Integrated intelligent stripping system and control method of heat sink assembly

Abstract

The application provides an integrated intelligent stripping system and control method for a heat sink assembly, which comprises a stripping detection unit, an intelligent control unit, a stripping execution unit and a recovery unit; the stripping detection unit collects distributed potential data and real-time thickness data on the surface of the heat sink assembly; the intelligent control unit generates a stripping process scheme for the heat sink assembly; whether an anti-corrosion linkage mechanism is triggered is determined according to the corrosion potential data and the coating thickness data; the stripping execution unit sprays stripping according to the differential stripping process scheme; and the recovery unit cyclically purifies and fractionally recovers metal ions from the stripping waste liquid. Through the linkage and differential process adaptation of multiple units, the application realizes the integrated cooperation of stripping detection, process execution and waste liquid recovery of the heat sink assembly, greatly improves the adaptability of the stripping process, the controllability of the stripping process and the environmental protection of resource utilization, and completes the integrated intelligent stripping of the heat sink assembly.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of heat sink plating, in particular, to an integrated intelligent plating removal system of a heat sink assembly and a control method. BACKGROUND

[0002] With the continuous improvement of the integration and power density of electronic equipment, the performance of the surface plating layer of the heat sink assembly as a core heat dissipation component directly determines the heat dissipation efficiency and the service life of the equipment. In the production, repair and recycling links of the heat sink assembly, the plating removal process is a key process for realizing plating layer updating and substrate recycling.

[0003] The existing heat sink assembly plating removal technology has various defects. For example, the existing technology mostly adopts unified solidification plating removal process and parameters, and does not dynamically adjust according to the adhesion grade, material composition and surface structure difference of the heat sink assembly plating layer. Moreover, the existing plating removal relies on manual observation or offline detection to judge the progress, which is difficult to stop work in time at the plating layer removal critical state, resulting in substrate damage and reducing the assembly reuse rate. In addition, the equipment integration of the plating removal device is low, and manual transfer of the assembly is required to complete the process flow, which not only prolongs the production rhythm, but also easily affects the plating removal precision due to positioning deviation, and the coordination of each unit is poor, further reducing the efficiency. SUMMARY

[0004] Based on the problems existing in the prior art, the present application provides an integrated intelligent plating removal system and control method of a heat sink assembly. The specific scheme is as follows:

[0005] An integrated intelligent plating removal system of a heat sink assembly, comprising a plating removal detection unit, an intelligent control unit, a plating removal execution unit and a recycling unit;

[0006] The plating removal detection unit comprises a potential detection assembly and a thickness measurement assembly, which are used to collect distributed potential data of the surface of the heat sink assembly through the potential detection assembly and collect real-time thickness data of the heat sink assembly through the thickness measurement assembly;

[0007] The intelligent control unit is used to obtain multi-modal data of the heat sink assembly to be plated and generate a plating removal process scheme of the heat sink assembly according to the multi-modal data; determine the plating removal progress according to the corrosion potential data and the plating layer thickness data, and determine whether to trigger the anti-corrosion linkage mechanism according to the plating removal progress;

[0008] The plating removal execution unit comprises a pretreatment assembly, a spraying assembly and a liquid preparation assembly, which are used to perform pretreatment through the pretreatment assembly according to the differential plating removal process scheme, adjust the concentration of the plating removal liquid through the liquid preparation assembly, and perform targeted spraying of the plating removal liquid through the spraying assembly to perform spraying plating removal;

[0009] The recycling unit is used to recycle and purify the plating removal waste liquid and grade the metal ions.

[0010] In some specific embodiments, the recovery unit comprises a metal separation membrane assembly and a gradient electric field module;

[0011] The gradient electric field module realizes the graded recovery of different metal ions in the plating layer of the heat sink assembly by sequentially switching different gradient voltages.

[0012] In some specific embodiments, the multi-modal data comprises appearance image data, three-dimensional size data, and plating layer spectrum data of the heat sink assembly;

[0013] The intelligent control unit obtains analysis results of each modality by performing semantic segmentation, contour fitting, and spectrum matching on the appearance image data, three-dimensional size data, and plating layer spectrum data respectively through the built-in multi-modal fusion recognition model; obtains a fusion determination result by fusing the analysis results of each modality using a weighted fusion algorithm; and determines the plating layer adhesion grade of the heat sink assembly according to the fusion determination result.

[0014] In some specific embodiments, pixel-level semantic segmentation is performed on the appearance image data to extract structural features of the heat sink assembly including fins, blind holes, and corner gaps, and to determine the key areas of the plating layer to be processed; surface reconstruction and thickness fitting are performed on the three-dimensional size data to obtain the initial thickness distribution of the plating layer and the surface contour parameters of the base material; and feature peak identification and matching are performed on the plating layer spectrum data to determine the material composition and proportion of each component of the plating layer.

[0015] In some specific embodiments, the intelligent control unit is also built-in with a process parameter mapping library, which pre-stores the parameter intervals of plasma bombardment power, stripping solution concentration, and spraying duration corresponding to different plating layer materials and adhesion grades; the intelligent control unit retrieves the matching parameters from the process parameter mapping library according to the fusion determination result of the multi-modal fusion recognition model, and then obtains the stripping process scheme.

[0016] In some specific embodiments, the pretreatment assembly comprises a plasma generating device and a chemical cleaning device;

[0017] For high-adhesion plating layers, a segmented pulse bombardment pretreatment is performed using the plasma generating device, and the spraying assembly sprays high-concentration stripping solution in a high-pressure directional mode; for medium-adhesion plating layers, a dilute acid etching pretreatment is performed using the chemical cleaning device, and the spraying assembly sprays medium-concentration stripping solution in a medium-pressure uniform mode; for low-adhesion plating layers, a dilute acid etching pretreatment is performed using the chemical cleaning device, and the spraying assembly sprays low-concentration stripping solution in a low-pressure mist mode.

