Monitoring and regulation system for preparation process of aluminum electrolytic capacitor electrode foil

Abstract

The invention belongs to the technical field of automatic control, and relates to a monitoring and regulation system for an aluminum electrolytic capacitor electrode foil preparation process, which comprises a dynamic thermal field on-line acquisition module used for generating a two-dimensional thermal distribution image flow; a reaction non-uniformity feature extraction module for analyzing the image stream to extract a reaction non-uniformity feature vector; the cathode configuration regulation and control strategy module is used for outputting a target inter-electrode distance compensation value of each cathode section; the multi-channel driving instruction generation module is used for generating a multi-channel mechanical driving instruction according to the compensation value; the segmented cathode array reconstruction module responds to an instruction to drive each cathode segment to displace, recombines the geometrical configuration of the cathode array and constructs a physical electric field with adjustable space distribution; the feedback locking module is used for locking the cathode configuration; the method solves the problems that static regulation and control are difficult to respond to reaction non-uniformity in real time, so that the spatial difference of the microstructure of the electrode foil is caused, and the product performance and reliability are influenced.

Description

Technical Field

[0001] This invention belongs to the technical field of automation control and relates to a monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil. Background Technology

[0002] In the preparation of aluminum electrolytic capacitor electrode foil, especially in the electrochemical etching and formation processes, the electrode foil needs to be immersed in the electrolyte and subjected to voltage to react. This process requires ensuring that the electrochemical environment experienced by all regions of the electrode foil along its width is consistent. Due to factors such as limitations of the electrolytic cell structure, the concentration effect of electric field lines at the electrode edges, and uneven electrolyte flow and concentration distribution, the current density on the electrode foil surface often exhibits a spatially non-uniform distribution, resulting in a significant difference in reaction rates between the central and edge regions, known as the edge effect.

[0003] To address the aforementioned non-uniformity issues, existing technologies generally employ measures such as optimizing electrolyte formulations, improving electrolyte circulation system design, and using static cathode structures with auxiliary electrodes or shielding plates. These methods aim to indirectly alleviate reaction non-uniformity by optimizing the macroscopic electrolysis environment or electric field distribution. They belong to static or open-loop control methods, and once the process equipment is installed and commissioned, its electric field distribution is basically fixed in a single production process, making it difficult to dynamically adjust according to real-time process fluctuations.

[0004] However, in actual long-term operation, process parameters such as electrolyte composition, temperature, and electrode state inevitably fluctuate. The aforementioned static control methods are insufficient to respond to real-time changes in reaction inhomogeneity. This causes the current density in local areas of the electrode foil to continuously deviate from the ideal value, resulting in spatial differences in the microstructure: over-reacting areas may experience excessive pit growth or even perforation due to excessively high current density, while under-reacting areas may suffer from insufficient pit depth or an excessively thin oxide film due to excessively low current density. Such inconsistencies in microstructure severely affect the specific capacitance performance and reliability of the electrode foil. The fundamental reason lies in the lack of online direct sensing capability for the spatial non-uniformity of the electrochemical reaction field, and the absence of a dynamic control mechanism based on sensing information in existing technologies. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides a monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil.

[0006] A monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil, comprising: The dynamic thermal field online acquisition module is used to acquire dynamic thermal field data of the moving electrode foil surface and generate a two-dimensional thermal distribution image stream. The reaction non-uniformity feature extraction module is used to analyze the two-dimensional heat distribution image stream, extract indicators that characterize the local differences in electrochemical reaction, and generate a reaction non-uniformity feature vector. The cathode configuration control strategy module, based on the reaction non-uniformity feature vector, performs inversion calculations through an electrochemical-thermal coupling control model and outputs the target electrode spacing compensation value for each independent cathode segment in the segmented cathode array. A multi-channel drive command generation module is used to generate multi-channel mechanical drive commands based on the target pole spacing compensation value; The segmented cathode array reconfiguration module is used to respond to multi-channel mechanical drive commands, drive each independent cathode segment to generate physical displacement, reconfigure the geometric configuration of the segmented cathode array, and construct a spatially adjustable physical electric field. The feedback locking module corrects the local current density flowing through different regions of the electrode foil based on the spatially adjustable physical electric field, continuously monitors the changes in the thermal field, and locks the current cathode array configuration when the thermal distribution gradient norm falls back to the preset uniformity judgment threshold.

[0007] In a further embodiment of the present invention, the dynamic thermal field online acquisition module is specifically configured to perform the following operations: The infrared thermal imaging acquisition unit is used to capture the infrared radiation signal emitted by the surface of the electrode foil immersed in the electrolyte in real time. The infrared radiation signal is mapped into a two-dimensional temperature matrix arranged in time order to form the raw data of the dynamic thermal field. The original dynamic thermal field data is subjected to joint time-domain and spatial-domain filtering and geometric correction to generate a two-dimensional thermal distribution image stream.

[0008] In a further embodiment of the present invention, the reaction unevenness feature extraction module is specifically configured to perform the following operations: Each frame of two-dimensional thermal distribution image is divided into multiple virtual monitoring zones along the width direction of the electrode foil; Calculate the average temperature value, temperature gradient norm, and migration rate of hotspot patches within each virtual monitoring zone; Based on the average temperature value, temperature gradient norm, and migration velocity, combined with the temperature difference between the edge and the center and the thermal field gradient dispersion, a response non-uniformity feature vector is generated.

[0009] In a further embodiment of the present invention, the cathode configuration control strategy module is specifically configured to perform the following operations: Based on the deviation between the average temperature value of each zone in the reaction non-uniformity feature vector and the preset benchmark value, the over-reaction zone and the under-reaction zone are identified. Calculate the positive inter-electrode spacing compensation value for the overreaction zone and the negative inter-electrode spacing compensation value for the underreaction zone to generate the target inter-electrode spacing compensation value; For partitions that are not identified as abnormal, their target interpole spacing compensation value is set to zero.

