the present invention.

[0023] An embodiment of the present application provides a method for laminating and forming an aluminum-based copper-clad laminate. Figure 1 A flow chart of a method for laminating and forming an aluminum-based copper-clad laminate provided in one embodiment of the present application. In this embodiment, the method includes: See also Figure 1 , broadband acoustic excitation and dual-band thermal excitation are applied simultaneously when the aluminum-based copper-clad laminate enters the hot pressing zone; In one embodiment of the present invention, the simultaneous application of broadband acoustic excitation and dual-band thermal excitation when the aluminum-based copper-clad laminate enters the hot pressing zone includes: According to the acoustic impedance ratio of the aluminum substrate to the copper foil in the aluminum-based copper-clad laminate, a 2-8 MHz linear frequency modulation acoustic pulse is divided into a 2-4 MHz low frequency band and a 6-8 MHz high frequency band, the low frequency band is amplitude-enhanced to detect thickness standing wave defects, and the high frequency band is amplitude-enhanced to detect interface void defects, thereby obtaining a defect-selective broadband acoustic excitation signal; According to the thickness of the insulating dielectric layer and the filler content in the aluminum-based copper-clad laminate, the 3-5 μm band infrared pulse is pulse-width modulated to detect the shallow filler distribution, and the 8-12 μm band infrared pulse is intensity modulated to detect the deep filler distribution, thereby obtaining a dual-band thermal excitation signal for layered detection; The defect-selective broadband acoustic excitation signal and the layered detection dual-band thermal excitation signal are applied synchronously in the resin gelation window to ensure that the excitation propagation path is consistent along the acoustic and thermal conduction channel between the copper surface and the aluminum surface.

[0024] The following is a detailed description of the steps involved in the above embodiment: The acoustic impedance ratio is obtained by reading the product specification of the aluminum-based copper-clad laminate to obtain the aluminum substrate thickness (usually 1.0-3.0mm), copper foil thickness (usually 18-105μm) and material grade information, and then querying the corresponding acoustic impedance value from the material property database. The acoustic impedance of the aluminum alloy substrate is about 17×10 6 kg / (m²·s), the acoustic impedance of copper foil is about 42×10 6 kg / (m²·s), resulting in a calculated acoustic impedance ratio of approximately 0.4. Based on this ratio, the signal processing system separates the 2-8 MHz linear frequency-modulated pulses into two frequency bands via a bandpass filter: a low-frequency band of 2-4 MHz and a high-frequency band of 6-8 MHz. Amplitude enhancement is achieved using a programmable gain amplifier. When the acoustic impedance ratio is within the range of 0.3-0.5, the gain of the low-frequency band is set to 6 dB (i.e., the amplitude is doubled), and the gain of the high-frequency band is set to 9 dB (i.e., the amplitude is increased approximately 1.8 times). The physical mechanism of enhancement is that the characteristic size of thickness standing wave defects is related to the overall thickness of the plate (millimeter-level), requiring long-wavelength, low-frequency acoustic waves for resonance. In contrast, the characteristic size of interfacial void defects is typically micron-level, requiring short-wavelength, high-frequency acoustic waves for effective detection. After differential enhancement, the two frequency bands are recombined by a signal synthesizer to form a defect-selective broadband acoustic excitation signal. This selective design enables optimized detection of different defect types in a single excitation, avoiding the time-consuming process of multiple excitations.

[0025] The thickness of the insulating dielectric layer is determined by product design parameters (typical range: 50-200 μm), and the filler content is determined by the material formulation (typical range: 20-60 weight percent). Infrared pulse modulation is achieved through pulse width and power control of the laser driver circuit. For the 3-5 μm band, the pulse width modulator adjusts proportionally to the dielectric layer thickness: for every 50 μm increase in thickness, the pulse width increases by 200 ns. This ensures that the heat diffusion depth is approximately 1 / 3 of the dielectric layer thickness, which is used for detecting shallow areas near the copper foil interface. For the 8-12 μm band, the power controller adjusts the laser intensity based on the filler content: for every 10% increase in filler content, the laser power increases by 15% to compensate for scattering losses of infrared radiation from filler particles. Shallow filler distribution refers to the filler distribution within 0-30% of the thickness from the top surface, while deep filler distribution refers to the filler distribution within 0-30% of the thickness from the bottom surface. This layered detection design is based on the physical law of filler sedimentation: heavy ceramic fillers tend to settle downward under the influence of gravity during resin flow, forming a gradient distribution with sparseness at the top and denseness at the bottom. This distribution unevenness can be quantified through differentiated detection at different heat penetration depths.

[0026] The resin gelation window is determined by temperature sensors and an online viscosity monitoring system. When the resin temperature reaches 120-140°C and the viscosity begins to rise significantly, the gelation process begins. This window typically lasts 15-25 seconds. Synchronous application is achieved through a unified clock signal from the main controller. The acoustic excitation generator and infrared laser receive the same trigger pulse, ensuring that the time deviation between the two excitations is less than 1 millisecond. The consistency of the propagation path is guaranteed by the equipment's mechanical positioning system: the acoustic transducer and infrared laser are arranged on the upper and lower sides of the hot pressing equipment with the same lateral spacing (typically 20-50mm) and longitudinal alignment accuracy (±1mm), so that both the sound waves and the heat pulses propagate along a straight path perpendicular to the plate surface. The sound and heat conduction channels utilize the metallic properties of the aluminum substrate and copper foil: the propagation speed of sound waves in aluminum is approximately 6420m / s and in copper is approximately 4760m / s; the diffusion coefficient of heat diffusion in aluminum is approximately 97×10 -6 m² / s, which is about 111×10 -6 m² / s. This synchronous application design ensures that the acoustic and thermal signals have the same time starting point and spatial coordinates, providing an accurate data basis for subsequent multi-physics field coupling analysis. At the same time, the application timing within the gelation window ensures that the excitation signal can obtain the maximum defect response signal at the stage when the material property changes are most sensitive.

[0027] In one embodiment of the present invention, according to the acoustic impedance ratio of the aluminum substrate to the copper foil in the aluminum-based copper clad laminate, the 2-8 MHz linear frequency modulation acoustic pulse is divided into a 2-4 MHz low frequency band and a 6-8 MHz high frequency band, the low frequency band is amplitude enhanced to detect thickness standing wave defects, and the high frequency band is amplitude enhanced to detect interface void defects, thereby obtaining a defect-selective broadband acoustic excitation signal, including: According to the thickness specification parameters and material model of the aluminum-based copper-clad laminate, the corresponding aluminum substrate acoustic impedance value and copper foil acoustic impedance value are obtained from a preset acoustic impedance database, and a ratio operation is performed to obtain a standard acoustic impedance ratio; According to the ratio range of the standard acoustic impedance ratio, the thickness standing wave sensitivity weight coefficient is set for the 2-4 MHz low frequency band, and the interface void sensitivity weight coefficient is set for the 6-8 MHz high frequency band. After amplitude modulation processing, they are respectively synthesized into defect-selective broadband acoustic excitation signals.