[0018] In some specific embodiments, the stripping execution unit further comprises a mechanical arm, and the potential detection assembly, thickness measurement assembly, plasma generating device, and chemical cleaning device are integrated on the mechanical arm and arranged in a central symmetric modular layout.

[0019] In some embodiments, the plasma generating device is arranged in a ring radiation layout, the spray nozzles of the chemical cleaning device are arranged outside the plasma generating device, and the spray nozzles are directed in the same direction as the bombardment direction of the plasma generating device; the detection probe of the thickness measuring assembly is located between the chemical cleaning device and the plasma generating device, and is isolated from the plasma generating device by a heat-shielding and light-shielding support; the array of micro reference electrodes of the potential detection assembly is embedded in the gap between the spray nozzles of the chemical cleaning device.

[0020] In some embodiments, the anti-corrosion linkage mechanism includes: controlling the spray assembly to switch to a mist spray mode, shortening the spray time of the easily-corroded area, controlling the liquid preparation assembly to reduce the active concentration of the stripping solution, and controlling the recovery unit to output a weak reducing current and form a passivation film on the surface of the substrate.

[0021] A control method of an integrated intelligent stripping system, for controlling the integrated intelligent stripping system of any one of the preceding embodiments; the control method comprises the following steps:

[0022] Obtaining multi-modal data of the heat sink assembly to be stripped, and the intelligent control unit generates a stripping process scheme for the heat sink assembly according to the multi-modal data;

[0023] The stripping execution unit performs pretreatment through the pretreatment assembly according to the stripping process scheme, adjusts the concentration of the stripping solution through the liquid preparation assembly, and then performs targeted spray stripping of the stripping solution through the spray assembly;

[0024] The recovery unit recovers and purifies the stripping waste liquid;

[0025] The potential detection assembly collects distributed potential data on the surface of the heat sink assembly, and the thickness measuring assembly collects real-time thickness data of the heat sink assembly;

[0026] The intelligent control unit determines the stripping progress according to the corrosion potential data and the coating thickness data, and determines whether to trigger the anti-corrosion linkage mechanism according to the stripping progress.

[0027] Beneficial effects: The application provides an integrated intelligent stripping system and control method for heat sink assemblies. Through the coordinated linkage of multiple units and the adaptation of differentiated processes, the integrated cooperation of heat sink assembly stripping detection, process execution, and waste liquid recovery is achieved, greatly improving the adaptability of the stripping process, the controllability of the stripping process, and the environmental friendliness of resource utilization, and completing the integrated intelligent stripping of the heat sink assembly. It can generate accurate differentiated stripping schemes according to the adhesion grade of the plating layer, material composition, and structural characteristics, and can monitor the stripping process in real time and trigger the corrosion prevention mechanism to avoid substrate damage. At the same time, through the recycling unit, waste liquid recycling and metal ion grading recovery are realized, solving the problems of insufficient process targeting, process monitoring, and low equipment integration in the prior art. It provides technical support for efficient repair and substrate recycling of heat sink assemblies, effectively guarantees the stripping quality and production efficiency of heat sink assemblies.

[0028] In order to make the above-mentioned purposes, features and advantages of the present application more obvious and easy to understand, the following preferred embodiments are described in detail below, and the accompanying drawings are described as follows. BRIEF DESCRIPTION OF DRAWINGS

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as a limitation on the scope. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0030] Figure 1 is a schematic diagram of an integrated intelligent stripping system module of the present application;

[0031] Figure 2 is a schematic diagram of the principle of the integrated intelligent stripping system of the present application;

[0032] Figure 3 is a schematic diagram of the multi-modal data processing flow of the present application;

[0033] Figure 4 is a schematic diagram of the construction flow of the stripping process scheme of the present application;

[0034] Figure 5 is a schematic diagram of the control method flow of the present application.

[0035] Reference signs: 1 - stripping detection unit; 2 - intelligent control unit; 3 - stripping execution unit; 4 - recycling unit; 11 - potential detection assembly; 12 - thickness measurement assembly; 31 - pretreatment assembly; 32 - jetting assembly; 33 - liquid preparation assembly; 41 - metal separation membrane assembly; 42 - gradient electric field module. DETAILED DESCRIPTION

[0036] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application. Based on the embodiments in the present application, all other embodiments obtained by a person of ordinary skill in the art without creative work fall within the scope of the present application.

[0037] The present application provides an integrated intelligent stripping system for a heat sink assembly. Through the deep cooperation of each unit, the stripping of the heat sink assembly is changed from passive execution to active regulation. The problems of detection fragmentation, traceability difficulty, and insufficient precision in traditional processes are solved. The intelligent control of the whole process ensures the stripping quality and the safety of the base material, while considering resource recycling and environmental protection requirements. The integrated intelligent stripping system provides a complete control scheme support for efficient repair and high-quality production of the heat sink assembly. Figure 1 The principle is shown in the accompanying Figure 2 The specific scheme is as follows:

[0038] An integrated intelligent stripping system for a heat sink assembly, comprising a stripping detection unit 1, an intelligent control unit 2, a stripping execution unit 3, and a recovery unit 4.

[0039] The stripping detection unit 1 comprises a potential detection assembly 11 and a thickness measurement assembly 12, which are used to collect distributed potential data on the surface of the heat sink assembly through the potential detection assembly 11 and to collect real-time thickness data of the heat sink assembly through the thickness measurement assembly 12.