[0010] In a further embodiment of the present invention, the multi-channel drive instruction generation module is specifically configured to perform the following operations: Calculate the target rotation angle or linear travel of each drive axis based on the type of servo actuator; Generate independent pulse control signals for each independent cathode segment, and add a unified action synchronization timestamp to all signals; Integrate the pulse control signals from each channel with the synchronization timestamp to generate multi-channel mechanical drive commands.

[0011] In a further embodiment of the present invention, the segmented cathode array reconfiguration module is specifically configured to perform the following operations: Each independent cathode segment is driven to produce independent displacement in a direction perpendicular to the electrode foil surface by a servo actuator; The electrolyte resistance distribution in the corresponding area is adjusted by changing the local anode-cathode spacing of each partition. Under constant power supply voltage, a spatially adjustable physical electric field is generated based on the adjusted electrolyte resistance distribution.

[0012] In a further embodiment of the present invention, the feedback locking module is specifically configured to perform the following operations: Continuously acquire new two-dimensional thermal distribution image streams and calculate their thermal distribution gradient norm in real time; Determine whether the gradient norm continues to decrease and falls below a preset uniformity threshold; If so, a locking command is sent to the servo actuator to lock and maintain the current cathode configuration and electric field distribution.

[0013] A further aspect of the present invention involves preprocessing the raw dynamic thermal field data, including: A Gaussian low-pass filter is used to smooth the data to suppress random temperature fluctuations caused by environmental radiation fluctuations or detector noise. Based on the optical distortion parameters of the infrared thermal imaging acquisition unit lens and its mounting pitch angle relative to the electrode foil plane, the temperature distribution image of each frame is corrected using an image transformation algorithm.

[0014] In a further embodiment of the present invention, the virtual monitoring zone includes at least a left edge region, a central main body region, and a right edge region, and the width of each region is set based on the total width of the electrode foil and engineering experience.

[0015] In a further embodiment of the present invention, the migration speed of hot spot patches is obtained by identifying the centroid displacement of high-temperature connected regions in consecutive frames and calculating it in combination with time intervals.

[0016] In summary, the present invention has the following beneficial technical effects: 1. By using non-contact infrared thermal imaging technology to collect dynamic thermal field data on the surface of the electrode foil online and generating a two-dimensional thermal distribution image stream, full-width and continuous monitoring of the spatial differences in the degree of reaction on the foil surface is realized, replacing the traditional offline sampling or single-point temperature measurement method, and providing a real-time data basis reflecting the local current density distribution for subsequent analysis and control.

[0017] 2. Based on the eigenvector of reaction non-uniformity, the electrode spacing compensation value of each cathode segment is calculated by inversion through the electrochemical-thermal coupling control model, and the segmented cathode array is driven to perform independent displacement adjustment. Under constant power supply voltage, the electrolyte resistance distribution can be adjusted by changing the local electrode spacing, thereby realizing the adaptive reconstruction of the physical electric field spatial distribution.

[0018] 3. By specifically modifying the local current density, the electrochemical reaction is guided towards spatial self-equilibrium, resulting in a more uniform temperature distribution on the electrode foil surface. Under these conditions, the corrosion tunnel growth or oxide film formation processes in different regions of the electrode foil tend to be synchronized, which helps improve the structural consistency of the finished electrode foil in terms of pore morphology, oxide film thickness, etc., thereby supporting the stability of the product's electrical performance. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. The drawings are used to provide a further understanding of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This discloses a schematic diagram of the framework in the embodiments of this application.

[0021] Figure 2 This discloses a flowchart of an embodiment of this application. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] The following is in conjunction with the appendix Figure 1 - Figure 2 A preferred description of the present invention is provided below.

[0024] See attached document Figure 1 - Figure 2This invention proposes a monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil, comprising the following modules: The system comprises the following modules: a dynamic thermal field online acquisition module for acquiring dynamic thermal field data of the moving electrode foil surface and generating a two-dimensional thermal distribution image stream; a reaction non-uniformity feature extraction module for analyzing the two-dimensional thermal distribution image stream, extracting indicators characterizing the local differences in electrochemical reaction, and generating a reaction non-uniformity feature vector; a cathode configuration control strategy module for performing inversion calculations based on the reaction non-uniformity feature vector using an electrochemical-thermal coupling control model, and outputting target inter-electrode spacing compensation values ​​for each independent cathode segment in the segmented cathode array; a multi-channel drive command generation module for generating multi-channel mechanical drive commands based on the target inter-electrode spacing compensation values; a segmented cathode array reconstruction module for responding to the multi-channel mechanical drive commands, driving each independent cathode segment to generate physical displacement, reconstructing the geometric configuration of the segmented cathode array, and constructing a spatially adjustable physical electric field; and a feedback locking module for correcting the local current density flowing through different regions of the electrode foil based on the spatially adjustable physical electric field, continuously monitoring thermal field changes, and locking the current cathode array configuration when the thermal distribution gradient norm falls back to within a preset uniformity judgment threshold.

[0025] In one embodiment of the present invention, the dynamic thermal field online acquisition module is specifically configured to perform the following operations: Infrared radiation signals emitted from the surface of electrode foil immersed in electrolyte are captured in real time using an infrared thermal imaging acquisition unit. The infrared radiation signals are mapped into a two-dimensional temperature matrix arranged in time sequence to form the original data of the dynamic thermal field. The original data of the dynamic thermal field is subjected to joint time-domain and spatial-domain filtering and geometric correction to generate a two-dimensional thermal distribution image stream.

[0026] Specifically, an infrared thermal imaging acquisition unit installed above the corrosion tank or formation tank operates continuously at a preset sampling frequency. The sampling frequency is set according to the production line speed to ensure continuous and uninterrupted capture of the same foil surface area. The installation position of the infrared thermal imaging acquisition unit ensures that its optical lens axis is perpendicular to the moving electrode foil surface below, and its field of view covers the entire width range and a certain travel length of the electrode foil. The infrared thermal imaging acquisition unit is the core component of the equipment. It is an integrated module that embeds an infrared detector focal plane array, optical lens, signal amplification and analog-to-digital conversion circuit, and a preliminary processor. It is used to convert the infrared radiation energy distribution of the object surface into a digital grayscale or pseudo-color image. The settings are based on the standard technical specifications of industrial-grade infrared thermal imagers, such as selecting long-wave infrared sensors with wavelengths of 8 to 14 micrometers to adapt to the measurement range from near room temperature to several hundred degrees Celsius.