[0028] The following is a detailed description of the steps involved in the above embodiment: The thickness specification parameters and material model are obtained by reading the product identification or technical documents of the aluminum-based copper clad laminate. The thickness specification parameters include the thickness of the aluminum substrate (range 0.8-3.0mm), the thickness of the insulating dielectric layer (range 50-200μm) and the thickness of the copper foil (range 18-105μm). The material model identifies the aluminum alloy grade (such as 1060, 6061) and the type of copper foil (such as electrolytic copper, rolled copper). The preset acoustic impedance database is a material property reference table stored in the system memory. It is indexed according to different material grades and thickness specifications and records the corresponding acoustic impedance values. Database query is achieved through a material model matching algorithm: the system searches for the corresponding aluminum substrate acoustic impedance value based on the input aluminum alloy grade (1060 aluminum alloy is approximately 17.2×10 6 kg / (m²·s), 6061 aluminum alloy is about 17.8×10 6 kg / (m²·s)), find the copper foil acoustic impedance value according to the copper foil type (electrolytic copper is about 41.6×10 6 kg / (m²·s), rolled copper is about 42.3×10 6 kg / (m²·s)). The ratio calculation is performed using a divider circuit, dividing the acoustic impedance of the aluminum substrate by the acoustic impedance of the copper foil to obtain a standard acoustic impedance ratio, which ranges from 0.40 to 0.43. For example, for a 1060 aluminum substrate and electrolytic copper foil combination, the standard acoustic impedance ratio is 17.2÷41.6≈0.414. This standard acoustic impedance ratio reflects the degree of acoustic impedance matching between the two metal materials. This ratio directly affects the reflection and transmission characteristics of sound waves at the interface, providing a physical basis for subsequent frequency weighting.

[0029] The standard acoustic impedance ratio is classified according to the preset range: 0.38-0.40 is the low impedance ratio range, 0.40-0.42 is the medium impedance ratio range, and 0.42-0.45 is the high impedance ratio range. The thickness standing wave sensitivity weight coefficient is an amplitude modulation multiple set for the low frequency band of 2-4MHz. This coefficient is determined according to the acoustic impedance ratio range: the low impedance ratio range is set to 1.2 times, the medium impedance ratio range is set to 1.5 times, and the high impedance ratio range is set to 1.8 times. The interface void sensitivity weight coefficient is an amplitude modulation multiple set for the high frequency band of 6-8MHz: the low impedance ratio range is set to 2.2 times, the medium impedance ratio range is set to 2.0 times, and the high impedance ratio range is set to 1.8 times. The amplitude modulation processing is achieved through a variable gain amplifier, which adjusts the signal amplitude of the corresponding frequency band according to the weight coefficient. For example, when the standard acoustic impedance ratio is 0.414, which belongs to the medium impedance ratio range, the 2-4MHz frequency band is modulated to 1.5 times the amplitude, and the 6-8MHz frequency band is modulated to 2.0 times the amplitude. Signal synthesis recombines the two modulated frequency band signals through a frequency domain adder to form a defect-selective broadband acoustic excitation signal with differentiated spectral characteristics. The setting of the weight coefficient is based on the physical sensitivity of different frequency sound waves to defect detection: low-frequency sound waves have a longer wavelength, which matches the overall thickness of the plate and has a stronger excitation ability for thickness standing wave defects; high-frequency sound waves have a shorter wavelength, which matches the micron-level interface defect size and has a higher detection sensitivity for interface void defects. This differentiated weight distribution realizes the simultaneous optimized detection of multiple defect types with a single excitation, avoiding the time consumption and equipment complexity of multiple excitations required by traditional methods.

[0030] Please continue reading Figure 1 , collecting the acoustic propagation response and thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone to obtain an acoustic-thermal synchronous response sequence; In one embodiment of the present invention, the step of collecting the acoustic propagation response and thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone to obtain an acoustic-thermal synchronous response sequence includes: performing path separation processing on the direct wave signal, the interface reflection wave signal, and the mode conversion wave signal in the acoustic propagation response according to the traveling speed of the aluminum-based copper-clad laminate and the length of the hot pressing zone to obtain multipath acoustic response data; According to the temperature distribution of the aluminum-based copper-clad laminate in the hot pressing zone, temperature compensation processing is performed on the 3-5 μm band signal and the 8-12 μm band signal in the thermal diffusion response to obtain corrected dual-band thermal response data; Performing spatiotemporal calibration processing on the multipath acoustic response data and the corrected dual-band thermal response data according to the longitudinal coordinates of the panel surface, establishing a correspondence between the response signal and the spatial position, and obtaining spatiotemporally correlated acoustic-thermal response data; The temporally and spatially correlated acoustic-thermal response data are serialized and arranged according to a uniform sampling time interval to obtain an acoustic-thermal synchronous response sequence.

[0031] The following is a detailed description of the steps involved in the above embodiment: The speed of the aluminum-clad copper laminate in the hot press is measured in real time using an encoder or laser rangefinder, typically ranging from 0.5-2.0 m / s. The hot press length is a fixed parameter of the equipment, ranging from 0.3-1.2 m. Path separation is achieved based on the time difference and amplitude characteristics of sound wave propagation: the direct wave signal is the sound wave that propagates directly from the transmitting point to the receiving point and has the shortest propagation time; the interface reflection wave signal is the sound wave that reaches the receiving point after reflection at the copper foil-insulating dielectric layer interface or the insulating dielectric layer-aluminum substrate interface and has a longer propagation time; the mode conversion wave signal is the sound wave that propagates after longitudinal-to-transverse wave conversion at the interface and has a characteristic frequency shift. Separation is achieved using a time-domain window function filter: the theoretical arrival time of each path is calculated based on the plate thickness and the corresponding time window is set. For example, for a plate with a total thickness of 1.5 mm, the arrival time of the direct wave is approximately 0.25 μs, the arrival time of the interface reflection wave is approximately 0.45 μs, and the arrival time of the mode conversion wave is approximately 0.38 μs. Multipath acoustic response data refers to three independent sets of acoustic wave signal sequences after path separation, each set containing amplitude, phase, and frequency information. For example, when a plate passes through a 0.6m long hot press zone at a speed of 1.0m / s, the system continuously collects and separates the acoustic responses of the three propagation paths within 0.6 seconds. This path separation process eliminates interference between signals from different propagation paths, allowing each defect type to be accurately identified on the corresponding most sensitive propagation path. The direct wave mainly reflects the overall uniformity of the material, the interface reflection wave is most sensitive to interface defects, and the mode conversion wave is sensitive to changes in the elastic constants within the material.

[0032] Temperature distribution is monitored in real time at various locations within the hot press zone using an array of infrared thermal imagers with a measurement accuracy of ±2°C and a sampling frequency of 50Hz. Temperature compensation addresses the effect of temperature variations on the signal amplitude in the thermal diffusion response. The radiant intensity of both the 3-5μm and 8-12μm bands varies exponentially with temperature, necessitating normalization of the raw signals based on the measured temperature. This compensation algorithm utilizes a table lookup method: the system pre-stores a table of emissivity corrections for different temperatures. The corresponding correction coefficient is then found based on the measured temperature. The raw signal is then divided by the correction coefficient to obtain the compensated signal. For example, when the measurement point temperature is 150°C, the correction coefficient for the 3-5μm band is 1.34, and the correction coefficient for the 8-12μm band is 1.28. The raw signal is then divided by the corresponding correction coefficient to obtain the normalized signal amplitude. The corrected dual-band thermal response data represents the normalized thermal radiation signal after temperature effects have been eliminated, effectively reflecting the changes in the thermal diffusion characteristics within the material. Temperature compensation processing eliminates the systematic deviation of thermal signals caused by the temperature gradient in the hot pressing zone, making the thermal response data collected at different locations comparable and ensuring that defect identification is not affected by ambient temperature fluctuations.