[0040] The intelligent control unit 2 is used to obtain multi-modal data of the heat sink assembly to be stripped and to generate a stripping process scheme for the heat sink assembly according to the multi-modal data. The stripping process is determined according to the corrosion potential data and the coating thickness data, and the anti-corrosion linkage mechanism is triggered according to the stripping process.

[0041] The stripping execution unit 3 comprises a pretreatment assembly 31, a spraying assembly 32, and a liquid preparation assembly 33, which are used to perform pretreatment through the pretreatment assembly 31 according to the differential stripping process scheme, to adjust the concentration of the stripping liquid through the liquid preparation assembly 33, and to perform targeted spraying stripping through the spraying assembly 32.

[0042] The recovery unit 4 is used for recycling and purifying the stripping waste liquid and grading the recovery of metal ions.

[0043] The integrated intelligent stripping system of the present application breaks the problems of detection, execution, and recovery fragmentation in traditional stripping processes through the cooperation of the stripping detection unit 1, the intelligent control unit 2, the stripping execution unit 3, and the recovery unit 4, and realizes the whole-process intelligent closed-loop control from feature recognition of the heat sink assembly to stripping process adaptation, to process monitoring and waste liquid treatment.

[0044] The stripping detection unit 1, as the sensing core of the system, is responsible for acquiring key data of the stripping process in real time. Its included potential detection component 11 and thickness measurement component 12 achieve precise monitoring from two dimensions: electrochemical characteristics and physical thickness. The potential detection component 11 typically employs a micro-reference electrode array structure. Utilizing the electrochemical potential difference between the electrode and the coating / substrate on the heat sink surface, it simultaneously collects distributed potential data from different regions. This data directly reflects the degree of coating dissolution and the exposure state of the substrate. When the potential value approaches the critical potential of the substrate, it can predict that the substrate is about to be corroded. The thickness measurement component 12 uses laser confocal technology. It scans the surface of the heat sink component by emitting a visible light laser beam. The laser beam is reflected from the coating surface and captured by the receiver. Based on the laser round-trip time difference and light intensity changes, combined with optical imaging algorithms, the real-time coating thickness is calculated.

[0045] The intelligent control unit 2 is responsible for coordinating data processing, process generation, and process control. First, the intelligent control unit 2 is responsible for generating the process plan before stripping. By acquiring multimodal data of the heat sink component to be stripped, it uses a built-in multimodal fusion recognition model to perform collaborative analysis on the data, and finally accurately determines the coating adhesion level, material composition, and surface structure characteristics of the heat sink component. Then, it generates a suitable differentiated stripping process plan. This process solves the problem that traditional single data recognition cannot fully reflect the coating characteristics, ensuring that the process plan is highly matched with the actual situation of the heat sink component.

[0046] During the stripping process, the intelligent control unit 2 is responsible for process control. It determines the stripping progress by receiving distributed potential data and real-time thickness data transmitted by the stripping detection unit 1 in real time. For example, by combining the built-in corrosion potential judgment model, the actual data is compared with the preset substrate critical potential threshold and the lower limit thickness value of the coating to dynamically determine the stripping progress. When the data reaches the threshold (such as the potential approaching the critical value or the thickness dropping to the lower limit), the anti-corrosion linkage mechanism is immediately triggered to avoid excessive corrosion of the substrate and achieve precise control and stop of the stripping process.

[0047] The stripping execution unit 3 completes the specific stripping operation according to the process plan generated by the intelligent control unit 2. Its pretreatment component 31, spraying component 32, and solution preparation component 33 form a collaborative execution chain. The pretreatment component 31 provides suitable pretreatment methods for coatings with different adhesion levels, weakening the adhesion between the coating and the substrate through physical or chemical action, laying the foundation for subsequent stripping. The solution preparation component 33 uses microfluidic mixing technology to precisely control the feed rate and mixing ratio of different raw solutions, achieving real-time adjustment of the stripping solution concentration to ensure that the concentration is completely consistent with the process plan requirements. It also has a built-in pressure buffer structure to prevent concentration fluctuations from affecting the stripping effect. The spraying component 32 adopts a multi-nozzle array structure, where each nozzle can be independently controlled for opening, closing, and spraying angle. Based on the spraying pressure and angle requirements in the process plan, it performs targeted spraying on the areas to be treated on the heat sink component. Especially for complex structural areas such as fins and blind holes, it can achieve precise coverage by adjusting the nozzle angle, solving the blind spot problem of traditional spray stripping and ensuring uniform stripping.

[0048] The recycling unit 4 addresses the wastewater generated during the stripping process by employing a two-step key treatment process to achieve both environmental protection and resource recovery goals: The first step is circulation purification, using a precision filtration device to remove solid impurities from the wastewater, while simultaneously adjusting the pH level using an automatic pH calibration module to ensure the treated wastewater meets reuse requirements, thus achieving the recycling of the stripping solution; the second step is metal ion fractional recovery, utilizing a metal separation membrane module 41 to preliminarily separate different metal ions in the wastewater, achieving high-purity fractional recovery. This avoids environmental pollution caused by direct discharge of wastewater and enables the secondary utilization of metal resources, improving the system's economic efficiency and environmental friendliness.

[0049] In some specific embodiments, the recycling unit 4 includes a metal separation membrane assembly 41 and a gradient electric field module 42; the gradient electric field module 42 realizes the graded recycling of different metal ions in the heat sink assembly coating by sequentially switching different gradient voltages.

