Metal deplating environment dynamic regulation and control method and related equipment

Abstract

The invention relates to the technical field of metal deplating, and provides a metal deplating environment dynamic regulation and control method and related equipment, and the method comprises the following steps: obtaining basic information of a to-be-deplated product; according to the basic information of the to-be-deplated product, performing simulation to obtain deplating process reference information; carrying out deplating treatment on the basis of the deplating process reference information, and dynamically acquiring multi-mode real-time deplating data; the stripping process reference information and the multi-mode real-time stripping data are combined, and stripping environment dynamic adjusting parameters are obtained; and regulating and controlling the deplating environment based on the dynamic regulation parameters of the deplating environment. According to the application, the deplating treatment is carried out according to the basic information of the to-be-deplated product, and the deplating treatment mode is set based on the actual condition of the product, so that differential treatment of different to-be-deplated products can be realized; and the deplating condition is obtained in real time in the deplating process, and the deplating environment is dynamically and adaptively adjusted, so that good deplating efficiency, deplating stability and deplating effect are realized.

Description

Technical Field

[0001] This application relates to the field of metal stripping technology, and in particular to a method and related equipment for dynamic control of the metal stripping environment. Background Technology

[0002] Metal stripping is a technique used in electroplating and metal surface treatment to remove plating layers from the surface of parts.

[0003] Currently, in metal stripping, it is difficult to effectively control stripping efficiency and stability, which seriously affects the metal stripping effect of the parts. Summary of the Invention

[0004] To address the above technical problems, this application proposes a method and related equipment for dynamic control of the metal stripping environment.

[0005] The first aspect of this application provides a method for dynamically controlling the metal stripping environment, the method comprising: Obtain basic information about the products to be de-plated; Based on the basic information of the product to be stripped, the baseline information of the stripping process is simulated. The plating removal process is performed based on the reference information of the plating removal process, and multimodal real-time plating removal data is dynamically acquired. By combining the reference information of the stripping process with the multimodal real-time stripping data, dynamic adjustment parameters for the stripping environment are obtained; The stripping environment is regulated based on the aforementioned dynamic adjustment parameters.

[0006] In an optional embodiment, simulating the baseline information for the plating removal process based on the basic information of the product to be stripped includes: Based on the basic information of the product to be deplated, detailed information on product features is obtained; Based on the basic information of the product to be stripped and the product feature enhancement information, the basic parameters of stripping are obtained. The basic parameters of stripping include one or any combination of stripping solution formulation, stripping temperature and current density. Based on the product feature enhancement information and the basic parameters of the stripping process, the stripping process simulation information is obtained. The stripping process simulation information includes one or more of the key time nodes, reaction rate, and target metal concentration. The basic parameters of the stripping process are integrated with the simulation information of the stripping process to obtain the baseline information of the stripping process.

[0007] In an optional embodiment, obtaining product feature refinement information based on the basic information of the product to be deplated includes: Obtain the substrate type information, coating metal type information, coating thickness information, product standard image and image description information from the basic information of the product to be deplated; Based on the substrate type information and the coating metal type information, material property information is obtained, and the material property information is converted into a numerical feature vector; Obtain image feature vectors based on the product standard image; Generate text feature vectors based on the image description information; By combining the coating thickness information with the image description information, physical constraints are obtained; The numerical feature vector, the image feature vector, and the text feature vector are weighted and fused together, and combined with the physical constraints, to obtain product feature enhancement information.

[0008] In an optional embodiment, the step of integrating the basic parameters of the stripping process with the simulation information of the stripping process to obtain the baseline information for the stripping process includes: Based on the relationship between the basic parameters of the stripping process and the corresponding simulation information of the stripping process, the initial reference information of the stripping process is obtained by integrating them. By using a preset optimization objective, the initial baseline information of the stripping process is iteratively updated according to a genetic algorithm to obtain the baseline information of the stripping process.

[0009] In an optional embodiment, the step of combining the reference information of the stripping process with the multimodal real-time stripping data to obtain the dynamic adjustment parameters of the stripping environment includes: The time node for acquiring the multimodal real-time deplating data is determined as the first time node; Obtain the first stripping process simulation information corresponding to the first time node from the stripping process reference information; The reaction rate information and target metal concentration information in the multimodal real-time stripping data are obtained to obtain the first real-time stripping data; Based on the first real-time stripping data and the first stripping process simulation information, a simulated deviation reference value is obtained; The real-time image of the stripped product, stripping temperature, and pH value are obtained from the multimodal real-time stripping data to obtain the second real-time stripping data; The simulated development trend value is obtained based on the second real-time deplating data; Based on the simulated deviation reference value and the simulated development trend value, the dynamic adjustment parameters of the stripping environment are obtained.

[0010] In an optional embodiment, obtaining the dynamic adjustment parameters of the stripping environment based on the simulated deviation reference value and the simulated development trend value includes: The first weight of the simulated deviation from the reference value and the second weight of the simulated development trend value are obtained through a preset node weight mapping table. The dynamic adjustment feature vector of the plating removal environment is obtained by weighting and fusing the first weight with the simulated deviation reference value and the second weight with the simulated development trend value; The dynamic adjustment parameters of the plating removal environment are obtained based on the dynamic adjustment feature vector of the plating removal environment. The control objects of the dynamic adjustment parameters of the stripping environment include one or any combination of stripping time adjustment, stripping solution flow rate, regenerator replenishment rate, stripping environment temperature, and electrowinning start / stop.

[0011] In an optional embodiment, the method further includes: After the plating on the product to be stripped is completed, a plating effect evaluation report is generated based on the baseline information of the plating stripping process and the dynamic adjustment parameters of the plating stripping environment.