[0033] The spatiotemporal calibration process achieves precise correspondence between the response signal and spatial position by synchronously recording the signal acquisition time and the plate position. The longitudinal coordinate of the plate surface is determined by measuring the plate's movement distance using a photoelectric encoder. The coordinate origin is set to the position where the front end of the plate enters the hot press zone, and the longitudinal coordinate range is from 0 to the plate length. The spatiotemporal correspondence is achieved through a linear mapping of timestamps and position coordinates: the system records the timestamp of each sampling point, calculates the corresponding longitudinal coordinate position based on the travel speed, and establishes a one-to-one correspondence between time and space. For example, when the plate travels at a speed of 1.5m / s, the longitudinal coordinate corresponding to the sampling time t=0.2s is 0.3m, and the longitudinal coordinate corresponding to the sampling time t=0.4s is 0.6m. Spatiotemporally correlated acoustic-thermal response data refers to a comprehensive data set in which each data point simultaneously contains multipath acoustic response data, corrected dual-band thermal response data, and corresponding spatial coordinates. The spatiotemporal calibration process ensures the spatial registration accuracy of acoustic and thermal signals, enabling multimodal signals to be accurately associated with the same physical location on the panel surface, providing a reliable spatial reference for subsequent defect spatial positioning and acoustic-thermal coupling analysis.

[0034] The sampling interval is uniformly set to 2ms, and the corresponding spatial resolution is determined by the travel speed. For a travel speed of 1.0m / s, the spatial resolution is 2mm. Serialization is achieved through a data rearrangement algorithm: the spatiotemporally correlated acoustic-thermal response data are arranged in chronological order according to the timestamps, forming a continuous data sequence. This arrangement process maintains the multidimensional integrity of each data point, meaning that each sequence element contains the complete acoustic and thermal characteristic information at the corresponding time and location. An acoustic-thermal synchronous response sequence is a multidimensional data sequence arranged in chronological order. The sequence length is equal to the total number of sampling points, and each sequence element contains three acoustic responses, two thermal responses, and a spatial coordinate. For example, for a plate 1.0m long and traveling at a speed of 1.0m / s, sampling at 2ms intervals over a 1.0-second transit time yields a sequence of 500 data points. Serialization converts the spatially distributed multimodal signals into a time-series data format, facilitating subsequent digital signal processing algorithms for time-frequency domain analysis and feature extraction, while maintaining the signal's temporal continuity and spatial integrity.

[0035] Please continue reading Figure 1 , performing differential transformation and time-frequency domain analysis on the acoustic-thermal synchronous response sequence and the defect-free reference spectrum, extracting the multi-frequency sound velocity drift rate, acoustic attenuation vector, dual-band thermal attenuation slope and thermal hysteresis ratio, and obtaining the acoustic-thermal coupling feature set; In one embodiment of the present invention, the acoustic-thermal synchronous response sequence and the defect-free reference spectrum are differentially transformed and analyzed in the time-frequency domain to extract the multi-frequency sound velocity drift rate, acoustic attenuation vector, dual-band thermal attenuation slope and thermal hysteresis ratio to obtain the acoustic-thermal coupling feature set, including: The acoustic-thermal synchronous response sequence is divided into a flow state response section, a gelation state response section, and a solidification state response section according to the resin gelation window, and differential transformation processing is performed on each of them with the defect-free reference spectrum of the corresponding state to obtain a staged acoustic difference spectrum and a staged thermal difference spectrum; Performing interface separation and analysis on the phased acoustic difference spectrum according to its frequency characteristics, extracting the acoustic velocity drift trends at the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface, and obtaining the dual-interface multi-frequency acoustic velocity drift rate; According to the variation pattern of the dual-interface multi-frequency acoustic velocity drift rate at different gelation window stages, the defect evolution trajectory is analyzed and processed in combination with the interface void formation mechanism to obtain the acoustic attenuation vector for void development prediction; A hierarchical identification process is performed on the thermal diffusion anomaly caused by filler sedimentation in the staged thermal differential spectrum. By comparing the attenuation differences between the 3-5μm band signal and the 8-12μm band signal at each gelation window stage, a dual-band thermal attenuation slope representing the filler gradient and a thermal hysteresis ratio representing the sedimentation dynamics are obtained. The dual-interface multi-frequency acoustic velocity drift rate, the acoustic attenuation vector for cavity development prediction, the dual-band thermal attenuation slope characterized by filler gradient, and the thermal hysteresis ratio characterized by sedimentation dynamics are subjected to multi-dimensional correlation fusion processing to obtain the acoustic-thermal coupling feature set.

[0036] The following is a detailed description of the steps involved in the above embodiment: The resin gelation window is divided based on real-time monitoring of resin viscosity and temperature changes, with the time boundaries of the three stages determined using an online viscometer and temperature sensor. The flow response phase corresponds to the period when the resin viscosity is below 1000 cP and the temperature is between 80-120°C, lasting approximately 5-8 seconds. The gelation response phase corresponds to the period when the viscosity rapidly rises to 1000-10,000 cP and the temperature reaches 120-150°C, lasting approximately 8-12 seconds. The curing response phase corresponds to the period when the viscosity exceeds 10,000 cP and the temperature exceeds 150°C, lasting approximately 10-15 seconds. The system divides the acoustic-thermal synchronous response sequence into three subsequences based on time stamps, each containing all sampled data for the corresponding phase. The defect-free reference spectrum is a pre-stored standard response signal corresponding to the ideal material state for each of the three gelation phases, obtained through statistical analysis of qualified products. Differential transformation processing is implemented using a digital subtractor: the response sequence for the current phase is subtracted point by point from the corresponding defect-free reference spectrum to generate a deviation signal. For example, the acoustic response at a sampling point in the flow dynamic response section is -12.3dB, while the corresponding defect-free baseline value is -10.8dB, resulting in a differential result of -1.5dB. The staged acoustic and thermal differential spectra represent differential signal sequences for each of the three gelation stages. Each differential spectrum contains deviation information in both the frequency and time domains. This staged processing identifies the evolution of defects in materials at different gelation states, avoiding misjudgments caused by misinterpreting dynamic changes as static. The use of dedicated baseline spectra for each stage ensures accurate comparisons across different material states.

[0037] Interface separation analysis analyzes frequency characteristics based on the differences in acoustic frequency sensitivity to different interfaces. The phased acoustic difference spectrum is separated into high-frequency components (6-8 MHz) and low-frequency components (2-4 MHz) using a bandpass filter. The high-frequency component primarily reflects changes in the state of the copper foil-insulating dielectric interface, while the low-frequency component primarily reflects changes in the state of the insulating dielectric-aluminum substrate interface. Sound velocity drift trends are extracted through phase difference analysis: phase changes at different times for the same frequency are compared to calculate the relative rate of change of sound velocity. The sound velocity drift rate of the copper foil-insulating dielectric interface is determined by analyzing the phase gradient in the 6-8 MHz frequency band, while the sound velocity drift rate of the insulating dielectric-aluminum substrate interface is determined by analyzing the phase gradient in the 2-4 MHz frequency band. The dual-interface multi-frequency sound velocity drift rate is a dataset of sound velocity change rates at multiple frequency points for both interfaces, including the high-frequency drift rate of the upper interface and the low-frequency drift rate of the lower interface. For example, in the gelled state response, the acoustic velocity drift rate of the upper interface at 6 MHz is -2.3%, and the acoustic velocity drift rate of the lower interface at 3 MHz is -1.8%. Interface separation analysis utilizes the corresponding relationship between acoustic wave frequency and interface depth. High-frequency acoustic waves have a shallow penetration depth and are sensitive to surface interfaces, while low-frequency acoustic waves have a deep penetration depth and are sensitive to deeper interfaces. This separation method can independently monitor the state changes of the two interfaces, avoiding signal aliasing between the interfaces.