[0050] The metal separation membrane module 41 performs preliminary purification and ion pre-separation of the stripping waste liquid. It employs a polymer separation membrane with specific pore size and selective permeation performance. Based on the particle size differences, charge properties, and chemical affinity of different metal ions in the waste liquid, it performs preliminary screening of mixed ions. For example, for common copper, nickel, and gold ions in the waste liquid, the separation membrane can first retain larger impurity particles while allowing the target metal ions to pass through. Furthermore, some selective membranes can preferentially allow certain types of metal ions to permeate, achieving preliminary separation of different metal ions and avoiding mutual interference between different ions during subsequent electric field recovery. This lays the foundation for the precise classification of the gradient electric field module 42. In addition, this module can also remove residual stripping solution additives, micro-plating debris, and other impurities from the waste liquid, ensuring that the waste liquid entering the gradient electric field module 42 has a more homogeneous composition, thereby improving subsequent recovery efficiency and metal purity.

[0051] The gradient electric field module 42 separates metal ions based on the differences in the electrochemical properties of different metal ions in an electric field. Different metal ions have different standard electrode potentials and migration rates. Under the same electric field strength, metal ions with higher electrode potentials are more likely to gain electrons and be deposited at the cathode, while metal ions with lower electrode potentials require a higher electric field strength to be deposited. Based on this characteristic, the gradient electric field module 42 constructs an adjustable electric field environment through a programmable power supply and an electrode array. During operation, an initial electric field gradient is first set. At this time, metal ions with higher electrode potentials preferentially deposit on the cathode surface to form elemental metals. Once the concentration of these ions drops to a threshold, the system automatically switches to a higher voltage gradient, allowing metal ions with the next higher electrode potential to begin deposition. This process is repeated, and by sequentially switching between 3-4 different voltage gradients, the staged and orderly deposition of different metal ions is achieved.

[0052] In some specific embodiments, the multimodal data includes the appearance image data, three-dimensional dimension data, and coating spectral data of the heat sink component. The intelligent control unit 2 uses a built-in multimodal fusion recognition model to perform semantic segmentation, contour fitting, and spectral matching on the appearance image data, three-dimensional dimension data, and coating spectral data respectively to obtain the analysis results of each modality. A weighted fusion algorithm is used to fuse the analysis results of each modality to obtain the fusion judgment result, and the coating adhesion level of the heat sink component is determined based on the fusion judgment result. By accurately analyzing the modalities and performing multi-dimensional fusion judgment, the limitation that a single data dimension cannot fully reflect the coating characteristics is overcome, providing a scientific basis for determining the coating adhesion level. The multimodal data fusion processing flow is attached. Figure 3 As shown.

[0053] In some specific embodiments, pixel-level semantic segmentation is performed on the appearance image data to extract structural features of the heat sink component, including fins, blind holes, and corner gaps, and to determine the key areas to be processed in the coating; surface reconstruction and thickness fitting are performed on the three-dimensional dimensional data to obtain the initial thickness distribution of the coating and the surface contour parameters of the substrate; feature peak identification and matching are performed on the coating spectral data to determine the material composition of the coating and the proportion of each component.

[0054] For the appearance image data, the intelligent control unit 2 uses pixel-level semantic segmentation technology for processing. Through deep learning algorithms, each pixel in the image is classified into its corresponding structural category (such as fin areas, blind hole areas, corner gap areas, and flat areas), achieving refined identification of the surface structure of the heat sink component. Semantic segmentation processing can accurately extract structural features such as the distribution density of fins, the diameter and depth of blind holes, and the width of corner gaps. These features are directly related to the coating adhesion environment; for example, uneven deposition inside blind holes can lead to differences in adhesion. This allows for the identification of key areas for coating processing, providing a structural basis for adjusting the spray angle and pressure in subsequent stripping processes.

[0055] For 3D dimensional data, the processing revolves around surface reconstruction and thickness fitting: First, point cloud data of the heat sink component surface is acquired using a 3D scanning device. Then, a 3D modeling algorithm is used to reconstruct the discrete point cloud into a continuous surface model. Simultaneously, combined with the original 3D data of the substrate, the initial thickness distribution of the coating in different regions is calculated using a thickness fitting algorithm. In addition, this processing can also obtain the surface contour parameters of the substrate, such as the flatness of the substrate surface and the presence of micro-dimples. These parameters can not only help determine the coating adhesion but also avoid excessive corrosion caused by differences in the substrate contour during subsequent decoating.

[0056] For coating spectral data, the core processing method is characteristic peak identification and matching: using spectral analysis technology to obtain the absorption spectrum or emission spectrum of the coating, identifying the position and intensity of characteristic peaks in the spectrum, and matching them with a preset standard spectral database to determine the material composition of the coating and the proportion of each component; since the bonding force between coatings of different materials and the substrate is essentially different, the material composition and proportion data are the key basis for determining the coating adhesion level.

[0057] Appearance image data can only reflect structural features and cannot determine the material; three-dimensional dimension data can only reflect thickness distribution and cannot identify weak areas; coating spectral data can only reflect material composition and cannot locate the treatment position. By assigning different weights to appearance image data, three-dimensional dimension data, and coating spectral data, multi-dimensional feature complementary verification can be achieved, accurately identifying the coating adhesion level and easily corroded areas of heat sink components, and thus generating a more adaptable and differentiated stripping process solution.