[0012] A second aspect of this application provides a dynamic control system for a metal stripping environment, the system comprising: The first information acquisition module is used to acquire basic information about the products to be deplated. The stripping process simulation module is used to simulate the baseline information of the stripping process based on the basic information of the product to be stripped. The second information acquisition module is used to perform stripping processing based on the stripping process reference information and dynamically acquire multimodal real-time stripping data. The stripping parameter acquisition module is used to combine the stripping process reference information with the multimodal real-time stripping data to obtain the stripping environment dynamic adjustment parameters; The environmental dynamic control module is used to control the stripping environment based on dynamic adjustment parameters.

[0013] A third aspect of this application provides an electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of a method for dynamically controlling the metal stripping environment.

[0014] A fourth aspect of this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a method for dynamically controlling the metal stripping environment.

[0015] Beneficial effects: This application provides a method and related equipment for dynamic control of the metal stripping environment, which can perform differentiated stripping treatment for different products to be stripped and achieve effective control of the stripping environment, thereby effectively improving the stripping effect and stripping stability. Specifically, the method of this application simulates the baseline information of the stripping process based on the basic information of the product to be stripped, and performs stripping treatment based on the baseline information of the stripping process; the stripping treatment method is set according to the actual situation of the product, which can realize differentiated treatment for different products to be stripped; and the stripping status is acquired in real time during the stripping process, and the stripping environment is dynamically and adaptively adjusted according to the actual situation during the stripping process, thereby achieving good stripping efficiency, stripping stability and stripping effect. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of this application and should not be regarded as a limitation of the scope. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart of a method for dynamically controlling the metal stripping environment provided in the embodiments of this application; Figure 2 The functional modules of the metal stripping environment dynamic control device provided in the embodiments of this application are as follows: Figure 1 ; Figure 3 Functional modules of the metal stripping environment dynamic control device provided in the embodiments of this application Figure 2 ; Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application.

[0018] Figure label: 1-First information acquisition module; 2-Removal process simulation module; 3-Second information acquisition module; 4-Removal parameter acquisition module; 5-Environmental dynamic control module; 6-Removal report generation module; 7-Electronic equipment; 71-Memory; 72-Processor; 73-Bus. Detailed Implementation

[0019] Various embodiments of this disclosure will be described more fully below. This disclosure may have various embodiments, and adjustments and changes may be made therein. However, it should be understood that there is no intention to limit the various embodiments of this disclosure to the specific embodiments disclosed herein, but rather this disclosure should be understood to cover all adjustments, equivalents, and / or alternatives falling within the spirit and scope of the various embodiments of this disclosure.

[0020] Reference Figure 1 As shown in the embodiments of this application, a method for dynamic control of the metal stripping environment is proposed, the method comprising: S100, obtain basic information about the product to be stripped of plating; The basic information of the product to be stripped includes one or more of the following: substrate type information, coating metal type information, coating thickness information, product standard image and image description information.

[0021] Specifically, in some embodiments of this application, the basic information of the product to be stripped of plating is obtained based on production orders or processing orders; the orders generally involve information such as the size, shape, material type, thickness, and image of the product to be stripped. Orders can exist in paper form or stored electronically in a database. If the order is paper, an image of the order needs to be captured using a photographic device, and then text and product images need to be extracted and parsed from the order image. In practical applications, deep learning networks or optical character recognition modules can be used to extract text and obtain relevant information from the order, ultimately obtaining the required basic information of the product to be stripped of plating.

[0022] Of course, there are no specific restrictions on the content of the basic information of the products to be deplated or on how the basic information is obtained.

[0023] S200, based on the basic information of the product to be stripped, simulates the baseline information of the stripping process; The baseline information for the stripping process can generally include basic stripping parameter information and stripping process simulation information corresponding to each basic stripping parameter.

[0024] Basic parameters for stripping typically include the stripping solution formulation, stripping temperature, and current density. Meanwhile, the information on the stripping process simulation generally refers to the information on the stripping rate, stripping time, changes in the metal ion content in the stripping solution, and changes in the image of the product to be stripped, based on the aforementioned basic parameters.

[0025] Of course, the above content is only a few specific examples of basic parameters and simulation information for stripping, illustrating only one possible implementation method, and does not limit the specific content of the basic parameters and simulation information for stripping.

[0026] For example, in some embodiments of this application, baseline information for the stripping process can be obtained from an empirical database of "products-processes-procedures and results". The empirical database of "products-processes-procedures and results" is established by collecting historical data of different types of products to be stripped.

[0027] After obtaining the basic information of the product to be stripped, the basic parameters of the stripping process corresponding to the product are obtained by case matching and analogy in the experience database of "product-process-process and result". The credibility of the inference results is judged by confidence calculation. Cases are ranked according to the confidence level. The specific information of the adaptive optimization process is used as the basic parameter information of stripping. Based on the cases matched in the experience database, the process and results of the cases are simulated and adjusted as the simulation information of the stripping process.

[0028] Furthermore, the process and results of simulating and adjusting cases can be achieved using software such as HSC Chemistry. Within software such as HSC Chemistry, chemical thermodynamic calculations are performed based on the basic information of the product to be stripped and / or the basic parameters of the stripping process to complete the simulation of the feasibility and equilibrium state of the stripping reaction. The process and results of simulating and adjusting cases can also be achieved by conducting small-scale tests on different products to be stripped in order to correct the simulation information of the stripping process.

[0029] Of course, there are no specific restrictions on the method for simulating the baseline information of the stripping process. The aforementioned solution can also be achieved by training a neural network model. For example, the basic information of the product to be stripped can be used as input data, and the corresponding baseline information of the stripping process can be set as output verification. Through the trained stripping process simulation model, the baseline information of the stripping process can be simulated and output through an intermediate layer based on the basic information of the product to be stripped. The data for training the neural network model can come from an empirical database.

[0030] The S300 performs stripping processing based on reference information of the stripping process and dynamically acquires multimodal real-time stripping data. Specifically, in some embodiments of this application, the stripping process is performed based on the reference information of the stripping process, that is, the parameters of the basic parameters of the stripping process in the reference information of the stripping process are used as the initial conditions to perform the stripping process on the product to be stripped; and multimodal real-time stripping data is dynamically acquired during the stripping process, such as continuous acquisition or intermittent acquisition. Intermittent acquisition is usually achieved by setting a certain time interval or acquisition frequency.