[0027] During sampling, the infrared detector array in the acquisition unit receives the infrared electromagnetic wave signal radiated from the surface of the electrode foil below the electrolyte surface and converts the analog signal into a digital electrical signal. The surface of the electrode foil below the electrolyte surface is the surface of the aluminum foil strip immersed in the corrosive or chemical electrolyte and undergoing an electrochemical reaction. It is the solid-liquid interface where the reaction occurs, and its temperature distribution is affected by the exothermic reaction, Joule heating, and convective heat transfer of the electrolyte. The infrared radiation signal is the electromagnetic wave continuously emitted outward from the surface of the electrode foil because its temperature is above absolute zero. Its radiation intensity and surface temperature follow the Stefan-Boltzmann law, and its wavelength is mainly located in the mid- and far-infrared band.

[0028] The digital electrical signal is converted into temperature values ​​pixel by pixel by the processor built into the acquisition unit, according to the pre-calibrated radiation intensity-temperature correspondence curve. These values ​​are then organized into a two-dimensional matrix according to the physical arrangement of the detector array. The value of each element in the matrix represents the instantaneous temperature of a corresponding area on the surface of the electrode foil. As time progresses, such two-dimensional temperature matrices are continuously generated and stacked in the storage buffer in chronological order, thus forming a data cube with three dimensions. The two spatial dimensions characterize the temperature distribution on the foil surface, and the temporal dimension characterizes the evolution of this distribution with the process. This data cube is the raw dynamic thermal field data. It should be noted that the two-dimensional temperature matrix refers to a data structure output by the infrared thermal imaging acquisition unit. It is a numerical matrix with the number of rows and columns corresponding to the number of rows and columns of the detector array, respectively. The value of each element in the matrix represents the temperature measurement value of the corresponding pixel in the image, and its physical unit is usually Celsius or Kelvin. The raw dynamic thermal field data is an uncorrected collection of two-dimensional temperature matrices stacked in chronological order. It is a three-dimensional array or data stream containing two-dimensional spatial information and one-dimensional temporal information, reflecting the original record of the temperature field evolution.

[0029] Subsequently, the raw data of the dynamic thermal field is preprocessed, including: first, joint temporal and spatial filtering is performed, using a Gaussian low-pass filter to smooth the data and suppress random temperature fluctuations caused by environmental radiation fluctuations or detector noise; then, geometric correction is performed, based on the optical distortion parameters of the infrared thermal imaging acquisition unit lens and its installation pitch angle relative to the electrode foil plane, using image transformation algorithms, such as perspective transformation algorithms, to perform perspective correction on each frame of the temperature distribution image. The algorithm's processing includes: calculating the projection transformation matrix from the image coordinate system to the actual foil surface physical coordinate system based on pre-calibrated lens intrinsic and extrinsic parameters, where the intrinsic parameters include: focal length, principal point, and distortion coefficients; and the extrinsic parameters include the rotation and translation matrices of the lens relative to the foil surface; then, this transformation matrix is ​​applied to the coordinates of each pixel in the image, and the corrected image is generated through bilinear interpolation resampling, eliminating image distortion caused by non-direct viewing angles, and establishing a precise linear mapping relationship between the pixel coordinates in the image and the physical position of the actual electrode foil surface. After filtering and geometric correction, the data is reorganized into a two-dimensional temperature image sequence that is temporally continuous and spatially corresponds to physical locations. This sequence is a two-dimensional thermal distribution image stream, which is a continuous temperature distribution image sequence with time-stamp alignment and spatial coordinate consistency between images. Each frame of the image can be directly used for visual analysis or as input for subsequent steps. Its data format is usually an industry standard image format that includes a temperature value matrix and metadata.

[0030] It should be noted that filtering and geometric correction are two operations that process the raw data of the dynamic thermal field. Filtering removes random noise to improve the signal-to-noise ratio, and is usually implemented using linear or nonlinear digital filters. Geometric correction corrects image geometric distortion caused by optical lens distortion and viewing angle inaccuracies, and is usually achieved through inverse mapping based on calibration parameters or polynomial transformation algorithms. Perspective transformation algorithms are well-known techniques in the fields of computer vision and image processing. They achieve coordinate mapping between the image plane and the physical plane by constructing a projection transformation matrix, and are often used to correct perspective distortion caused by the shooting angle.

[0031] For example, an infrared thermal imager of model FLIRA655sc can be selected as the infrared thermal imaging acquisition unit. It has an infrared resolution of 640 pixels by 480 pixels, a thermal sensitivity of 0.03 degrees Celsius, and a spectral range of 7.5 micrometers to 14 micrometers. The thermal imager is fixed 1.5 meters above the corrosion tank using a rigid bracket, with the lens pointing vertically downwards, ensuring its field of view covers the full width of the electrode foil (1.2 meters) and approximately 0.5 meters along the direction of travel. The sampling frequency of the thermal imager is set to 10 Hz to match the electrode foil's moving speed of 2 meters per minute. The thermal imager is calibrated at the factory using a blackbody radiation source and has a built-in radiation intensity to temperature conversion curve.

[0032] Next, the thermal imager begins acquiring raw radiation intensity data, which is then converted into temperature value matrices in real time by the built-in processor. Each matrix consists of 640 rows and 480 columns. A buffer is created, and 1000 consecutive temperature matrices are stacked in chronological order of acquisition time to form an initial dynamic thermal field raw data cube. Subsequently, a preset image processing library is invoked to apply a spatial Gaussian filter with a standard deviation of 1.5 pixels to the data cube, followed by temporal median filtering to eliminate transient noise. Simultaneously, the lens distortion parameters of the thermal imager, pre-obtained using a checkerboard calibration method, are loaded. Based on the perspective transformation matrix calculated using the installation height and angle, geometric correction is performed on each frame of the temperature image to align the edges of the electrode foil in the image with the actual physical edges, and to ensure that the pixel spacing corresponds to the actual size. After the above processing, a two-dimensional thermal distribution image stream is output. In this image stream, any frame can clearly show the temperature distribution details on the electrode foil surface from left to right and from front to back. For example, it can be observed that the temperature in the central area of ​​the foil is 85.3 degrees Celsius, while the temperature in the left edge area is 92.1 degrees Celsius, forming a clear temperature contrast. This directly verifies that the non-contact monitoring device has successfully acquired dynamic thermal field data reflecting the non-uniformity of the reaction.