[0058] For example, if the spectral data indicates a high-hardness alloy coating (suggesting high adhesion), the appearance image shows a complex blind hole structure (suggesting the need for enhanced pretreatment), and the three-dimensional dimensions show a uniform coating thickness (suggesting relatively stable stripping parameters), then after fusion, it is determined to be a high-adhesion coating, requiring targeted pretreatment and stripping parameters. If the spectral data indicates a single metal coating (suggesting medium to low adhesion), the appearance image shows a flat surface (suggesting low pretreatment requirements), and the three-dimensional dimensions show a thin coating thickness (suggesting the need for low-concentration stripping solution), then it is determined to be a low-adhesion coating. Through this multimodal data processing and fusion determination method, the intelligent control unit 2 can overcome the limitations of traditional single data recognition, achieving a comprehensive and accurate characterization of coating characteristics. This ensures the reliability of the coating adhesion level determination results, providing a scientific and effective decision-making basis for subsequently generating differentiated stripping process solutions, and avoiding problems such as incomplete stripping or substrate damage caused by deviations in the judgment of coating characteristics.

[0059] In some specific embodiments, the intelligent control unit 2 also incorporates a built-in process parameter mapping library. This library pre-stores parameter ranges for plasma bombardment power, stripping solution concentration, and spraying duration corresponding to different coating materials and adhesion levels. Based on the fusion judgment results of the multimodal fusion recognition model, the intelligent control unit 2 retrieves matching parameters from the process parameter mapping library to obtain the stripping process scheme. The process parameter mapping library is the core data support for the intelligent control unit 2 in generating the stripping process scheme. Essentially, it is a standardized parameter set built upon extensive experimental data and engineering experience. Its core function is to provide directly accessible and verified process parameter ranges for heat sink components with different coating characteristics, avoiding the subjectivity and errors of traditional manual parameter setting. The construction process of the stripping process scheme is shown in the attached figure. Figure 4 As shown.

[0060] The pre-stored content in the mapping library revolves around plating material and adhesion level, deeply binding these two key characteristics with specific process parameters. Plating materials cover common heat sink components such as copper, nickel, gold, and copper-nickel alloys. Adhesion levels are divided into high, medium, and low. The corresponding process parameters focus on the core operational indicators of the stripping execution unit 3, including plasma bombardment power, stripping solution concentration, and spraying mode. Each parameter is pre-stored in range format, allowing for fine-tuning based on the specific conditions of the heat sink components. For example, for copper plating, a combination with high adhesion levels, the mapping library pre-stores parameter ranges for plasma bombardment power of 80-100W, stripping solution concentration of 15%-20%, and spraying time of 15-20 minutes.

[0061] The intelligent control unit 2 decomposes the multimodal fusion judgment result into coating material and adhesion level, which serve as the retrieval index for the mapping library. Next, it locates the corresponding parameter range in the mapping library through the built-in matching algorithm. The algorithm will prioritize matching completely consistent "material-adhesion" combinations. If a special alloy material is encountered, it will perform similarity matching based on the material composition ratio. Then, after retrieving the parameter range, the intelligent control unit 2 will combine other analysis results of the multimodal data to precisely fine-tune the parameters within the range. Finally, it integrates the fine-tuned plasma bombardment parameters, stripping solution concentration parameters, and spraying time parameters with parameters such as pretreatment method and spraying method to form a complete and targeted stripping process solution.

[0062] In some specific embodiments, the pretreatment component 31 includes a plasma generator and a chemical cleaning device. For high-adhesion coatings, a segmented pulse bombardment pretreatment is performed using the plasma generator, and the spraying component 32 sprays a high-concentration stripping solution in a high-pressure directional mode. For medium-adhesion coatings, a dilute acid micro-etching pretreatment is performed using the chemical cleaning device, and the spraying component 32 sprays a medium-concentration stripping solution in a medium-pressure uniform mode. For low-adhesion coatings, a dilute acid micro-etching pretreatment is performed using the chemical cleaning device, and the spraying component 32 sprays a low-concentration stripping solution in a low-pressure mist mode. By functionally dividing the plasma generator and the chemical cleaning device, the pretreatment component 31 constructs a pretreatment system adapted to coatings with different adhesion, avoiding overtreatment that damages the substrate or undertreatment that leads to stripping difficulties. At the same time, the pretreatment operation is coordinated with the spraying mode and stripping solution concentration depth of the spraying component 32 to ensure a balance between stripping efficiency and quality.

[0063] For high-adhesion coatings, a segmented pulse bombardment method using a plasma generator is selected for pretreatment. This method utilizes a high-energy particle stream of plasma to periodically and discontinuously bombard the coating surface. The segmented pulse mode can both disrupt the dense oxide layer and microstructure of the coating surface through high-energy particles, weakening the interface between the coating and the substrate, and avoid excessively high local temperatures caused by continuous bombardment, preventing deformation or performance damage to the heat sink component substrate due to high temperatures. The corresponding jetting component 32 employs a high-pressure directional mode, using a jetting pressure of 0.6-0.8 MPa to precisely impact the coating surface with a high-concentration stripping solution in a directional stream. The impact force of the high-pressure stream assists in peeling off the plasma-weakened coating, while the high-concentration stripping solution rapidly reacts chemically with the coating, accelerating its dissolution. Together, these two methods address the problem of low stripping efficiency for high-adhesion coatings.

[0064] For coatings with medium adhesion, a dilute acid micro-etching method using a chemical cleaning device is selected for pretreatment. This method achieves pretreatment through gentle chemical action. For example, using 8%-12% dilute sulfuric acid or 5% dilute hydrochloric acid, the dilute acid can slowly dissolve the thin oxide layer on the coating surface, while forming a micro-uneven structure on the coating surface. This method does not damage the substrate and increases the contact area between the subsequent stripping solution and the coating, improving reaction efficiency. The matching spray mode is a medium-pressure uniform mode. The medium pressure ensures that the stripping solution fully covers the coating surface while avoiding excessive impact from high pressure on the micro-etched coating. A medium concentration of stripping solution is selected, which is suitable for the dissolution requirements of coatings with medium adhesion, ensuring the stripping rate while reducing the risk of substrate corrosion caused by excessive stripping solution.