[0031] For example, multimodal real-time stripping data may include audio data, image data, and text data. Audio data, such as the sound of bubbles generated by the stripping solution, is used to reflect the intensity of the stripping process; image data, such as images of the stripped product, is used to reflect the stripping effect and degree; text data, such as the concentration of metal ions in the stripping solution, the reaction rate of stripping, and the saturation of the stripping solution, are used.

[0032] Of course, no restrictions are placed on the specific format and content of the multimodal real-time deplating data.

[0033] S400 combines baseline information from the stripping process with real-time stripping data from multiple modes to obtain dynamic adjustment parameters for the stripping environment; Specifically, in some embodiments of this application, by combining the reference information of the stripping process with multimodal real-time stripping data, the degree of deviation in the real-time stripping process is obtained by comparing the difference between the reference information of the stripping process and the multimodal real-time stripping data, and the cause of the deviation and the corresponding dynamic adjustment of the stripping environment can be obtained based on the degree of deviation and / or the multimodal real-time stripping data.

[0034] The S500 regulates the stripping environment based on dynamic adjustment parameters.

[0035] Understandably, the metal stripping environment dynamic control method provided in this application embodiment simulates the stripping process reference information based on the basic information of the product to be stripped. It can generate different stripping process reference information for different products to be stripped, which can reduce the influence of the operator's subjective factors and optimize the product stripping process. The stripping process reference information is used as a reference for the product stripping process, realizing differentiated treatment for different products to be stripped. Furthermore, it dynamically acquires multimodal real-time stripping data, continuously monitors the stripping process, enhances the controllability of the stripping process, improves the ability to handle complex workpieces, and adapts to the stripping needs of complex shapes, local deep holes, or masks. Based on the difference between multimodal real-time data and the baseline information of the stripping process, it can achieve targeted and precise dynamic control of the stripping environment, promptly solve problems that occur during the stripping process, improve the anti-interference ability of the stripping process, keep the stripping environment of the product to be stripped relatively stable, and keep the stripping rate of the product to be stripped stable, thus achieving a good and stable stripping effect. In addition, it also helps to improve the consistency of the stripping effect of different batches of products to be stripped to a large extent, which can reduce the rework rate of products and thus increase production capacity.

[0036] In an optional embodiment, the baseline information for the plating removal process, simulated based on the basic information of the product to be stripped, includes: Based on the basic information of the product to be deplated, detailed information on product characteristics is obtained; Basic product information consists of the product's objective description, inherent attributes, and initial data. It is usually static and superficial information. In contrast, in-depth product feature information is obtained by analyzing the basic information of the product to be deplated. It is usually in-depth and potential information.

[0037] For example, in some embodiments of this application, when the type of plating metal is obtained, information such as plating hardness, adhesion, and wear resistance can be further analyzed, as well as information such as conductivity, activation energy, and molar mass. Similarly, when a standard image of the product is obtained, detailed information on product characteristics such as the degree of masking, surface roughness, porosity, and shape complexity of the product to be stripped can be analyzed, and the difficulty of stripping the plating can be further analyzed.

[0038] Of course, the above are just some examples of product feature enhancement information. There are no restrictions on the specific content and acquisition process of product feature enhancement information, and adjustments can be made according to the actual situation.

[0039] Understandably, the product feature enhancement information is obtained by mining or expanding the basic information of the product to be deplated, in order to improve the depth and richness of the product features.

[0040] Based on the basic information and detailed information on the product characteristics to be stripped, the basic parameters of stripping are obtained. The basic parameters of stripping include one or more of the stripping solution formula, stripping temperature and current density. Specifically, in some embodiments of this application, a preset experience database is invoked. The experience database may be the "product-process-procedure and result" experience database as described above. The basic information and product feature refinement information of the product to be deplated are matched with the cases in the experience database, and the basic deplating parameter information of the corresponding product to be deplated is determined according to the degree of matching.

[0041] The empirical database includes a process formulation sub-library, theoretical models and / or empirical models, and a material property sub-library.

[0042] More specifically, the process formulation sub-library can include standard stripping parameters corresponding to different materials, thicknesses, and shapes, such as stripping solution formulation, stripping temperature, stripping solution concentration, stripped layer density, and stripping time. The theoretical and / or empirical models can be diffusion models based on Fick's law and / or reaction rate models based on the Arrhenius equation, describing the relationship between parameters and stripping rates. The material property sub-library stores data on the substrate and coating, such as hardness, activation energy, and molar mass.

[0043] Furthermore, the theoretical and / or empirical models embed functional units, specifically including reaction kinetics units, mass transfer process units, temperature field simulation units, and geometric evolution units. The reaction kinetics unit calculates the coating dissolution rate based on hydrogen ion concentration and metal ion concentration. The mass transfer process unit evaluates diffusion and transport time. The temperature field simulation unit simulates the effect of temperature on all parameters (reaction rate, mass transfer coefficient, solubility, etc.). The geometric evolution unit simulates the shape changes and dynamic changes in the dissolution contact area of ​​the product to be stripped, and introduces randomness or cellular automata concepts to simulate non-uniform dissolution caused by local defects, generating a simulated image of the product to be stripped. For example, the geometric evolution unit can be implemented through customized development based on the COMSOL Multiphysics platform; similarly, the temperature field simulation unit can be directly implemented using the mature physics interface built into COMSOL Multiphysics.

[0044] Of course, no specific restrictions are placed on whether or not functional units are embedded in theoretical and / or empirical models, or on the data, functions, and specific implementation methods of kinetic energy units.