[0033] In one embodiment of the present invention, the uneven reaction feature extraction module is specifically configured to perform the following operations: Each frame of two-dimensional thermal distribution image is divided into multiple virtual monitoring zones along the width direction of the electrode foil; the average temperature value, temperature gradient norm, and migration speed of hot spot patches in each virtual monitoring zone are calculated; based on the average temperature value, temperature gradient norm, and migration speed, combined with the temperature difference between the edge and the center and the thermal field gradient dispersion, a response non-uniformity feature vector is generated.

[0034] Specifically, firstly, each frame of the two-dimensional thermal distribution image is partitioned. Based on the actual physical width parameter of the electrode foil, three continuous virtual monitoring zones are defined along the width direction of the image, i.e., the column direction. These virtual monitoring zones are rectangular image regions artificially defined on the image data, corresponding to different physical width segments on the electrode foil surface. They have clearly defined pixel coordinate boundaries and are used for local analysis. Their setting is based on engineering experience regarding the typical influence range of edge effects during electrode foil production; for example, the width of the edge area is usually set to 10% to 15% of the total width. For instance, if the total width of the electrode foil is 1200 mm, then the left edge area corresponds to an image column with a physical width of 0 to 150 mm, the central main area corresponds to an image column with a physical width of 150 mm to 1050 mm, and the right edge area corresponds to an image column with a physical width of 1050 mm to 1200 mm. Based on the mapping relationship between image pixel coordinates and actual physical dimensions, the pixel row range covered by each virtual monitoring zone is automatically calculated.

[0035] Next, for each virtual monitoring zone in each frame of the image, three characterization indicators are calculated simultaneously: The first item is the average temperature value, which is obtained by summing the temperature values ​​of all pixels in the monitoring band and dividing by the total number of pixels. The average temperature value refers to the arithmetic mean of the temperature values ​​of all pixels in the virtual monitoring band, and it is used to characterize the macroscopic thermal intensity of the reaction in this area. The second term is the temperature gradient norm. First, the Sobel operator is used to calculate the temperature gradient components of each pixel in the monitoring zone in the x and y directions. Then, the gradient magnitude of each pixel is calculated. Finally, the average gradient magnitude of all pixels in the monitoring zone is taken as the temperature gradient norm of the region. The temperature gradient norm is an overall measure of the degree of temperature spatial change within the virtual monitoring zone. It is achieved by calculating the average gradient magnitude of pixels in the region. It is a non-negative scalar, and the larger the value, the more rapid the temperature change in the region.

[0036] The third parameter is the migration speed of hotspot patches. This is achieved by identifying connected pixel regions in the current frame image whose temperature values ​​exceed a preset temperature threshold (e.g., regions with a temperature 10 degrees Celsius higher than the global average temperature of the frame). A hotspot patch is a set of consecutive pixels in a single frame of thermal image whose temperature is higher than the surrounding area, determined through connected component analysis. The centroid coordinates of the patch are calculated, and the position change of the centroid is tracked in the next frame image. Bit removal is applied over the time interval between two frames to obtain the instantaneous migration speed of the patch within the current monitoring zone. The migration speed is the rate at which the centroid of the hotspot patch is obtained by bit removal over the time interval between two consecutive frames, and it is used to characterize the dynamic behavior of locally volatile reaction areas.

[0037] Then, the three indicators calculated for each virtual monitoring zone at the current moment are compared with preset benchmark values. The preset benchmark values ​​can be the historical average values ​​or theoretical values ​​of the indicators for each zone, derived from long-term stable production data. The temperature difference between the edge and the center is calculated, namely, the absolute value of the difference between the average temperature of the left edge zone and the average temperature of the central main zone, and the absolute value of the difference between the average temperature of the right edge zone and the average temperature of the central main zone. The larger of the two is taken as the temperature difference value characterizing the edge effect.

[0038] Simultaneously, the dispersion of the thermal field gradient is calculated, that is, the temperature gradient norm of the three regions of the left edge region, the central main region, and the right edge region is calculated, and then the standard deviation of these three values ​​is obtained. This standard deviation is the dispersion of the thermal field gradient, which is used to quantify the non-uniformity of the overall thermal field spatial change.

[0039] Finally, the eleven scalar values ​​obtained from the current calculations—the average temperature of the left edge region, the temperature gradient norm of the left edge region, the migration velocity of hot spots in the left edge region, the average temperature of the central body region, the temperature gradient norm of the central body region, the migration velocity of hot spots in the central body region, the average temperature of the right edge region, the temperature gradient norm of the right edge region, the migration velocity of hot spots in the right edge region, the temperature difference between the edge and the center, and the dispersion of the thermal field gradient—are assembled into a one-dimensional array in a fixed order. This one-dimensional array is the quantitative output characterizing the local intensity differences and spatial consistency of the electrochemical reaction at the current moment, i.e., the reaction inhomogeneity feature vector. Its values ​​comprehensively characterize the spatial inhomogeneity of the electrochemical reaction on the electrode foil surface at the current moment.