[0038] The defect evolution trajectory analysis and processing is achieved by tracking the time-series changes of the multi-frequency sound velocity drift rate of the dual interface in the three gelation window stages. The change pattern recognition is achieved through the pattern matching algorithm: the measured sound velocity drift rate time series curve is correlated with the preset defect evolution pattern library to identify the most matching evolution pattern. The interface void formation mechanism is based on the physical process of insufficient resin degassing and poor interface wetting: in the flow state stage, bubbles begin to gather at the interface; in the gelation state stage, the increase in viscosity restricts the discharge of bubbles; in the solidification state stage, bubbles are solidified in the interface to form permanent voids. The trajectory analysis algorithm calculates the predicted growth trend of the void size and estimates the development trajectory of the void volume based on the amplitude of the sound velocity decrease. The acoustic attenuation vector for the prediction of void development refers to the acoustic energy loss vector caused by the predicted void in the subsequent solidification process, which includes two components: amplitude attenuation and frequency attenuation. For example, when the acoustic velocity drift rate of the upper interface is -1.2% in the flowing state, -3.5% in the gelled state, and -5.8% in the solidified state, the predicted acoustic attenuation vector amplitude component is -8.2 dB, and the frequency component is +150 Hz. This evolution trajectory analysis upgrades static defect detection to dynamic defect prediction, enabling the identification of development trends before defects fully form, achieving a technological leap from passive detection to active early warning.

[0039] Hierarchical identification processing identifies thermal diffusion anomalies based on the vertical distribution characteristics of filler sedimentation. The 3-5μm band signal primarily reflects the thermal diffusion characteristics of the upper layer (0-30% thickness from the copper surface), while the 8-12μm band signal primarily reflects the thermal diffusion characteristics of the lower layer (0-30% thickness from the aluminum surface). Attenuation difference comparison is achieved by calculating the difference in the attenuation slopes of the two band signals at each gelation window stage: the upper attenuation slope minus the lower attenuation slope yields the interlayer gradient index. The dual-band thermal attenuation slope, representing the filler gradient, is a quantitative indicator of the uneven distribution of filler across the thickness. A positive value indicates downward sedimentation, while a negative value indicates upward buoyancy. The thermal hysteresis ratio, representing sedimentation dynamics, is the ratio of the time difference between the attenuation peaks of the two band signals to the average arrival time. This ratio reflects the dynamic characteristics of the filler sedimentation process. For example, when the attenuation slope of the upper 3-5μm band in the gelled state is -0.8dB / s, and the attenuation slope of the lower 8-12μm band is -1.4dB / s, the filler gradient index is 0.6dB / s, indicating filler sedimentation. A thermal hysteresis ratio of 0.15 means that the lower signal reaches its attenuation peak 15% later than the upper signal. Hierarchical identification processing fully utilizes the differential penetration of different infrared bands into material depth, enabling vertical cross-sectional analysis of filler distribution and quantifying the extent and rate of filler sedimentation.

[0040] Multidimensional correlation fusion processing combines four characteristic parameters into a unified feature set using a multivariate correlation analysis algorithm. The correlation matrix calculates the Pearson correlation coefficient between each parameter and identifies linear and nonlinear correlations between them. Fusion weights are determined based on the contribution of each parameter to defect detection: a weight of 0.35 is assigned to the dual-interface multi-frequency acoustic velocity drift rate, a weight of 0.25 is assigned to the acoustic attenuation vector for cavity development prediction, a weight of 0.25 is assigned to the dual-band thermal attenuation slope for filler gradient characterization, and a weight of 0.15 is assigned to the thermal hysteresis ratio for sedimentation dynamics characterization. The acoustic-thermal coupling feature set is the fused multidimensional feature vector, where each vector element contains a weighted composite eigenvalue and the corresponding spatial coordinate. For example, the acoustic-thermal coupling feature set for a given location is [acoustic velocity drift rate: -2.1%, acoustic attenuation vector: -6.8 dB, thermal attenuation slope: 0.4 dB / s, thermal hysteresis ratio: 0.12, coordinate: 0.35 m]. Multi-dimensional correlation fusion processing realizes the collaborative analysis of acoustic and thermal features, and uses different physical fields to cross-validate the differentiated responses to the same defect, significantly improving the accuracy and reliability of defect identification. At the same time, the fused feature set maintains the integrity and spatial correlation of the original information.

[0041] In one embodiment of the present invention, the defect evolution trajectory analysis is performed based on the variation pattern of the dual-interface multi-frequency acoustic velocity drift rate at different gelation window stages in combination with the interface void formation mechanism to obtain the acoustic attenuation vector for void development prediction, including: The evolution trajectory of the dual-interface multi-frequency sound velocity drift rate is reconstructed according to the three gelation window stages of the flow state, gelation state, and solidification state, and the difference in the sound velocity change trend of the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface is analyzed to obtain the interface acoustic evolution trajectory data; According to the abnormal change points and amplitudes of the sound velocity in the interface acoustic evolution trajectory data, combined with the resin degassing process and the interface infiltration mechanism, the cavity initiation moment, expansion path and final size are predicted and calculated to obtain the acoustic attenuation vector for cavity development prediction.

[0042] The following is a detailed description of the steps involved in the above embodiment: The evolution trajectory reconstruction process uses a time-series data analysis algorithm to continuously reconstruct the multi-frequency sound velocity drift rate of the dual interfaces according to three gelation window stages. The system first classifies the sound velocity drift rate data for the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface into three time periods: the flowing state (0-8 seconds), the gelation state (8-20 seconds), and the solidification state (20-35 seconds). The reconstruction algorithm uses cubic spline interpolation to fill in the data gaps between stages, forming a continuous sound velocity variation curve. Trend difference analysis is achieved through slope calculation and inflection point identification: the slope of the sound velocity change of the upper and lower interfaces at each stage is calculated to identify the difference in the sound velocity change patterns between the two interfaces. For example, in the gelation stage, the sound velocity drift rate of the upper interface changes from -1.5% to -4.2%, with a slope of -0.225% / second, while the lower interface changes from -0.8% to -2.1%, with a slope of -0.108% / second, and a difference slope of -0.117% / second. Interface acoustic evolution trajectory data is a complete dataset describing the temporal evolution of the sound velocity at two interfaces, including the slope of change, inflection point moments, and difference magnitudes at each stage. This evolution trajectory reconstruction process converts discrete sound velocity measurement data into a continuous description of the dynamic process, enabling the identification of the temporal patterns of interface state changes and the mutual influence between the two interfaces, achieving a technological upgrade from static feature extraction to dynamic process analysis.