[0045] For example, based on substrate type information and coating metal type information, information such as stripping solution formulation, corrosion inhibitor type, stripping method, main salt and complexing agent can be obtained from the experience database to achieve effective dissolution of metal coatings and protect the substrate to the greatest extent possible, taking into account the chemical properties of different metals. Based on coating thickness information, product standard images and image description information, information such as current density, stripping solution concentration, stripping time, stripping temperature and stripping solution concentration can be determined or adjusted in the experience database.

[0046] For example, in experience databases, for highly reactive substrate types such as magnesium and aluminum, it is generally indicated that a mild stripping solution should be used and a highly effective corrosion inhibitor should be added appropriately; for substrate types with strong stability such as stainless steel and alloys, it is generally indicated that a stripping solution with strong oxidizing properties should be used; if the product shape is complex, the experience database will generally indicate that the stripping speed is slower and the required stripping time is longer; if the product has low complexity and a simple structure, the experience database will generally indicate that the stripping speed is faster and the required stripping time is shorter.

[0047] Based on the product feature enhancement information and basic stripping parameter information, the stripping process simulation information is obtained. The stripping process simulation information includes one or more of the key time nodes, reaction rate and target metal concentration. Specifically, in some embodiments of this application, product feature refinement information and basic stripping parameter information can be used as data input. A theoretical model and / or empirical model from an empirical database can be used as a framework to perform dynamic simulation to simulate and predict the stripping process. Furthermore, by identifying abrupt change points in the stripping process, each abrupt change point is designated as a key time point. Simulation data for each key time point, including reaction rate and target metal concentration, is obtained. Finally, a simulation data chart of the stripping process is generated based on the simulation data for each key time point.

[0048] Among them, the mutation node or key time node may include one or any number of the following: the stripping start node, the interface structure destruction node, the stripping rate mutation node, the stripping solution composition mutation node, and the stripping end node.

[0049] The stripping start point refers to the point where the stripping reaction truly begins, the stripping end point refers to the point where the stripping reaction completely ends, the stripping rate mutation point refers to the point where the stripping rate suddenly changes from a stable state, such as the point where the stripping rate changes from continuous growth to a point where it tends to stabilize, or the point where it changes from a stable state to a point where it shows a downward trend, and the stripping solution composition mutation point can be based on the change in the concentration of the target metal ions in the stripping solution, such as the point where the target metal ions are saturated or tend to be saturated.

[0050] By integrating the basic parameters of the stripping process with the simulation information of the stripping process, the baseline information of the stripping process is obtained.

[0051] For the same type of coated metal material, there can be multiple sets of basic stripping parameters, such as different selection of stripping solution and different setting of stripping temperature. Different basic stripping parameters will generate different stripping process simulation information. Based on the preset target, a set of basic stripping parameters and stripping process simulation information can be selected and combined to obtain the stripping process reference information.

[0052] In an optional embodiment, based on the basic information of the product to be deplated, the product feature enhancement information obtained includes: Obtain the basic information of the product to be stripped, including substrate type, coating metal type, coating thickness, product standard image, and image description. Based on the substrate type information and coating metal type information, material property information is obtained and then converted into a numerical feature vector. For example, in some embodiments of this application, the substrate material and the coating material can be converted into numerical vectors. Electrochemical properties of the materials, such as standard electrode potential, corrosion resistance, atomic weight, and the electrode potential difference between the substrate and the coating, can be embedded within these numerical vectors to represent the basic reaction tendencies of different material combinations in the stripping solution, which helps in the systematic understanding of the material properties of the product to be stripped.

[0053] Image feature vectors are obtained based on product standard images; For example, in some embodiments of this application, the standard product image is first preprocessed, such as scaling the image to a fixed size to ensure consistent input image size, converting the image to grayscale, and scaling the pixel values ​​from an integer range of 0-255 to a floating-point range of [0, 1] or [-1, 1]. Then, a pre-trained convolutional neural network (such as VGG, ResNET, etc.) is used to automatically learn the hierarchical features of the image, extracting visual features such as texture and shape, and performing feature normalization and fusion with calculated shape complexity factors, occlusion coefficients, sphericity, and other data. Finally, the output of an intermediate layer of the model is selected as the vector representation of the image. For example, the generated image feature vector can be generated using a VGG16 model with a dimension of 4096.

[0054] The shape complexity factor is represented by the ratio of the surface area of ​​the product to be stripped to the surface area of ​​the smallest circumscribed sphere. The ratio is positively correlated with the shape complexity. The larger the ratio, the more irregular and complex the shape of the product to be stripped. Naturally, the more difficult it is to achieve uniformity in the flow of the stripping solution and ion diffusion, resulting in a slower reaction rate and a longer reaction time.

[0055] The shading coefficient is determined by identifying deep holes and narrow slits in the image and expressed as the ratio of opening size to depth. A smaller ratio indicates weaker absorption capacity of the stripping solution, more difficult ion diffusion, and a greater likelihood of forming stripping dead zones. Furthermore, multiple thresholds can be set, and the shading coefficient can be assigned values ​​based on the differences between these thresholds.

[0056] Sphericity is used to quantify the topological and geometric characteristics of the product to be stripped. If the product to be stripped has a small sphericity, it is more likely to produce a significant edge effect during the stripping process, which will have a certain impact on the stripping effect and stripping rate.

[0057] Generate text feature vectors based on image description information; The text feature vectors are generated based on a pre-trained deep learning model. The pre-trained model is trained based on the Transformer architecture and can be any of the following: an encoder-only model, a decoder-only model, or an encoder-decoder model.

[0058] More specifically, in some embodiments of this application, the model used to generate text feature vectors may be a neural network model of the Qwen3-Embedding series, EmbeddingGemma, or Seed1.6-Embedding.

[0059] By combining coating thickness information with image description information, physical constraints are obtained; Specifically, in some embodiments of this application, key features, such as the shape and outline of the coating, texture state and defect areas, are first extracted from the image description information; then, the total amount of coating metal of the product to be stripped can be calculated by combining the coating thickness information; finally, the total amount of coating metal is used as a physical constraint.