[0040] For example, the aforementioned two-dimensional thermal distribution image stream is received. One frame of the image is partitioned, with a total image width corresponding to a physical width of 1200 mm and a pixel column count of 480. The left edge region is defined as columns 1 to 60 of the image (corresponding to 0-150 mm), the central main body region as columns 61 to 420 (corresponding to 150-1050 mm), and the right edge region as columns 421 to 480 (corresponding to 1050-1200 mm). The calculated average pixel temperature in the left edge region of this image frame is 92.1 degrees Celsius, with a temperature gradient norm of 8.5 degrees Celsius per millimeter. A hot spot patch was identified, and its centroid moved 0.3 millimeters from the previous frame to the current frame over a time interval of 0.1 seconds, resulting in a migration speed of 3.0 millimeters per second. The average temperature in the central main body region is 85.3 degrees Celsius, with a temperature gradient norm of 2.1 degrees Celsius per millimeter. No hot spots were identified, so the migration speed is recorded as 0. The average temperature in the right edge region is 90.8 degrees Celsius, with a temperature gradient norm of 7.8 degrees Celsius per millimeter, resulting in a migration speed of 2.5 millimeters per second. The preset baseline average temperature values ​​are assumed to be 88.0 degrees Celsius for the edge region and 86.0 degrees Celsius for the central region. Calculate the temperature difference between the edge and the center: Take the larger of the temperature differences at the left edge (|92.1-85.3|=6.8 degrees Celsius) and the right edge (|90.8-85.3|=5.5 degrees Celsius), which is 6.8 degrees Celsius. Calculate the dispersion of the thermal gradient: The gradient norms for the three zones are 8.5, 2.1, and 7.8, respectively, with a standard deviation of approximately 3.2 degrees Celsius per millimeter.

[0041] Finally, the generated response inhomogeneity feature vector is: [92.1, 8.5, 3.0, 85.3, 2.1, 0, 90.8, 7.8, 2.5, 6.8, 3.2]. The high average temperature, high gradient norm, and non-zero migration velocity in the left and right edge regions of this vector contrast with the low values ​​in the central region. Furthermore, the calculated temperature difference and gradient dispersion are both non-zero, directly demonstrating the successful analysis and quantification of significant response space inhomogeneity features from the two-dimensional thermal distribution image stream.

[0042] In one embodiment of the present invention, the cathode configuration control strategy module is specifically configured to perform the following operations: Based on the deviation between the average temperature value of each partition in the reaction non-uniformity feature vector and the preset benchmark value, the overreaction zone and the underreaction zone are identified; the positive interpole spacing compensation value is calculated for the overreaction zone and the negative interpole spacing compensation value is calculated for the underreaction zone, and the target interpole spacing compensation value is generated; among them, for partitions that are not identified as abnormal, their target interpole spacing compensation value is set to zero.

[0043] Specifically, firstly, the feature vector is input into a pre-set electrochemical-thermal coupled control model. This model is an algorithm program that correlates the current distribution effect and thermal effect of the electrochemical reaction and is used to calculate the control action. It is a mathematical mapping model based on simplified physical equations, and its main function is to convert the thermal feature vector reflecting the non-uniformity of the reaction into executable mechanical adjustment commands. This model is an algorithm module stored in the control computer, which encapsulates simplified physical equations describing the relationship between current density, Joule heat generation rate, and anode-cathode spacing. The core of the model is a nonlinear mapping relationship built based on Ohm's law and the Joule heating principle, used to invert the observed thermal anomalies into geometric adjustment quantities. Here, current density is the current passing through a unit electrode area, a key electrochemical parameter determining the local reaction rate; Joule heat generation rate is the heat generated per unit time when current passes through the electrolyte resistance, which is proportional to the square of the current density and the electrolyte resistivity; and electrode spacing is the perpendicular distance between the surface of the electrode foil and the surface of the opposite cathode section, a key geometric parameter determining the local electric field strength and current distribution. The nonlinear mapping relationship, a functional relationship between current density, Joule heat generation rate, and electrode spacing that cannot be described by a simple linear proportion, is simplified in this model to a linear compensation relationship based on the experimental calibration coefficient k for engineering implementation.

[0044] The electrochemical-thermal coupling control model first identifies abnormal regions based on the average temperature values ​​of each region in the input feature vector. If the average temperature value of a region (such as the left edge region) is consistently higher than its preset reference temperature value and exceeds the set first deviation threshold, and the temperature gradient norm of that region is also high, then that region is marked as a "high-temperature hot spot" or "overreacting" region. Conversely, if the average temperature value of a region is consistently lower than its preset reference temperature value and exceeds the set second deviation threshold, then it is marked as a "low-temperature cold zone" or "underreacting" region. It should be noted that "high-temperature hot spot" or "overreacting" is one judgment state of the feature vector by the model, which is that the average temperature value of the corresponding region is consistently higher than the preset reference, indicating that the local current density in that region is too high and the reaction is too violent. "Low-temperature cold zone" or "underreacting" is another judgment state of the feature vector by the model, which is that the average temperature value of the corresponding region is consistently lower than the preset reference, indicating that the local current density in that region is too low and the reaction activity is insufficient.

[0045] For each zone marked as a "high-temperature hotspot," the model performs inversion calculations. Inversion calculations are the process of deriving the causes and solving for corrective measures from the observation results. The underlying principle is that, under the condition that the power supply output voltage and electrolyte conductivity are approximately constant, the local current density is approximately inversely proportional to the local anode-cathode distance, while the Joule heat generation rate is directly proportional to the square of the current density.

[0046] Model establishment simplifies the relational expression: .in, This represents the increase in the target electrode spacing required to reduce the local current density of this region, in millimeters. This represents the positive deviation between the average temperature value of the zone and the preset reference temperature value, in degrees Celsius. It is the inversion coefficient, and the unit is millimeters per degree Celsius. The values ​​were set based on preliminary process experiment data, achieved by manually adjusting the distance between the cathode sections of specific zones during the experiments. And observe the steady change in the average temperature of the region. The results were obtained by fitting multiple experimental data. On a specific production line, the following steps can be used for offline calibration: Select a representative electrolytic cell and process conditions, and manually adjust the spacing of a certain independent cathode section (such as the edge region) by a small amount. For example, ±0.05mm, ±0.10mm. After the system is running stably, the average temperature change of the corresponding foil area of ​​the cathode section is measured through the dynamic thermal field online acquisition module. Inversion coefficients It can be calculated To improve accuracy, multiple adjustment experiments in different directions can be conducted, and the results obtained can be... The average value. For a specific electrolyte system and voltage, The typical range of values ​​is between 0.015 and 0.025 mm / ℃. Based on this, the model calculates... For example, if the temperature deviation in the left edge region... The temperature was +6.8 degrees Celsius, as calibrated in the experiment. If the value is 0.015 mm per degree Celsius, then the increase in the target interelectrode spacing is calculated. It is 0.102 mm.