[0065] For low-adhesion coatings, the pretreatment also employs dilute acid micro-etching using a chemical cleaning device, but with gentler parameters compared to medium-adhesion coatings. For example, the dilute acid concentration is reduced to 5%-8%, requiring only a light removal of oil and a very thin oxide layer from the coating surface, avoiding excessive micro-etching that could damage the fragile interface between the coating and the substrate. The corresponding spraying component 32 uses a low-pressure mist mode, which allows the low-concentration stripping solution to cover the coating surface in a uniform droplet form. The combination of low pressure and low concentration can slowly and controllably dissolve the low-adhesion coating, avoiding both high-pressure impact that could cause coating fragments to embed into the substrate gaps and unnecessary corrosion of the substrate by the high-concentration stripping solution.

[0066] In some specific embodiments, the stripping execution unit 3 also includes a robotic arm. The potential detection component 11, thickness measurement component 12, plasma generator, and chemical cleaning device are integrated on the robotic arm and arranged in a centrally symmetrical modular layout. The stripping execution unit 3 achieves the integrated layout of the potential detection component 11, thickness measurement component 12, plasma generator, and chemical cleaning device through the robotic arm, balancing the functional synergy of each component and avoiding mutual interference. At the same time, with the flexible movement capability of the robotic arm, it realizes the integrated continuous operation of detection, pretreatment, and stripping during the stripping process of the heat sink component, breaking the problems of long operation intervals and large positioning deviations caused by the dispersed layout of traditional equipment.

[0067] In some specific embodiments, the plasma generator is arranged in a ring-radiation pattern, with the spray nozzles of the chemical cleaning device surrounding the outside of the plasma generator and the spray nozzles facing the same direction as the bombardment direction of the plasma generator; the detection probe of the thickness measuring component 12 is located between the chemical cleaning device and the plasma generator and is isolated from the plasma generator by a heat-insulating and light-shielding bracket; the micro reference electrode array of the potential detection component 11 is embedded in the gap of the spray nozzles of the chemical cleaning device.

[0068] The plasma generator is positioned at the core of the layout for two reasons. Firstly, the plasma generator is only used for coatings with high adhesion. Placing it at the center ensures a more stable bombardment direction on the heat sink components, avoiding deviations in the pretreatment area caused by layout offsets. Secondly, the annular radial layout allows other components to be evenly distributed around the core device, shortening the distance between each component and the area to be treated on the heat sink components, reducing the movement path of the robotic arm, and improving work efficiency.

[0069] The chemical cleaning device's spray nozzles are arranged around the outside of the plasma generator, and the direction of the spray nozzles is consistent with the bombardment direction of the plasma generator, ensuring the continuity of the pretreatment operation. When treating high-adhesion coatings, the plasma generator first performs pulse bombardment on the target area. After the bombardment, the robotic arm does not need to be significantly adjusted in position, and the chemical cleaning device can perform auxiliary cleaning on the same area through the same-direction spray nozzles to remove coating debris generated by the bombardment, avoiding debris residue from affecting subsequent stripping. When treating medium- to low-adhesion coatings, the chemical cleaning device can independently spray dilute acid through the spray nozzles for micro-etching pretreatment. The same-direction layout ensures that the spray range overlaps with the coverage area of ​​the stripping solution of the subsequent spraying component 32, improving the connection accuracy between pretreatment and stripping.

[0070] The detection probe of the thickness measuring component 12 is located between the chemical cleaning device and the plasma generator, and is isolated from the plasma generator by a heat-insulating and light-shielding bracket. This location and protective design combine functionality and safety. From a functional perspective, this position is located at the core periphery of the pretreatment area, enabling rapid detection of the coating thickness of the target area before and after pretreatment. Pre-treatment detection can obtain initial thickness data, providing a basis for the intelligent control unit 2 to set the stripping parameters. Post-treatment detection can verify the pretreatment effect, while avoiding detection delays caused by the detection probe being too far from the pretreatment area. From a protective perspective, the plasma generator generates high temperature and strong light during operation. The heat-insulating and light-shielding bracket can block the transmission of high temperature and prevent strong light from interfering with the laser detection signal of the thickness measuring component 12, ensuring the accuracy of the thickness measurement data.

[0071] The micro reference electrode array of the potential detection component 11 is embedded in the gap of the spray nozzle of the chemical cleaning device. The micro reference electrode array needs to maintain an effective detection distance of 1-2 mm from the surface of the heat sink component. Embedding it in the gap of the spray nozzle can use the positioning reference of the spray nozzle to ensure that the distance between the electrode probe and the heat sink surface is stable, avoiding detection distance deviation caused by slight shaking of the robotic arm. At the same time, the position of the spray nozzle gap allows the electrode probe to be simultaneously within the coverage area of ​​the stripping solution or pretreatment solution, and to collect the potential change data of the coating under chemical action in real time. For example, during the chemical cleaning micro-etching process, the electrode can monitor the potential fluctuation caused by the dissolution of the coating in real time, providing a basis for the intelligent control unit 2 to determine whether the pretreatment meets the standards; during the stripping process, it can also simultaneously monitor the potential difference between the coating and the substrate, and promptly warn of the risk of the substrate being exposed, realizing a closed loop of potential monitoring throughout the stripping process.

[0072] In some specific embodiments, the anti-corrosion linkage mechanism includes: controlling the spraying component 32 to switch to a mist spraying mode, shortening the spraying time in easily corroded areas, controlling the liquid preparation component 33 to reduce the active concentration of the stripping solution, and controlling the recovery unit 4 to output a weak reduction current and form a passivation film on the substrate surface.