[0060] Physical constraints, used to guide model training or as inherent rules limiting the simulation of the stripping process, can greatly enhance the model's rationality and generalization ability, improving the accuracy and reliability of the stripping process simulation. For example, based on the coating thickness and image morphology, the total amount of coating metal can be obtained; the amount of coating metal ions dissolved in the stripping solution will not exceed this total amount, and it also helps to further predict the stripping completion time. As a kind of "prior knowledge," physical constraints greatly reduce the model's search space, guiding the model to learn within the correct physical laws, and can also improve the level of intelligence in coating quality monitoring and process control.

[0061] By weighted fusion of numerical feature vectors, image feature vectors, and text feature vectors, and combined with physical constraints, product feature enhancement information is obtained.

[0062] For example, in some embodiments of this application, numerical feature vectors, image feature vectors and text feature vectors can be concatenated to form a longer vector; or weights of numerical feature vectors, image feature vectors and text feature vectors can be preset, with the total weights being 1, and then weighted summed; or element-wise multiplication can be used, that is, multiplying corresponding elements of vectors of the same dimension to emphasize the parts of multiple features that are activated at the same time, thereby achieving the effect of interactive enhancement.

[0063] Specifically, in some embodiments of this application, the model dynamically learns weights and utilizes an attention mechanism to evaluate the importance of each part in the modal features, thereby generating weights for corresponding numerical feature vectors, image feature vectors, and text feature vectors. Further, based on Transformer fusion, the features of numerical feature vectors, image feature vectors, and text feature vectors from different modalities can be transformed into a series of tokens, which are then deeply fused in a deep network through their internal self-attention and cross-attention mechanisms. Physical constraints can be further translated into mathematical language to constrain the simulation results of the stripping process, ensuring that the simulated information matches the actual condition of the product to be stripped. Physical constraints can also be added to the model as a custom network layer.

[0064] The weighted and fused numerical feature vector, image feature vector, and text feature vector are combined with physical constraints to form product feature enhancement information.

[0065] In an optional embodiment, the basic parameters of the stripping process are integrated with the simulation information of the stripping process to obtain the baseline information of the stripping process, including: Based on the relationship between the basic parameters of the stripping process and the corresponding simulation information of the stripping process, the initial reference information of the stripping process is obtained by integrating them. By using a preset optimization objective, the initial baseline information of the stripping process is iteratively updated based on a genetic algorithm to obtain the baseline information of the stripping process.

[0066] Among them, there may still be multiple plating removal schemes after matching for the same product to be stripped of plating, and the specific basic parameters of plating removal in each scheme are different. Therefore, a large amount of initial reference information for the plating removal process corresponding to different plating removal schemes will be generated. It is necessary to screen, optimize and combine the initial reference information of the plating removal process to obtain the reference information of the plating removal process.

[0067] Specifically, the process of obtaining the baseline information for the stripping process by iteratively updating the initial baseline information based on the genetic algorithm according to the preset optimization objective includes: First, determine the optimization objective, such as maximizing stripping efficiency or minimizing cost, and use the basic stripping parameters in the initial baseline information of the stripping process as explicit decision variables, such as stripping temperature, stripping time, current density, stripping solution formula, and stripping solution concentration; and use encoding methods such as binary or real number encoding to represent these variables as chromosomes.

[0068] Secondly, design a fitness function to measure the quality of basic parameters for stripping. For example, the fitness function can be designed as fitness f = w1 × stripping efficiency - w2 × cost. w1 is the weight for optimizing the target stripping efficiency, and w2 is the weight for optimizing the target cost.

[0069] The weighting of stripping efficiency and cost can be adaptively set according to the priority or importance of the desired outcome.

[0070] Next, the iterative optimization algorithm begins to iterate, gradually eliminating undesirable basic parameters for deplating and combining and optimizing excellent basic parameters for deplating. Convergence curves can be plotted to observe the change in fitness with each generation and to judge the optimization process.

[0071] Finally, when the maximum number of iterations is reached or the fitness is stable, the chromosome with the highest fitness in the final population is decoded to obtain the basic parameters of the deplating process, and the initial reference information of the deplating process corresponding to the basic parameters of the deplating process is used as the reference information of the deplating process.

[0072] Understandably, iteratively updating the initial baseline information of the stripping process using a genetic algorithm yields baseline information for the stripping process. This leverages the powerful global search capability of the genetic algorithm to efficiently search within a wide range of parameter combinations of the initial baseline information, avoiding getting trapped in local optima and contributing to finding a globally optimal or near-optimal stripping scheme. Furthermore, the genetic algorithm does not depend on the mathematical properties of the objective function, making it suitable for handling the complex nonlinear relationships involving multiple coupled parameters and intricate mechanisms in the stripping process.

[0073] In an optional embodiment, by combining baseline information from the stripping process with multimodal real-time stripping data, the dynamic adjustment parameters for the stripping environment are obtained, including: The first time point is determined as the time node for acquiring multimodal real-time deplating data. Obtain the first stripping process simulation information corresponding to the first time node from the stripping process reference information; The reaction rate information and target metal concentration information in the multimodal real-time stripping data are obtained to obtain the first real-time stripping data; The simulated deviation reference value is obtained based on the first real-time stripping data and the simulation information of the first stripping process.

[0074] Of course, the time node for acquiring multimodal real-time stripping data may not completely coincide with the key time node of the stripping process reference information. The first time node may be located between two key time nodes. The simulation information of the first stripping process can be obtained by further acquiring simulation data, or the simulation information of the first stripping process can be inferred based on the distance between the first time node and the two key time nodes and the simulation information of the stripping process at the two key time nodes.