[0047] Similarly, for each partition marked as a "low-temperature cold zone," the model uses the same principle but in the opposite direction of calculation. At this point, This represents the negative deviation (absolute value) between the average temperature of this zone and the preset reference temperature. The model calculates the reduction in the target electrode spacing required to increase the local current density of this zone; the calculation formula remains the same. However, the physical meaning of the calculation results is that the spacing needs to be reduced. For example, if the temperature deviation of a certain zone... It is -4.0 degrees Celsius. If the values ​​are the same, the reduction in the target interelectrode spacing is calculated. The target inter-electrode spacing compensation value is 0.060 mm. For partitions not marked as abnormal, the target inter-electrode spacing compensation value is set to zero. After the model traverses all partitions, it outputs a list of values ​​corresponding to each partition, containing positive (increase), negative (decrease), or zero values. This list represents the target inter-electrode spacing compensation value for each cathode segment. Each element in the list corresponds to a cathode segment, and its value indicates the distance that the cathode segment needs to be adjusted relative to its current position. The positive and negative signs indicate the adjustment direction.

[0048] For example, the received response non-uniformity feature vector is [92.1, 8.5, 3.0, 85.3, 2.1, 0, 90.8, 7.8, 2.5, 6.8, 3.2]. The preset reference temperature for the left edge region is 88.0 degrees Celsius, the reference temperature for the central body region is 86.0 degrees Celsius, and the reference temperature for the right edge region is 88.0 degrees Celsius. The model judges: the temperature of the left edge region is 92.1 degrees Celsius, which is higher than the reference temperature of 88.0 degrees Celsius, indicating a deviation. The temperature was +4.1 degrees Celsius, which, although below the first deviation threshold of 5.0 degrees Celsius, was still labeled as "overreacted" by the model due to its high gradient norm of 8.5 and high migration velocity of 3.0. The temperature in the central body region was 85.3 degrees Celsius, lower than the baseline of 86.0 degrees Celsius, indicating a deviation. The temperature is -0.7 degrees Celsius, and the absolute value does not exceed the second deviation threshold, so it is marked as normal. The temperature of the right edge area is 90.8 degrees Celsius, which is higher than the baseline of 88.0 degrees Celsius, indicating a deviation. The temperature was +2.8 degrees Celsius, which did not exceed the first bias threshold, but considering its high gradient norm of 7.8, the model was conservatively labeled as "overreacting." Assuming the inversion coefficients were calibrated through historical experiments... The value is 0.018 mm per degree Celsius. For the left edge region, the increase in target interpolar spacing is calculated. Millimeters. For the right edge region, calculate the increase in target interpolar spacing. Millimeters. For the central main body region, the target electrode spacing compensation value is 0 millimeters. Finally, the target electrode spacing compensation value list output by the model is as follows: left edge region corresponding to cathode segment +0.074 millimeters, central main body region corresponding to cathode segment 0 millimeters, right edge region corresponding to cathode segment +0.050 millimeters.

[0049] In one embodiment of the present invention, the multi-channel drive instruction generation module is specifically configured to perform the following operations: Based on the type of servo actuator, calculate the target rotation angle or linear travel of each drive axis; generate independent pulse control signals for each independent cathode segment, and add a unified motion synchronization timestamp to all signals; integrate the pulse control signals and synchronization timestamps of each channel to generate multi-channel mechanical drive commands.

[0050] Based on each value in the target electrode spacing compensation value list, the required spacing adjustment is the amount of the cathode segment of a specific width partition. The servo actuator is a device that converts electrical control signals into precise mechanical motion. In this step, it specifically refers to the servo motor system or piezoelectric actuator system that drives each independent cathode segment to perform linear motion. It usually has high-precision positioning capability, fast response and programmable control interface.

[0051] Specifically, firstly, based on the target pole spacing compensation value for each partition, and combined with the specific mechanical transmission parameters of the servo actuator driving the cathode segment of that partition, the required physical motion parameters for each independent drive shaft are calculated. Taking a linear drive mechanism using a servo motor and ball screw as an example, the system pre-stores the transmission ratio i of this mechanism, in millimeters per revolution, which represents the distance the cathode segment moves linearly per revolution of the servo motor. Among them, the physical displacement parameter refers to the actual mechanical motion that the drive mechanism needs to complete. For a rotary motor, it is expressed as the target rotation angle θ, and for a linear actuator, it is expressed as the target linear stroke Δd. It has a clear value and unit and serves as a bridge connecting control commands and physical actions. The mechanical transmission ratio is the proportional relationship between the input motion and the output motion in the drive mechanism. In a ball screw mechanism, it is the screw lead, which is a fixed constant determined by the mechanical design. The independent drive shaft is a power shaft that can be independently controlled and drive the motion of a cathode segment. It is mechanically and electrically decoupled from each other and can be independently programmed.

[0052] For each non-zero target interpolar spacing compensation value Calculate the corresponding target rotation angle of the servo motor. The calculation formula is: .in, The target rotation angle is the angle that the servo motor needs to rotate through, and its setting is based on the target pole spacing compensation value. and mechanical transmission ratio Calculations show that The unit is millimeters. Represents the mechanical transmission ratio, which is the linear displacement corresponding to one revolution of the servo motor. It is set based on the specific design parameters of the ball screw lead or gear and rack transmission system used, and is a fixed mechanical characteristic constant.

[0053] This calculation converts linear displacement commands into rotary motion commands. If a piezoelectric actuator or other mechanism that directly generates linear displacement is used, the target linear stroke is calculated, and its value is directly equal to... Next, an independent pulse control signal is generated for each individual cathode segment that requires action. Taking a servo motor as an example, the control signal is typically a series of pulses, with the number of pulses corresponding to the target rotation angle. The pulse frequency corresponds to the motor's rotational speed. Based on a preset motor speed curve, such as uniform acceleration, uniform speed, or uniform deceleration, the timing sequence of pulse occurrences is calculated.