[0073] The anti-corrosion linkage mechanism is an active protection strategy implemented by the intelligent control unit 2 for scenarios where the substrate is about to be exposed or the risk of localized corrosion increases during the stripping process. Its core is to form a protective closed loop through multi-unit collaborative adjustment of process parameters, addressing three dimensions: slowing the corrosion rate, blocking the corrosion reaction, and strengthening substrate protection. This avoids the problem of excessive substrate corrosion caused by fixed process parameters in traditional stripping processes. The trigger condition is that the intelligent control unit 2 determines, through real-time data acquired by the stripping detection unit 1, that the stripping process has reached a critical threshold. For example, if the potential data shows that the potential value in a certain area is close to the critical corrosion potential of the substrate, or if the thickness data shows that the local coating thickness has dropped to near the substrate surface, the system immediately activates the anti-corrosion linkage mechanism.

[0074] The spray assembly 32 is switched to a mist spray mode to slow down the corrosion rate from the perspective of the contact method of the stripping solution. In traditional directional spraying or medium-high pressure spraying, the stripping solution impacts the coating surface at a high flow rate. If the coating is locally thin, the high-flow-rate liquid can easily directly wash away the substrate and accelerate corrosion. After switching to mist spraying, the nozzle of the spray assembly 32 atomizes the stripping solution into uniform droplets with a diameter of 50-100μm through the internal flow channel design. These droplets cover the substrate surface at a low pressure of 0.1-0.2MPa. The mist shape significantly reduces the impact of the stripping solution on the substrate and also reduces the amount of stripping solution remaining on the substrate surface, slowing down the chemical reaction rate between the stripping solution and the substrate, and allowing time for other protective measures to take effect.

[0075] Shortening the spraying time in easily corroded areas is a localized protection method based on precise quantity control. The intelligent control unit 2 has pre-located the easily corroded areas of the heat sink component through semantic segmentation results of appearance image data; when the corrosion risk in these areas is detected to be increased, the system shortens the single spraying time of the easily corroded area from the conventional 5-8 seconds to 2-3 seconds by adjusting the movement path of the robotic arm or the opening and closing of the nozzle of the spraying component 32, while extending the spraying interval to reduce the cumulative action time of the stripping solution in the easily corroded area.

[0076] The solution mixing component 33 reduces the active concentration of the stripping solution, thereby weakening its corrosive power from the perspective of the corrosive medium intensity. The solution mixing component 33 has multiple sets of raw solution storage tanks and precision metering pumps built in. During normal stripping, the solutions are mixed in a preset ratio to form a highly active stripping solution. After the anti-corrosion mechanism is activated, the intelligent control unit 2 sends a concentration adjustment command to the solution mixing component 33. The metering pump reduces the feed amount of the main corrosive agent while increasing the amount of corrosion inhibitor added, thereby reducing the active concentration of the stripping solution.

[0077] The recovery unit 4 outputs a weak reducing current and forms a passivation film on the substrate surface, achieving long-term corrosion protection from the substrate surface protection level. In addition to conventional waste liquid treatment functions, the recovery unit 4 also has a built-in low-voltage DC power supply and a retractable electrode assembly. When this measure is activated, the system extends the electrode assembly of the recovery unit 4 and approaches it close to the substrate surface of the heat sink assembly, outputting a weak reducing current of 1-2V. Simultaneously, a small amount of passivation liquid is sprayed onto the substrate surface through the liquid distribution assembly 33. Under the action of the weak reducing current, the metal ions in the passivation liquid undergo a reduction reaction on the substrate surface, forming a dense passivation film with a thickness of 1-2μm. This passivation film can physically isolate the stripping solution from contact with the substrate, fundamentally blocking subsequent corrosion reactions. Furthermore, the passivation film thickness is controllable and will not affect subsequent re-plating processing of the substrate.

[0078] A control method for an integrated intelligent stripping system, used to control any of the above-mentioned integrated intelligent stripping systems; a flowchart of the control method is attached. Figure 5 As shown, the control method includes the following steps:

[0079] 101. Acquire multimodal data of the heat sink component to be deplated, and the intelligent control unit generates a deplating process plan for the heat sink component based on the multimodal data;

[0080] 102. According to the stripping process plan, the stripping unit performs pretreatment through the pretreatment component, and after the concentration of the stripping solution is adjusted by the solution mixing component, the spraying component performs targeted spraying of the stripping solution for stripping.

[0081] 103. The stripping waste liquid is recycled and purified through a recycling unit;

[0082] 104. Collect distributed potential data on the surface of the heat sink component through the potential detection component, and collect real-time thickness data of the heat sink component through the thickness measurement component;

[0083] 105. The intelligent control unit determines the stripping process based on corrosion potential data and coating thickness data, and determines whether to trigger the anti-corrosion linkage mechanism based on the stripping process.

[0084] Those skilled in the art will understand that the components of this application described above can be implemented using general-purpose computing systems. They can be centralized on a single computing system or distributed across a network of multiple computing systems. Optionally, they can be implemented using computer-executable program code, thereby allowing them to be stored in a storage system for execution by the computing system. Alternatively, they can be fabricated as separate integrated circuit components, or multiple components or steps can be fabricated as a single integrated circuit component. Thus, this application is not limited to any particular combination of hardware and software.

[0085] Note that the above description is merely a preferred embodiment and the technical principles employed in this application. Those skilled in the art will understand that this application is not limited to the specific embodiments described herein, and various obvious changes, readjustments, and substitutions can be made without departing from the scope of protection of this application. Therefore, although this application has been described in detail through the above embodiments, this application is not limited to the above embodiments. Many other equivalent embodiments may be included without departing from the concept of this application, and the scope of this application is determined by the scope of the appended claims.