[0075] Specifically, the reaction rate and target metal concentration can be extracted from the simulation information of the first stripping process. The reaction rate in the simulation information is compared with the reaction rate in the first real-time stripping data to obtain a first deviation rate. The target metal concentration in the simulation information is compared with the target metal concentration in the first real-time stripping data to obtain a second deviation rate. The first and second deviation rates are normalized and combined to obtain a simulation deviation reference value. The target metal corresponds to the type of plating metal, such as one or more of platinum, gold, silver, copper, nickel, and chromium.

[0076] For example, if the stripping rate of the first real-time stripping data is 3.2 μm / min, and the stripping rate of the corresponding simulated first stripping process information is 5 μm / min, then the first deviation rate can be calculated to be 0.3; if the target metal concentration of the first real-time stripping data is 0.7 g / L, and the target metal concentration of the corresponding simulated first stripping process information is 0.6 g / L, then the first deviation rate can be calculated to be 0.3; then the first deviation rate can be calculated to be 0.167; if the weights of the first deviation rate and the second deviation rate are set to be the same, then the simulated deviation reference value is 0.2335.

[0077] The real-time images, stripping temperature, and pH value of the stripped product are obtained from the multimodal real-time stripping data to obtain the second real-time stripping data; The simulated development trend value was obtained based on the second real-time stripping data; Based on the simulated deviation from the reference value and the simulated development trend value, the dynamic adjustment parameters of the stripping environment are obtained.

[0078] Specifically, the aforementioned "Product-Process-Result" experience database can be accessed, and case information matching the second real-time stripping data can be found within the database. Based on this case information, a simulated trend value can be predicted. Furthermore, expert experience can be used to make secondary adjustments to the simulated trend value. The simulated trend value can be positive or negative; a positive value indicates an expansion of the simulated deviation from the reference value, while a negative value indicates a reduction in the simulated deviation from the reference value. Expert experience can be stored in an expert experience database.

[0079] Based on the simulated deviation from the reference value and the simulated development trend value, the control strategy used in obtaining the dynamic adjustment parameters of the stripping environment can be one of PID control, fuzzy logic control, or a combination of fuzzy logic control strategies. Of course, there are no specific restrictions on the specific control strategy used. The simulated development trend value can be obtained through theoretical model prediction, and the theoretical model can be trained using cases from the "product-process-procedure and result" experience database and expert experience from the expert experience database as datasets.

[0080] More specifically, in some embodiments of this application, a PID control strategy is employed in the process of obtaining the dynamic adjustment parameters of the stripping environment based on the dynamic adjustment feature vector of the stripping environment. The proportional component P in the PID controller is used to adjust the parameters proportionally according to the simulated deviation from the reference value. The integral component I is used to accumulate the total deviation over a period of time, eliminating small, persistent deviations and enabling the system to accurately stabilize at the target value. The derivative component D is used to adjust according to the simulated trend value, predicting future deviation trends in advance and applying inhibitory control to improve stability. The PID controller combines the rapid response of P, the precise elimination of steady-state error by I, and the anticipatory prediction by D in the three components to jointly calculate the adjustment amount. For example, if the trend value predicts that the temperature will rise too quickly, the heating power can be reduced in advance to regulate the stripping temperature.

[0081] In an optional embodiment, the dynamic adjustment parameters of the stripping environment are obtained based on the simulated deviation from the reference value and the simulated development trend value, including: The first weight of the simulated deviation from the reference value and the second weight of the simulated development trend value are obtained through a preset node weight mapping table. The dynamic adjustment feature vector of the plating removal environment is obtained by weighted fusion of the first weight and the simulated deviation reference value, and the second weight and the simulated development trend value. The dynamic adjustment parameters of the stripping environment are obtained based on the feature vector of the dynamic adjustment of the stripping environment. The controllable objects of the dynamic adjustment parameters of the stripping environment include one or more of the following: stripping time adjustment, stripping solution flow rate, regenerator replenishment rate, stripping environment temperature, and electrowinning start / stop.

[0082] The node weight mapping table stores multiple key time nodes, and each key time node is assigned a first weight for the simulated deviation from the reference value and a second weight for the simulated development trend value.

[0083] For example, during the stripping process of a product to be stripped, the first weight can be set to approach 0 and the second weight to approach 1 for the first 20% of the time, so that the stripping system focuses on the correct trend of change and ignores minor deviations; while in the latter 15% of the time, the weight value of the first weight is increased and the weight value of the second weight is decreased to reduce the situation of excessive corrosion or waste of resources.

[0084] Specifically, in some embodiments of this application, the first weight and the second weight can be determined through simulation optimization based on historical data stored in an empirical database and parameters such as steepness and midpoint of the transition function. Furthermore, parameters such as steepness and midpoint can be trained using reinforcement learning to maximize the final product yield or minimize the risk of over-corrosion.

[0085] Understandably, the dynamic adjustment parameters for the stripping environment can also be empty or remain unchanged, meaning there is no deviation or a small deviation from the simulated situation, making simulation unnecessary. Of course, an adjustment threshold can be set, and adjustments are only made when the simulated deviation from the reference value exceeds this threshold. For example, the adjustment threshold can be 0.05. If the simulated deviation from the reference value is less than 0.05, then no adjustment of the stripping environment parameters is needed at that time point.

[0086] Furthermore, the dynamic adjustment parameters of the stripping environment can be obtained from the output of the trained dynamic adjustment parameter model of the stripping environment. The dynamic adjustment parameter model of the stripping environment is essentially a multi-input, single-output intelligent prediction system, which aims to map two input variables to the optimal adjustment parameters.

[0087] The dynamic adjustment parameter model for the decoating environment has two input ports, which receive normalized simulated deviation from the reference value and simulated development trend value, respectively. The feature fusion layer of the dynamic adjustment parameter model for the decoating environment concatenates or weights the two input vectors to fuse them into a comprehensive feature vector (dynamic adjustment feature vector for the decoating environment) that contains the current state and future trend, providing complete information for subsequent decision-making. The weights used can be the first weight and the second weight in the aforementioned node weight mapping table.