[0054] At the same time, a unified action synchronization timestamp is added to this set of control signals generated for all partitions. It is an absolute time value or a delay time relative to the time of command transmission, used to ensure that multiple independent movements start to execute at the same time. This timestamp indicates the absolute start time of all drive shafts starting to execute their respective movement commands, so as to ensure that the adjustment actions of multiple cathode segments start in time in a coordinated manner, and avoid asynchronous adjustment of the foil surface electric field distribution due to sequential execution.

[0055] Finally, the pulse control signal sequences of all partitions, the corresponding drive axis identifiers, and the synchronization timestamps are compiled. This data is integrated and encapsulated into a data packet or command sequence that follows a specific communication protocol, such as EtherCAT. This data packet or command sequence is the multi-channel mechanical drive command used to reconstruct the cathode's geometric morphology. It is sent to the lower-level multi-axis motion controller via an industrial fieldbus. The information from each channel in the command independently controls the drive mechanism of one cathode segment. It should be noted that the multi-channel mechanical drive command is a structured command set containing information from multiple independent control channels, which is ultimately sent to the motion control system. It is in the form of a data packet, and the data packet contains the motion parameters, timing information, and synchronization commands for each drive axis, used to coordinate the synchronous actions of multiple actuators.

[0056] For example, the target electrode spacing compensation value list is as follows: left edge region +0.074 mm, central main body region 0 mm, right edge region +0.050 mm. Assume the transmission ratio i of the servo motor ball screw mechanism driving each cathode segment is 0.1 mm per revolution. For the left edge region, Calculate the target rotation angle to +0.074 mm. Degree. For the right edge region, For +0.050 mm, calculate Degree. The compensation value for the central main area is 0, so no motion signal needs to be generated.

[0057] Subsequently, pulse control signals are generated for the drive shafts of the left and right edge zones. The motor speed is set to 100 revolutions per minute, or 1.667 revolutions per second. For the left edge zone motor, which needs to rotate 266.4 degrees (approximately 0.74 revolutions), the required time is approximately 0.44 seconds. A corresponding number of pulse sequences are generated based on this. Simultaneously, a motion synchronization timestamp is set. It is 1000 milliseconds after the current system time.

[0058] Ultimately, the integrated multi-channel mechanical drive instructions are a data packet containing the following core information: Channel 1 (left edge) target angle 266.4 degrees, start time. Channel 2 (right edge) target angle 180.0 degrees, start time T_sync; Channel 3 (center) target angle 0 degrees. This instruction is sent to the motion controller via the industrial network, demonstrating the accurate conversion of the target polarity compensation value into a multi-channel control instruction with synchronous timing that can drive the physical mechanical structure.

[0059] In one embodiment of the present invention, the segmented cathode array reconfiguration module is specifically configured to perform the following operations: The servo actuator drives each independent cathode segment to generate independent displacement in a direction perpendicular to the electrode foil surface; by changing the local anode-cathode spacing corresponding to each partition, the electrolyte resistance distribution in the corresponding area is adjusted; under the condition of constant power supply voltage, a spatially adjustable physical electric field is generated based on the adjusted electrolyte resistance distribution.

[0060] Specifically, the servo actuator receives multi-channel mechanical drive commands via an industrial fieldbus network. The servo actuator is a complete motion control system comprising a multi-axis motion controller, servo drivers, servo motors, and mechanical transmission mechanisms. It is capable of receiving digital commands and outputting high-precision, synchronized mechanical motion. The motion controller parses the multi-channel mechanical drive commands, extracting the target motion parameters and a unified motion synchronization timestamp for each channel. When the timestamp indicates the arrival time, the motion controller synchronously sends enable signals and specific motion control signals to the corresponding servo motor drivers or piezoelectric drivers for each zone. Each driver then drives the mechanical actuator connected to its corresponding independent cathode segment to produce motion.

[0061] Taking a lifting mechanism employing a servo motor and a ball screw as an example, the motor starts rotating based on the received pulse sequence, driving the ball screw to rotate via a coupling, converting the rotational motion into linear motion of the screw nut. The screw nut is rigidly connected to the corresponding independent cathode segment, such as the left edge cathode segment, through a rigid connector, thereby driving the cathode segment to generate independent and precise physical displacement in a direction perpendicular to the electrode foil surface. Physical displacement refers to the actual change in the position of the cathode segment in three-dimensional space, specifically linear motion along the direction perpendicular to the electrode foil surface in this step. The direction of displacement is determined by the sign of the target electrode spacing compensation value: when the compensation value is positive, the cathode segment is driven to move away from the electrode foil surface, increasing the local electrode spacing; when the compensation value is negative, the cathode segment is driven to move closer to the electrode foil surface, decreasing the local electrode spacing.

[0062] Each independent cathode segment in each zone completes its displacement adjustment synchronously or quasi-synchronously according to its own instructions. When all cathode segments stop at the new positions determined by the instructions, the geometric and physical shape of the entire cathode array is reorganized, and the local anode-cathode spacing corresponding to each zone is independently adjusted to the new set value. This series of changes in physical spacing directly and instantly reshapes the geometry of the electrolyte channels within the tank. According to the law of resistance... In the electrolyte resistivity and effective conductive cross-sectional area The resistance of the local electrolyte changes negligibly over a short period of time. Mainly related to the path length of the current flow It is proportional to the local anode-cathode distance, which refers to the vertical distance between a point on the electrode foil surface and the surface of the cathode section directly opposite it after adjustment.

[0063] Therefore, by changing the local spacing, the local electrolyte resistance corresponding to each zone is essentially dynamically adjusted. With the power supply output voltage remaining constant, according to the voltage division principle of a series circuit, the current will tend to flow towards the path with lower resistance. By finely adjusting the spacing (i.e., resistance) of each zone, the current distribution ratio flowing through different areas of the electrode foil can be forcibly changed, thereby forming an electric field pattern with preset non-uniform distribution characteristics above the electrode foil surface—that is, a spatially adjustable physical electric field.