[0086] The above disclosures are only a few specific implementation scenarios of this application. However, this application is not limited to these. Any variations that can be conceived by those skilled in the art should fall within the protection scope of this application.

Claims

1. An integrated intelligent stripping system for heat sink components, characterized in that, The device comprises a stripping detection unit, an intelligent control unit, a stripping execution unit and a recycling unit. The stripping detection unit comprises a potential detection assembly and a thickness measurement assembly, which are used to collect distributed potential data of the surface of the heat sink assembly through the potential detection assembly and collect real-time thickness data of the heat sink assembly through the thickness measurement assembly. The intelligent control unit is used to obtain multi-modal data of the heat sink assembly to be stripped and generate a stripping process scheme of the heat sink assembly according to the multi-modal data. The corrosion potential data and the coating thickness data are used to determine the stripping progress, and the stripping progress is used to determine whether to trigger the anti-corrosion linkage mechanism. The stripping execution unit comprises a pretreatment assembly, a spraying assembly and a liquid preparation assembly, which are used to perform pretreatment through the pretreatment assembly according to the differential stripping process scheme, adjust the concentration of the stripping solution through the liquid preparation assembly, and perform targeted spraying of the stripping solution through the spraying assembly to realize spraying stripping. The recycling unit is used to recycle and purify the stripping waste liquid and grade the metal ions. The multi-modal data comprises appearance image data, three-dimensional size data and coating spectrum data of the heat sink assembly. The intelligent control unit performs semantic segmentation, contour fitting and spectrum matching on the appearance image data, the three-dimensional size data and the coating spectrum data respectively through a built-in multi-modal fusion recognition model to obtain analysis results of each modality. A weighted fusion algorithm is used to fuse the analysis results of each modality to obtain a fusion determination result. The fusion determination result is used to determine the coating adhesion level of the heat sink assembly. The appearance image data is subjected to pixel-level semantic segmentation to extract structural features of the heat sink assembly including fins, blind holes and corner gaps, and determine the key areas of the coating to be treated. The three-dimensional size data is subjected to surface reconstruction and thickness fitting to obtain the initial thickness distribution of the coating and the surface contour parameters of the base material. The coating spectrum data is subjected to feature peak identification and matching to determine the material composition of the coating and the proportion of each component. The intelligent control unit further comprises a process parameter mapping library. The process parameter mapping library pre-stores the parameter intervals of plasma bombardment power, stripping solution concentration and spraying time corresponding to different coating materials and adhesion levels. The intelligent control unit retrieves the matching parameters from the process parameter mapping library according to the fusion determination result of the multi-modal fusion recognition model to obtain the stripping process scheme.

2. The integrated intelligent stripping system of claim 1, wherein The recycling unit comprises a metal separation membrane assembly and a gradient electric field module. The gradient electric field module realizes the graded recovery of different metal ions in the coating of the heat sink assembly by sequentially switching different gradient voltages.

3. The integrated intelligent stripping system of claim 1, wherein The pretreatment assembly comprises a plasma generating device and a chemical cleaning device. For high-adhesion coatings, the plasma generating device is used for segmented pulse bombardment pretreatment, and the spraying assembly sprays high-concentration stripping solution in a high-pressure directional mode. For medium-adhesion coatings, the chemical cleaning device is used for dilute acid etching pretreatment, and the spraying assembly sprays medium-concentration stripping solution in a medium-pressure uniform mode. For low-adhesion coatings, the chemical cleaning device is used for dilute acid etching pretreatment, and the spraying assembly sprays low-concentration stripping solution in a low-pressure mist mode.

4. The integrated intelligent stripping system of claim 1, wherein The stripping execution unit further comprises a mechanical arm, and the potential detection assembly, the thickness measurement assembly, the plasma generating device and the chemical cleaning device are integrated on the mechanical arm and arranged in a central symmetrical modular layout.

5. The integrated smart stripping system of claim 4, wherein, The plasma generating device is arranged in a ring-shaped radiation layout, the spray openings of the chemical cleaning device are arranged outside the plasma generating device, and the spray openings are directed in the same direction as the bombardment direction of the plasma generating device; the detection probe of the thickness measurement assembly is located between the chemical cleaning device and the plasma generating device and is isolated from the plasma generating device by a heat-shielding and light-shielding support; and the micro reference electrode array of the potential detection assembly is embedded in the gap between the spray openings of the chemical cleaning device.

6. The integrated intelligent stripping system of claim 1, wherein The anti-corrosion linkage mechanism comprises: controlling the spray assembly to switch to a mist spray mode, shortening the spraying time of the easily-corroded area, controlling the liquid preparation assembly to reduce the active concentration of the stripping liquid, and controlling the recovery unit to output a weak reducing current and form a passivation film on the surface of the substrate.

7. A control method of an integrated intelligent stripping system, characterized by, The control method comprises the following steps: acquiring multi-modal data of the heat sink assembly to be stripped, and generating a stripping process scheme for the heat sink assembly according to the multi-modal data by the intelligent control unit; performing pretreatment on the heat sink assembly by the pretreatment assembly according to the stripping process scheme, adjusting the concentration of the stripping liquid by the liquid preparation assembly, and performing targeted spray stripping of the heat sink assembly by the spray assembly; recovering and purifying the stripping waste liquid by the recovery unit; collecting distributed potential data on the surface of the heat sink assembly by the potential detection assembly and collecting real-time thickness data of the heat sink assembly by the thickness measurement assembly; determining the stripping progress according to the corrosion potential data and the coating thickness data by the intelligent control unit, and determining whether to trigger the anti-corrosion linkage mechanism according to the stripping progress.

Images