[0088] The hidden layers of the dynamic adjustment parameter model for the stripping environment can use recurrent gated units (GRUs) or long short-term memory networks (LSTMs) to capture the time dependencies during the stripping process. The output layer of the model outputs one or more dynamic adjustment parameters, such as stripping time adjustment, stripping solution flow rate, regenerator replenishment rate, stripping environment temperature, and electrodeposition start / stop. Depending on the parameter characteristics, linear activation functions or Sigmoid / Tanh functions can be used.

[0089] The weighted fusion of simulated deviation from the reference value and simulated development trend value can be calculated before being input into the dynamic adjustment parameter model of the stripping environment, and the calculation result is used as the input to the model. The dynamic adjustment parameter model of the stripping environment has three input ports, taking the simulated deviation from the reference value, the simulated development trend value, and multimodal real-time stripping data as data inputs, and obtaining the dynamic adjustment parameters of the stripping environment based on the dynamic adjustment feature vector of the stripping environment and combined with the multimodal real-time stripping data.

[0090] The data sources for training the dynamic adjustment parameter model of the stripping environment can come from the empirical database of "product-process-process and result", as well as data generated by simulation through theoretical models and / or empirical models; physical constraints and federated learning frameworks can also be further embedded in the dynamic adjustment parameter model of the stripping environment.

[0091] In an optional embodiment, the method further includes: After the plating removal process is completed, a plating removal effect evaluation report is generated based on the baseline information of the plating removal process and the dynamic adjustment parameters of the plating removal environment.

[0092] A pre-set template for stripping effect evaluation report is provided. After the stripping of the product is completed, the system automatically generates or guides the user to fill out the report based on the template. The stripping effect evaluation report can include images, names, and models of the product to be stripped, multimodal real-time stripping data at various time points, simulated deviation reference values, simulated trend values, and dynamic adjustment parameters of the stripping environment, as well as one or more of the stripping effect score and simulation accuracy. The stripping effect evaluation report and related data are stored in the "Product-Process-Process and Results" experience database.

[0093] This application provides a method for dynamic control of the metal stripping environment, which can perform differentiated stripping treatment for different products to be stripped and achieve effective control of the stripping environment, thereby effectively improving the stripping effect and stripping stability. Specifically, the method of this application simulates the baseline information of the stripping process based on the basic information of the product to be stripped, and performs stripping treatment based on the baseline information of the stripping process; the stripping treatment method is set according to the actual situation of the product, which can realize differentiated treatment for different products to be stripped; and the stripping status is acquired in real time during the stripping process, and the stripping environment is dynamically and adaptively adjusted according to the actual situation during the stripping process, thereby achieving good stripping efficiency, stripping stability and stripping effect.

[0094] Example 2 Figure 2 This is a schematic diagram of the module of the dynamic control device for metal stripping environment provided in Embodiment 2 of this application.

[0095] In some embodiments, the dynamic control device for the metal stripping environment may include multiple functional modules composed of computer program segments. The computer programs for each program segment of the dynamic control device for the metal stripping environment may be stored in the memory of an electronic device and executed by at least one processor to perform (see details). Figure 1 (Description) Function of dynamic control of metal stripping environment.

[0096] Based on the functions they perform, the system can be divided into multiple functional modules. The system's functional modules may include: a first information acquisition module 1, a stripping process simulation module 2, a second information acquisition module 3, a stripping parameter acquisition module 4, and an environmental dynamic control module 5. The term "module" in this application refers to a series of computer program segments that can be executed by at least one processor and perform a fixed function, stored in memory. In this embodiment, the functions of each module will be detailed in subsequent embodiments.

[0097] The system includes: The first information acquisition module 1 is used to acquire basic information about the product to be deplated. The stripping process simulation module 2 is used to simulate the baseline information of the stripping process based on the basic information of the product to be stripped. The second information acquisition module 3 is used to perform stripping processing based on the reference information of the stripping process and dynamically acquire multimodal real-time stripping data. The stripping parameter acquisition module 4 is used to combine the baseline information of the stripping process with the multimodal real-time stripping data to obtain the dynamic adjustment parameters of the stripping environment; The environmental dynamic control module 5 is used to control the stripping environment based on the dynamic adjustment parameters of the stripping environment.

[0098] like Figure 3 As shown, the system also includes: The stripping report generation module 6 generates a stripping effect evaluation report based on the baseline information of the stripping process and the dynamic adjustment parameters of the stripping environment after the stripping of the product is completed.

[0099] It should be understood that the various variations and specific embodiments of the metal stripping environment dynamic control method provided in Embodiment 1 above are also applicable to the metal stripping environment dynamic control device in this embodiment. Through the detailed description of the aforementioned metal stripping environment dynamic control method, those skilled in the art can clearly understand the implementation process of the metal stripping environment dynamic control device in this embodiment. For the sake of brevity, it will not be described in detail here.

[0100] Example 3 See Figure 4 The diagram shown is a structural schematic of an electronic device according to an embodiment of this application.

[0101] Electronic device 7 includes a memory 71, at least one processor 72, and at least one communication bus 73. Those skilled in the art should understand that... Figure 4 The structure of the electronic device shown does not constitute a limitation of the embodiments of this application. It can be a bus structure or a star structure. The electronic device may also include more or fewer other hardware or software than shown, or different component arrangements.

[0102] In some embodiments, the electronic device is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions. Its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), digital processors, and embedded devices. The electronic device may also include user equipment, which includes, but is not limited to, any electronic product capable of human-computer interaction with a user via a keyboard, mouse, remote control, touchpad, or voice control device, such as a personal computer, tablet computer, smartphone, or digital camera.

[0103] In the embodiments provided in this application, it should be understood that the disclosed methods, apparatuses, computer-readable storage media, and electronic devices can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple components or modules may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices, components, or modules may be electrical, mechanical, or other forms.

[0104] The components described as separate parts may or may not be physically separate. The components shown as components may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the components can be selected to achieve the purpose of this embodiment according to actual needs.