[0043] Abnormal sound velocity change points are identified using a change point detection algorithm based on the principles of cumulative sum control charts. When the cumulative deviation of the sound velocity change rate exceeds a preset threshold, it is marked as an abnormal change point. The magnitude of the change is calculated by calculating the difference in sound velocity before and after the abnormal change point. The threshold is set at three standard deviations of normal fluctuations, approximately 0.5%. The resin degassing process describes the release of solvents and volatiles from insulating resin during heating: degassing is greatest during the fluidization phase, limited during the gelation phase, and virtually ceases during the solidification phase. The interface wetting mechanism is based on a dynamic balance between surface tension and viscosity: at low viscosity, the resin can fully wet the microscopic asperities of the interface, while at high viscosity, this ability decreases, resulting in the retention of voids. Void initiation is predicted by analyzing the temporal correlation between abnormal sound velocity change points and the degassing peak. A change point detected within 2-3 seconds of the degassing peak is considered the moment of void initiation. The propagation path is predicted based on analysis of the interface geometry and stress distribution, with voids preferentially propagating along directions of stress concentration. The final size prediction is calculated based on the empirical relationship between the decrease in sound velocity and the cavity volume: every 1% decrease in sound velocity corresponds to an increase in cavity volume of approximately 0.02mm. 3 The acoustic attenuation vector of the cavity development prediction refers to the energy loss vector caused by the cavity to the sound wave propagation after it is fully formed. It includes two components: scattering loss and absorption loss. Scattering loss mainly affects the high-frequency component, while absorption loss mainly affects the low-frequency component. For example, the final size of the cavity is predicted to be 0.15mm. 3 When the sound attenuation vector is [scattering loss: -3.2dB@7MHz, absorption loss: -1.8dB@3MHz]. This predictive computing process enables a leap from process monitoring to result prediction, and can estimate the impact of voids on product performance before they fully form, providing forward-looking information for quality control decisions.

[0044] Please continue reading Figure 1 Based on the correlation pattern of acoustic features and thermal features in the acoustic-thermal coupling feature set, defect identification and grading evaluation are performed in combination with the dynamically adjusted process tolerance threshold to obtain a spatially distributed defect grade map.

[0045] In one embodiment of the present invention, the defect identification and grading evaluation based on the correlation pattern of acoustic features and thermal features in the acoustic-thermal coupling feature set, combined with the dynamically adjusted process tolerance threshold, to obtain a spatially distributed defect grade map includes: Based on the interface difference pattern of the dual-interface multi-frequency acoustic velocity drift rate in the acoustic-thermal coupling feature set and the interlayer difference pattern of the dual-band thermal attenuation slope characterized by the filler gradient, acoustic-thermal feature cross-matching processing is performed to identify three defect types: interface voids, thickness standing waves, and filler settlement, and obtain defect type identification results; According to the real-time temperature, pressure and travel speed of the aluminum-based copper-clad laminate in the hot pressing zone, a process state compensation adjustment process is performed on the preset reference tolerance threshold to obtain a dynamically adjusted process tolerance threshold; The acoustic attenuation vector amplitude predicted by the cavity development at each defect location in the defect type identification result and the thermal hysteresis ratio value characterized by the sedimentation dynamics are quantitatively compared with the dynamically adjusted process tolerance threshold, and the defect degree is divided into three levels: mild, moderate, and severe to obtain a defect grading assessment result; The defect grading assessment results are spatially mapped according to the longitudinal coordinates of the panel surface, and connected domain analysis is performed on similar defects at adjacent positions to obtain a spatially distributed defect grade map.

[0046] The following is a detailed description of the steps involved in the above embodiment: The interface difference mode is identified by analyzing the difference in multi-frequency acoustic velocity drift rates at the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface. This difference is calculated by subtracting the drift rate of the lower interface from the drift rate of the upper interface using a digital subtractor. The interlayer difference mode is identified by comparing the thermal attenuation slope difference of the filler gradient in the 3-5μm band and the 8-12μm band. The difference is the upper layer attenuation slope minus the lower layer attenuation slope. Acoustic-thermal feature cross-matching is achieved using a decision tree algorithm: When the interface difference pattern indicates a difference in the acoustic velocity drift rate between the upper and lower interfaces greater than 2%, and the interlayer difference pattern indicates a difference in the thermal attenuation slope between the upper and lower layers less than 0.1 dB / s, the defect is identified as an interface void. When the interface difference pattern shows periodic variations in the acoustic velocity drift rate (amplitude exceeding 1.5%) and the interlayer difference pattern indicates a steady change in the thermal attenuation slope, the defect is identified as a thickness standing wave defect. When the interface difference pattern indicates a steady change in the acoustic velocity and a difference in the thermal attenuation slope greater than 0.3 dB / s, the defect is identified as a filler sedimentation defect. For example, if the acoustic velocity drift rate at a certain location is detected to be -3.8% for the upper interface and -1.2% for the lower interface, with a difference of -2.6%, and the thermal attenuation slopes of the upper layer are -0.9 dB / s and -0.95 dB / s for the lower layer, with a difference of 0.05 dB / s, the defect is identified as an interface void according to the decision tree. The defect type identification result is a dataset containing the defect type identifier, location coordinates, and confidence level. The acoustic-thermal feature cross-matching process utilizes the differentiated response mechanism of different physical fields to the same defect. The acoustic signal is sensitive to changes in interface state, and the thermal signal is sensitive to changes in material distribution. Through cross-validation, it can eliminate misjudgments of single physical field detection and significantly improve the accuracy and reliability of defect type identification.

[0047] The process state compensation adjustment process dynamically corrects the preset baseline tolerance threshold based on real-time monitored process parameters. Real-time temperature is monitored by an infrared thermal imager with a range of 130-180°C, pressure is monitored by a pressure sensor with a range of 2-8 MPa, and travel speed is monitored by an encoder with a range of 0.5-2.0 m / s. The compensation algorithm is implemented through multiple linear regression: the baseline tolerance threshold is multiplied by the product of the temperature compensation coefficient, the pressure compensation coefficient, and the speed compensation coefficient to obtain the dynamically adjusted process tolerance threshold. The temperature compensation coefficient is calculated by increasing by 5% for every 10°C increase, the pressure compensation coefficient is calculated by decreasing by 3% for every 1 MPa increase, and the speed compensation coefficient is calculated by decreasing by 2% for every 0.1 m / s increase. For example, when the real-time temperature is 160°C, the pressure is 5 MPa, and the travel speed is 1.2 m / s, the compensation coefficients relative to the standard operating conditions (150°C, 4 MPa, 1.0 m / s) are 1.05, 0.97, and 0.96, respectively, for a combined compensation coefficient of 1.05 × 0.97 × 0.96 = 0.978. If the preset baseline tolerance threshold is -5.0 dB, the dynamically adjusted process tolerance threshold is -5.0 × 0.978 = -4.89 dB. The dynamically adjusted process tolerance threshold refers to a defect judgment standard that is modified based on real-time process conditions, adapting to the impact of process fluctuations on defect detection sensitivity. This process state compensation adjustment avoids the misjudgment of fixed thresholds under varying process conditions, ensuring that the defect assessment standard always matches the actual process state, improving the inspection system's adaptability to process fluctuations and the stability of the judgment results.

[0048] Quantitative comparison processing assesses defect severity by numerically comparing defect characteristic parameters with dynamically adjusted process tolerance thresholds. The acoustic attenuation vector amplitude of the void development prediction and the thermal hysteresis ratio of the sedimentation dynamics characterization are compared with the corresponding dynamically adjusted process tolerance thresholds. A minor defect is determined when the absolute value of the characteristic parameter is less than 50% of the threshold; a moderate defect is determined when the absolute value of the characteristic parameter is between 50% and 100% of the threshold; and a severe defect is determined when the absolute value of the characteristic parameter exceeds the threshold. For example, the acoustic attenuation vector amplitude of the void development prediction at a certain interface void location is -3.2dB, and the corresponding dynamically adjusted process tolerance threshold is -4.89dB. The ratio is 3.2 / 4.89 = 0.65, which falls within the 50%-100% range and is classified as a moderate defect. The thermal hysteresis ratio of the sedimentation dynamics characterization at a filler sedimentation location is 0.28, and the corresponding threshold is 0.20. The ratio is 0.28 / 0.20 = 1.4, which exceeds 100% and is classified as a severe defect. Defect grading assessment results are comprehensive assessment data that includes information on defect location, type, and grade. The three-tier grading system is based on a quantitative assessment of the degree to which defects impact product performance: minor defects have a performance impact of less than 10%, moderate defects have an impact of 10%-30%, and severe defects have an impact of more than 30%. This quantitative comparison transforms qualitative defect identification into a quantitative grade assessment, providing a clear numerical basis for quality control decisions. The grading system is directly linked to product performance requirements, ensuring the engineering practicality of the assessment results.

[0049] Spatial mapping visualizes the defect grading results by converting their location coordinates into a two-dimensional spatial distribution map of the panel surface. The panel's vertical coordinate serves as the X-axis, and the defect grade is color-coded: minor defects appear green, moderate defects appear yellow, and severe defects appear red. Connected domain analysis uses image processing algorithms to identify similar defect regions in adjacent locations. Adjacent pixels (spatial distances less than 5mm) with the same defect type are grouped into a connected domain. The connected domain labeling algorithm uses the eight-connectivity rule to calculate the area, length, and shape parameters of each connected domain. For example, six medium interface void defects detected within the 0.15-0.25m section of the panel surface, all spaced less than 3mm apart, are identified by connected domain analysis as a single strip-shaped defect region with an area of ​​100mm² and a length of 80mm. A spatially distributed defect grade map is a pseudo-color visualization of the defect distribution. Each pixel in the image corresponds to a spatial location on the panel surface, and the pixel color indicates the defect grade at that location. This spatial mapping and connected domain analysis process converts discrete defect detection data into intuitive spatial distribution images, making it easier for operators to quickly identify the spatial distribution patterns and concentrated areas of defects. At the same time, connected domain analysis can identify the spatial correlation of defects, providing spatial dimension information support for defect cause analysis and process optimization.

[0050] In one embodiment of the present invention, the acoustic-thermal feature cross-matching process is performed based on the interface difference pattern of the dual-interface multi-frequency acoustic velocity drift rate and the interlayer difference pattern of the dual-band thermal attenuation slope characterized by the filler gradient in the acoustic-thermal coupling feature set to identify three defect types: interface voids, thickness standing waves, and filler settlement. The defect type identification results obtained include: The multi-frequency sound velocity drift rate of the dual interface in the acoustic-thermal coupling feature set is subjected to interface difference pattern recognition processing to extract the sound velocity drift difference characteristics of the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface; the dual-band thermal attenuation slope characterized by the filler gradient in the acoustic-thermal coupling feature set is subjected to interlayer difference pattern recognition processing to extract the thermal attenuation difference characteristics of the 3-5μm band and the 8-12μm band; According to the coupled combination mode of the acoustic velocity drift difference feature and the thermal attenuation difference feature, pattern matching processing is performed with the preset interface void feature template, thickness standing wave feature template and filler sedimentation feature template to determine the dominant defect type at each position and obtain the defect type identification result.

[0051] The following is a detailed description of the steps involved in the above embodiment: Interface difference pattern recognition uses difference calculation and statistical analysis algorithms to extract the acoustic velocity drift difference characteristics at the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface. The system first separates the dual-interface multi-frequency acoustic velocity drift rate data from the acoustic-thermal coupling feature set and categorizes it into upper and lower interface data based on interface location. The difference calculation uses a digital subtractor to subtract the upper interface acoustic velocity drift rate from the lower interface acoustic velocity drift rate to obtain the interface difference value. A statistical analysis algorithm calculates the mean, standard deviation, and trend of the difference values ​​to identify the difference pattern type: a stable difference pattern is identified when the mean difference is greater than 1.5% and the standard deviation is less than 0.3%; an oscillating difference pattern is identified when the difference value exhibits periodic fluctuations with an amplitude exceeding 2%; and a trending difference pattern is identified when the difference trend is monotonically increasing or decreasing with a slope exceeding 0.5% / second. The acoustic velocity drift difference characteristic is a feature vector that quantitatively describes the degree and pattern of the difference in acoustic velocity variation between the two interfaces. It includes the difference amplitude, trend, and stability indicators. The interlayer difference pattern recognition process utilizes the same algorithmic framework to process the dual-band thermal attenuation slopes of the filler gradient representation: the 8-12μm band thermal attenuation slope is subtracted from the 3-5μm band thermal attenuation slope to calculate the difference in interlayer thermal diffusion characteristics. The thermal attenuation difference signature is a characteristic vector that describes the difference in filler distribution and the change in thermal diffusion characteristics between the upper and lower layers. For example, the acoustic velocity drift rate at a certain location is -3.2% at the upper interface and -1.8% at the lower interface, a difference of -1.4% with a standard deviation of 0.2%, which is identified as a stable difference pattern. The thermal attenuation slopes of the upper layer and -0.9dB / s at the lower layer are -1.3dB / s, a difference of 0.4dB / s, which is identified as a positive interlayer difference. This difference pattern recognition process converts multi-dimensional physical response data into structured difference signatures that can independently describe material state changes at different interfaces and depths, providing precise feature input for subsequent defect type identification.

[0052] Coupled combined pattern analysis achieves feature fusion by mapping the acoustic velocity drift difference and thermal attenuation difference features into a two-dimensional feature space. The system uses the acoustic velocity drift difference feature as the X-axis coordinate and the thermal attenuation difference feature as the Y-axis coordinate to form two-dimensional feature points. The preset interface void feature template, thickness standing wave feature template, and filler sedimentation feature template are pre-stored standard feature patterns. Each template defines the typical distribution area of ​​the corresponding defect type in the two-dimensional feature space. The interface void feature template corresponds to the feature space region where the acoustic velocity drift difference is greater than 1.2% and the thermal attenuation difference is less than 0.2dB / s; the thickness standing wave feature template corresponds to the region where the acoustic velocity drift difference exhibits periodic changes and the thermal attenuation difference varies steadily; and the filler sedimentation feature template corresponds to the region where the acoustic velocity drift difference is less than 0.8% and the thermal attenuation difference is greater than 0.3dB / s. Pattern matching processing uses the nearest neighbor classification algorithm to calculate the Euclidean distance between the measured feature points and each feature template, and selects the template with the smallest distance as the matching result. For example, the coordinates of a feature point at a certain location are (-1.4%, 0.1dB / s), the distance to the interface void feature template is 0.28, the distance to the thickness standing wave feature template is 0.65, and the distance to the filler sedimentation feature template is 0.82. Therefore, the dominant defect type is determined to be an interface void. The defect type identification result refers to the identification data set that includes the defect type identification, matching confidence, and location coordinates. This pattern matching processing realizes the automatic identification of defect types through the collaborative analysis of multi-physical field features, avoiding the limitations of single physical quantity discrimination, significantly improving the accuracy and reliability of defect identification, and especially significantly enhancing the recognition ability of composite defects and boundary defects.