[0105] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing module, or each component can exist physically separately, or two or more modules can be integrated into one module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0106] If the integrated module is implemented as a software functional module 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, in essence, or the part 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 to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the intelligent data backup method for electronic devices described in the various embodiments of the present invention. The aforementioned storage medium includes: USB flash drive, portable hard drive, read-only memory (ROM). Various media that can store program code, such as only memory, random access memory (RAM), magnetic disks, or optical disks.

[0107] Example 4 This application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the metal stripping environment dynamic control method according to Embodiment 1.

[0108] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that the present invention is not limited to the described order of actions, because according to the present invention, some steps can be performed in other orders or simultaneously. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to the present invention.

[0109] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0110] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A method for dynamically controlling the metal stripping environment, characterized in that, The method includes: Obtain basic information about the products to be de-plated; Based on the basic information of the product to be stripped, the baseline information of the stripping process is simulated. The plating removal process is performed based on the reference information of the plating removal process, and multimodal real-time plating removal data is dynamically acquired. By combining the reference information of the stripping process with the multimodal real-time stripping data, dynamic adjustment parameters for the stripping environment are obtained; The stripping environment is regulated based on the aforementioned dynamic adjustment parameters.

2. The method for dynamic control of the metal stripping environment according to claim 1, characterized in that, The baseline information for simulating the plating removal process based on the basic information of the product to be stripped includes: Based on the basic information of the product to be deplated, detailed information on product features is obtained; Based on the basic information of the product to be stripped and the product feature enhancement information, the basic parameters of stripping are obtained. The basic parameters of stripping include one or any combination of stripping solution formulation, stripping temperature and current density. Based on the product feature enhancement information and the basic parameters of the stripping process, the stripping process simulation information is obtained. The stripping process simulation information includes one or more of the key time nodes, reaction rate, and target metal concentration. The basic parameters of the stripping process are integrated with the simulation information of the stripping process to obtain the baseline information of the stripping process.

3. The method for dynamic control of the metal stripping environment according to claim 2, characterized in that, The process of obtaining product feature refinement information based on the basic information of the product to be deplated includes: Obtain the substrate type information, coating metal type information, coating thickness information, product standard image and image description information from the basic information of the product to be deplated; Based on the substrate type information and the coating metal type information, material property information is obtained, and the material property information is converted into a numerical feature vector; Obtain image feature vectors based on the product standard image; Generate text feature vectors based on the image description information; By combining the coating thickness information with the image description information, physical constraints are obtained; The numerical feature vector, the image feature vector, and the text feature vector are weighted and fused together, and combined with the physical constraints, to obtain product feature enhancement information.

4. The method for dynamic control of the metal stripping environment according to claim 2, characterized in that, The process of integrating the basic parameters of the plating stripping process with the simulation information of the plating stripping process to obtain the baseline information of the plating stripping process includes: Based on the relationship between the basic parameters of the stripping process and the corresponding simulation information of the stripping process, the initial reference information of the stripping process is obtained by integrating them. By using a preset optimization objective, the initial baseline information of the stripping process is iteratively updated according to a genetic algorithm to obtain the baseline information of the stripping process.

5. The method for dynamic control of the metal stripping environment according to claim 4, characterized in that, The dynamic adjustment parameters for the stripping environment obtained by combining the baseline information of the stripping process with the multimodal real-time stripping data include: The time node for acquiring the multimodal real-time deplating data is determined as the first time node; Obtain the first stripping process simulation information corresponding to the first time node from the stripping process reference information; The reaction rate information and target metal concentration information in the multimodal real-time stripping data are obtained to obtain the first real-time stripping data; Based on the first real-time stripping data and the first stripping process simulation information, a simulated deviation reference value is obtained; The real-time image of the stripped product, stripping temperature, and pH value are obtained from the multimodal real-time stripping data to obtain the second real-time stripping data; The simulated development trend value is obtained based on the second real-time deplating data; Based on the simulated deviation reference value and the simulated development trend value, the dynamic adjustment parameters of the stripping environment are obtained.

6. The method for dynamic control of the metal stripping environment according to claim 5, characterized in that, The dynamic adjustment parameters for the stripping environment, obtained based on the simulated deviation reference value and the simulated development trend value, include: The first weight of the simulated deviation from the reference value and the second weight of the simulated development trend value are obtained through a preset node weight mapping table. The dynamic adjustment feature vector of the plating removal environment is obtained by weighting and fusing the first weight with the simulated deviation reference value and the second weight with the simulated development trend value; The dynamic adjustment parameters of the plating removal environment are obtained based on the dynamic adjustment feature vector of the plating removal environment. The control objects of the dynamic adjustment parameters of the stripping environment include one or any combination of stripping time adjustment, stripping solution flow rate, regenerator replenishment rate, stripping environment temperature, and electrowinning start / stop.

7. The method for dynamic control of the metal stripping environment according to any one of claims 1 to 6, characterized in that, The method further includes: After the plating on the product to be stripped is completed, a plating effect evaluation report is generated based on the baseline information of the plating stripping process and the dynamic adjustment parameters of the plating stripping environment.

8. A dynamic control system for the metal stripping environment, characterized in that, The system includes: The first information acquisition module is used to acquire basic information about the products to be deplated. The stripping process simulation module is used to simulate the baseline information of the stripping process based on the basic information of the product to be stripped. The second information acquisition module is used to perform stripping processing based on the stripping process reference information and dynamically acquire multimodal real-time stripping data. The stripping parameter acquisition module is used to combine the stripping process reference information with the multimodal real-time stripping data to obtain the stripping environment dynamic adjustment parameters; The environmental dynamic control module is used to control the stripping environment based on the aforementioned stripping environment dynamic adjustment parameters.

9. An electronic device, characterized in that, The electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the steps of the dynamic control method for metal stripping environment according to any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the dynamic control method for metal stripping environment according to any one of claims 1 to 7.

Images