[0053] The above describes the film forming method of the aluminum-based copper-clad laminate according to the embodiment of the present invention. The following describes the film forming device of the aluminum-based copper-clad laminate according to the embodiment of the present invention. Figure 2 An embodiment of the film forming device of the aluminum-based copper-clad laminate according to the present invention includes: The excitation applying module 101 is used to simultaneously apply broadband acoustic excitation and dual-band thermal excitation when the aluminum-based copper-clad laminate enters the hot pressing zone; The response acquisition module 102 is used to acquire the acoustic propagation response and thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone to obtain an acoustic-thermal synchronous response sequence; The feature extraction module 103 is used to perform differential transformation and time-frequency domain analysis on the acoustic-thermal synchronous response sequence and the defect-free reference spectrum, extract the multi-frequency sound velocity drift rate, the acoustic attenuation vector, the dual-band thermal attenuation slope and the thermal hysteresis ratio, and obtain the acoustic-thermal coupling feature set; The defect recognition module 104 is used to perform defect recognition and grading evaluation based on the correlation pattern of acoustic features and thermal features in the acoustic-thermal coupling feature set in combination with the dynamically adjusted process tolerance threshold to obtain a spatially distributed defect grade map.

[0054] above Figure 2 The aluminum-based copper-clad laminate film forming device in the embodiment of the present invention is described in detail from the perspective of modular functional entities. The aluminum-based copper-clad laminate film forming equipment in the embodiment of the present invention is described in detail from the perspective of hardware processing.

[0055] Figure 3 The figure is a schematic diagram of the structure of an aluminum-based copper-clad laminate film forming apparatus provided by an embodiment of the present invention. The aluminum-based copper-clad laminate film forming apparatus 200 may vary significantly depending on configuration or performance. It may include one or more processors 210 (e.g., one or more processors), memory 220, and one or more storage media 230 (e.g., one or more mass storage devices) storing application programs 233 or data 232. The memory 220 and storage media 230 may be either transient or persistent storage. The program stored in the storage medium 230 may include one or more modules (not shown), each of which may include a series of instructions and operations within the aluminum-based copper-clad laminate film forming apparatus 200. Furthermore, the processor 210 may be configured to communicate with the storage medium 230, executing the series of instructions and operations stored in the storage medium 230 on the aluminum-based copper-clad laminate film forming apparatus 200 to implement the steps of the aluminum-based copper-clad laminate film forming method described above.

[0056] The aluminum-based copper-clad laminate film forming device 200 may further include one or more power supplies 240, one or more wired or wireless network interfaces 250, one or more input and output interfaces 260, and / or one or more operating systems 231, such as Windows Server, Mac OS X, Unix, Linux, FreeBSD, etc. It will be understood by those skilled in the art that Figure 3 The structure of the aluminum-based copper-clad laminate film forming equipment shown does not constitute a limitation on the aluminum-based copper-clad laminate film forming equipment provided by the present invention, and may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0057] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium. The computer-readable storage medium stores instructions. When the instructions are run on a computer, the computer executes the steps of the aluminum-based copper-clad laminate film forming method.

[0058] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the above-described systems, devices, and units can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0059] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes various media that can store program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0060] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made by using the contents of the present invention description and drawings under the inventive concept of the present invention, or direct / indirect application in other related technical fields are included in the patent protection scope of the present invention.

Claims

1. A method for laminating an aluminum-based copper-clad laminate, characterized in that: include: Broadband acoustic excitation and dual-band thermal excitation are applied simultaneously when the aluminum-based copper-clad laminate enters the hot pressing zone; Acquiring an acoustic propagation response and a thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone to obtain an acoustic-thermal synchronous response sequence; Performing differential transformation and time-frequency domain analysis on the acoustic-thermal synchronous response sequence and the defect-free reference spectrum, extracting the multi-frequency sound velocity drift rate, acoustic attenuation vector, dual-band thermal attenuation slope, and thermal hysteresis ratio, and obtaining an acoustic-thermal coupling feature set; Based on the correlation pattern of acoustic features and thermal features in the acoustic-thermal coupling feature set, defect identification and grading evaluation are performed in combination with a dynamically adjusted process tolerance threshold to obtain a spatially distributed defect grade map.

2. The film forming method of the aluminum-based copper-clad laminate according to claim 1, characterized in that: The method of simultaneously applying broadband acoustic excitation and dual-band thermal excitation when the aluminum-based copper-clad laminate enters the hot pressing zone includes: According to the acoustic impedance ratio of the aluminum substrate to the copper foil in the aluminum-based copper-clad laminate, a 2-8 MHz linear frequency modulation acoustic pulse is divided into a 2-4 MHz low frequency band and a 6-8 MHz high frequency band, the low frequency band is amplitude-enhanced to detect thickness standing wave defects, and the high frequency band is amplitude-enhanced to detect interface void defects, thereby obtaining a defect-selective broadband acoustic excitation signal; According to the thickness of the insulating dielectric layer and the filler content in the aluminum-based copper-clad laminate, the 3-5 μm band infrared pulse is pulse-width modulated to detect the shallow filler distribution, and the 8-12 μm band infrared pulse is intensity modulated to detect the deep filler distribution, thereby obtaining a dual-band thermal excitation signal for layered detection; The defect-selective broadband acoustic excitation signal and the layered detection dual-band thermal excitation signal are applied synchronously in the resin gelation window to ensure that the excitation propagation path is consistent along the acoustic and thermal conduction channel between the copper surface and the aluminum surface.

3. The film forming method of the aluminum-based copper-clad laminate according to claim 2, characterized in that: The method divides the 2-8 MHz linear frequency modulation acoustic pulse into a 2-4 MHz low frequency band and a 6-8 MHz high frequency band according to the acoustic impedance ratio of the aluminum substrate to the copper foil in the aluminum-based copper clad laminate, performs amplitude enhancement on the low frequency band to detect thickness standing wave defects, and performs amplitude enhancement on the high frequency band to detect interface void defects, thereby obtaining a defect-selective broadband acoustic excitation signal, including: According to the thickness specification parameters and material model of the aluminum-based copper-clad laminate, the corresponding aluminum substrate acoustic impedance value and copper foil acoustic impedance value are obtained from a preset acoustic impedance database, and a ratio operation is performed to obtain a standard acoustic impedance ratio; According to the ratio range of the standard acoustic impedance ratio, the thickness standing wave sensitivity weight coefficient is set for the 2-4 MHz low frequency band, and the interface void sensitivity weight coefficient is set for the 6-8 MHz high frequency band. After amplitude modulation processing, they are respectively synthesized into defect-selective broadband acoustic excitation signals.

4. The film forming method of the aluminum-based copper-clad laminate according to claim 1, characterized in that: The collecting of the acoustic propagation response and thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone to obtain an acoustic-thermal synchronous response sequence includes: performing path separation processing on the direct wave signal, the interface reflection wave signal, and the mode conversion wave signal in the acoustic propagation response according to the traveling speed of the aluminum-based copper-clad laminate and the length of the hot pressing zone to obtain multipath acoustic response data; According to the temperature distribution of the aluminum-based copper-clad laminate in the hot pressing zone, temperature compensation processing is performed on the 3-5 μm band signal and the 8-12 μm band signal in the thermal diffusion response to obtain corrected dual-band thermal response data; Performing spatiotemporal calibration processing on the multipath acoustic response data and the corrected dual-band thermal response data according to the longitudinal coordinates of the panel surface, establishing a correspondence between the response signal and the spatial position, and obtaining spatiotemporally correlated acoustic-thermal response data; The temporally and spatially correlated acoustic-thermal response data are serialized and arranged according to a uniform sampling time interval to obtain an acoustic-thermal synchronous response sequence.

5. The film forming method of the aluminum-based copper-clad laminate according to claim 1, characterized in that: The acoustic-thermal synchronous response sequence and the defect-free reference spectrum are subjected to differential transformation and time-frequency domain analysis to extract the multi-frequency sound velocity drift rate, acoustic attenuation vector, dual-band thermal attenuation slope and thermal hysteresis ratio to obtain the acoustic-thermal coupling feature set, including: The acoustic-thermal synchronous response sequence is divided into a flow state response section, a gelation state response section, and a solidification state response section according to the resin gelation window, and differential transformation processing is performed on each of them with the defect-free reference spectrum of the corresponding state to obtain a staged acoustic difference spectrum and a staged thermal difference spectrum; Performing interface separation and analysis on the phased acoustic difference spectrum according to its frequency characteristics, extracting the acoustic velocity drift trends at the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface, and obtaining the dual-interface multi-frequency acoustic velocity drift rate; According to the variation pattern of the dual-interface multi-frequency acoustic velocity drift rate at different gelation window stages, the defect evolution trajectory is analyzed and processed in combination with the interface void formation mechanism to obtain the acoustic attenuation vector for void development prediction; A hierarchical identification process is performed on the thermal diffusion anomaly caused by filler sedimentation in the staged thermal differential spectrum. By comparing the attenuation differences between the 3-5μm band signal and the 8-12μm band signal at each gelation window stage, a dual-band thermal attenuation slope representing the filler gradient and a thermal hysteresis ratio representing the sedimentation dynamics are obtained. The dual-interface multi-frequency acoustic velocity drift rate, the acoustic attenuation vector for cavity development prediction, the dual-band thermal attenuation slope characterized by filler gradient, and the thermal hysteresis ratio characterized by sedimentation dynamics are subjected to multi-dimensional correlation fusion processing to obtain the acoustic-thermal coupling feature set.

6. The film forming method of the aluminum-based copper-clad laminate according to claim 5, characterized in that: The defect evolution trajectory analysis is performed based on the change pattern of the dual-interface multi-frequency acoustic velocity drift rate at different gelation window stages in combination with the interface void formation mechanism to obtain the acoustic attenuation vector for void development prediction, including: The evolution trajectory of the dual-interface multi-frequency sound velocity drift rate is reconstructed according to the three gelation window stages of the flow state, gelation state, and solidification state, and the difference in the sound velocity change trend of the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface is analyzed to obtain the interface acoustic evolution trajectory data; According to the abnormal change points and amplitudes of the sound velocity in the interface acoustic evolution trajectory data, combined with the resin degassing process and the interface infiltration mechanism, the cavity initiation moment, expansion path and final size are predicted and calculated to obtain the acoustic attenuation vector for cavity development prediction.

7. The film forming method of the aluminum-based copper-clad laminate according to claim 1, characterized in that: The defect identification and grading evaluation is performed based on the correlation pattern of the acoustic features and thermal features in the acoustic-thermal coupling feature set, combined with the dynamically adjusted process tolerance threshold, to obtain a spatially distributed defect grade map, including: Based on the interface difference pattern of the dual-interface multi-frequency acoustic velocity drift rate in the acoustic-thermal coupling feature set and the interlayer difference pattern of the dual-band thermal attenuation slope characterized by the filler gradient, acoustic-thermal feature cross-matching processing is performed to identify three defect types: interface voids, thickness standing waves, and filler settlement, and obtain defect type identification results; According to the real-time temperature, pressure and travel speed of the aluminum-based copper-clad laminate in the hot pressing zone, a process state compensation adjustment process is performed on the preset reference tolerance threshold to obtain a dynamically adjusted process tolerance threshold; The acoustic attenuation vector amplitude predicted by the cavity development at each defect location in the defect type identification result and the thermal hysteresis ratio value characterized by the sedimentation dynamics are quantitatively compared with the dynamically adjusted process tolerance threshold, and the defect degree is divided into three levels: mild, moderate, and severe to obtain a defect grading assessment result; The defect grading assessment results are spatially mapped according to the longitudinal coordinates of the panel surface, and connected domain analysis is performed on similar defects at adjacent positions to obtain a spatially distributed defect grade map.

8. The film forming method of the aluminum-based copper-clad laminate according to claim 7, characterized in that: According to the interface difference mode of the dual-interface multi-frequency acoustic velocity drift rate and the interlayer difference mode of the dual-band thermal attenuation slope characterized by the filler gradient in the acoustic-thermal coupling feature set, acoustic-thermal feature cross-matching processing is performed to identify three defect types: interface voids, thickness standing waves, and filler settlement. The defect type identification results are obtained, including: The multi-frequency sound velocity drift rate of the dual interface in the acoustic-thermal coupling feature set is subjected to interface difference pattern recognition processing to extract the sound velocity drift difference characteristics of the copper foil-insulating dielectric layer interface and the insulating dielectric layer-aluminum substrate interface; the dual-band thermal attenuation slope characterized by the filler gradient in the acoustic-thermal coupling feature set is subjected to interlayer difference pattern recognition processing to extract the thermal attenuation difference characteristics of the 3-5μm band and the 8-12μm band; According to the coupled combination mode of the acoustic velocity drift difference feature and the thermal attenuation difference feature, pattern matching processing is performed with the preset interface void feature template, thickness standing wave feature template and filler sedimentation feature template to determine the dominant defect type at each position and obtain the defect type identification result.

9. A film forming device for aluminum-based copper-clad laminate, characterized in that: The aluminum-based copper-clad laminate film forming device adopts the aluminum-based copper-clad laminate film forming method according to any one of claims 1 to 8, and the aluminum-based copper-clad laminate film forming device comprises: An excitation application module is used to simultaneously apply broadband acoustic excitation and dual-band thermal excitation when the aluminum-based copper-clad laminate enters the hot pressing zone; a response acquisition module, configured to acquire the acoustic propagation response and thermal diffusion response of the aluminum-based copper-clad laminate during its movement in the hot pressing zone, and obtain an acoustic-thermal synchronous response sequence; a feature extraction module for performing differential transformation and time-frequency domain analysis on the acoustic-thermal synchronous response sequence and the defect-free reference spectrum, extracting the multi-frequency sound velocity drift rate, the acoustic attenuation vector, the dual-band thermal attenuation slope, and the thermal hysteresis ratio to obtain an acoustic-thermal coupling feature set; The defect recognition module is used to perform defect recognition and grade evaluation based on the correlation pattern of acoustic features and thermal features in the acoustic-thermal coupling feature set and in combination with the dynamically adjusted process tolerance threshold to obtain a spatially distributed defect grade map.

10. An aluminum-based copper-clad laminate film forming device, characterized in that: The aluminum-based copper-clad laminate film forming device includes: a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the aluminum-based copper-clad laminate film forming equipment to perform the steps of the aluminum-based copper-clad laminate film forming method according to any one of claims 1 to 8.

Images