[0064] It should be noted that the electrolyte resistance distribution is due to the different anode-cathode spacing in each zone, which leads to different current path lengths in the electrolyte, resulting in a spatial distribution of resistance values ​​across the entire electrode foil width.

[0065] For example, the servo actuator receives a multi-channel mechanical drive command requiring the left edge cathode segment to move upwards by 0.074 mm, the right edge cathode segment to move upwards by 0.050 mm, and the central cathode segment to remain stationary. When the synchronization timestamp arrives, the motion controller drives the corresponding servo motors on the left and right edges. The left edge servo motor rotates 266.4 degrees, precisely lifting its driven cathode segment upwards by 0.074 mm via a ball screw with a lead of 0.1 mm per revolution. The right edge servo motor rotates 180.0 degrees, lifting its driven cathode segment upwards by 0.050 mm. After adjustment, the local electrode spacing in the left edge region changes from the original 2.000 mm to 2.074 mm, the central region remains at 2.000 mm, and the right edge region changes from 2.000 mm to 2.050 mm. This change alters the resistance ratio of the electrolyte channels corresponding to the three zones, with the left edge region showing the largest relative increase in resistance, the central region remaining unchanged, and the right edge region showing a smaller relative increase in resistance. Driven by a constant voltage source, the current proportion at the left edge decreases, the current proportion at the right edge decreases slightly, and the current proportion at the center increases relatively. Thus, a spatially adjustable physical electric field with a strong center and weak sides is formed on the electrode foil surface, demonstrating the process of actively and precisely generating a spatially adjustable electric field to correct reaction inhomogeneities through mechanical configuration reorganization.

[0066] In one embodiment of the present invention, the feedback locking module is specifically configured to perform the following operations: Continuously acquire new two-dimensional thermal distribution image streams and calculate their thermal distribution gradient norm in real time; determine whether the gradient norm continues to decrease and falls below the preset uniformity judgment threshold; if so, send a locking command to the servo actuator to lock and maintain the current cathode configuration and electric field distribution.

[0067] Specifically, after the generated spatially adjustable physical electric field is applied to the electrode foil, the electric field modifies the local current density flowing through different width regions of the electrode foil by changing the local electrolyte resistance distribution. The local current density is the current on the electrode per unit area, and its correction is accomplished by adjusting the electric field by changing the local electrode spacing.

[0068] For the high-temperature hot spot zone identified as overreacting, since its corresponding cathode section has been moved further away, the increased local electrode spacing leads to increased resistance. With the total voltage remaining constant, the voltage drop obtained increases, which reduces the actual effective overpotential of the region. This inhibits the electrochemical reaction rate in the region, reduces heat generation, and the temperature begins to drop.

[0069] For the low-temperature cold zone identified as having insufficient reaction, its cathode section was brought closer, reducing local resistance and increasing the effective reaction overpotential, thereby activating the reaction activity in that region, increasing heat generation, and causing the temperature to rise. This adjusted thermal field change was continuously monitored using an infrared thermal imaging acquisition unit, acquiring new two-dimensional heat distribution image streams in real time, and calculating its key indicators, particularly the norm of the thermal field gradient across the entire foil surface. The gradient norm of the heat distribution image stream is a scalar indicator calculated from the real-time thermal field image, characterizing the overall intensity of temperature spatial changes. Its decline indicates that the temperature distribution tends to be flatter and more uniform; this gradient norm comprehensively reflects the severity of temperature spatial changes.

[0070] The real-time value of the gradient norm is compared with a preset uniformity threshold. This preset uniformity threshold is a gradient norm threshold used to determine whether the reaction has reached sufficient uniformity. It is set based on the statistical upper limit of the thermal field gradient norm during the stable phase of historical qualified product production. This threshold is determined statistically based on typical thermal field uniformity data from the production of qualified electrode foil. For example, under normal process conditions of stable production and qualified products, a two-dimensional thermal distribution image stream is acquired over a continuous production cycle; the global thermal distribution gradient norm Gr corresponding to each frame is calculated; statistical analysis is performed on the obtained Gr sequence, such as calculating the mean μ and standard deviation σ, and the threshold is set accordingly. This covers over 95% of qualified operating conditions, or a conservative upper limit can be set based on experience. According to statistics from multiple production lines, for specific types of electrode foil etching processes, The example range is 0.8-1.2℃ / mm; the present invention uses 1.0℃ / mm as an example.

[0071] When the gradient norm of the thermal distribution image stream is continuously reduced and eventually falls below the preset uniformity threshold, it is determined that the electrochemical reaction has reached a self-healing equilibrium state, and the current cathode array geometry effectively balances the reaction.

[0072] At this point, a lock command is sent to the motion controller to disable all servo drives or switch to position holding mode, locking the current positions of each cathode segment and maintaining the existing spatially adjustable physical electric field distribution. After adjustment, on the macroscopic scale of the electrode foil, the differences in effective electric field, current density, and reaction driving force experienced by each point are minimal. That is, in this macroscopically homogeneous electrochemical environment, the local current density differences at each point across the entire width of the electrode foil are controlled within a minimal range. This ensures that during the etching process, the initiation, growth, and branching processes of tunnel pits on the aluminum foil surface tend to be synchronized; and during the formation process, the nucleation, thickening, and densification processes of the oxide film tend to be synchronized. Finally, after a complete process cycle, the depth and distribution density of the tunnel structure or the thickness and crystal structure of the oxide film are optimized throughout the width and length of the product, resulting in a finished electrode foil with optimized microstructure consistency; it is an electrode foil product with improved spatial uniformity of surface micromorphology and composition.

[0073] It should be noted that guiding the electrochemical reaction to self-healing equilibrium is a process of automatically converging and stabilizing the non-uniform reaction state towards a uniform state through closed-loop regulation established in the preceding steps; tunnel hole growth is the process of forming micro-pore structures by anodic dissolution of aluminum foil during the corrosion process; and oxide film formation is the process of generating an insulating alumina dielectric layer on the surface of aluminum foil during the formation process.

[0074] For example, after the spatially adjustable physical electric field is formed, the thermal field is continuously monitored. Due to the increased spacing, the average temperature in the left edge region gradually decreases from 92.1 degrees Celsius, while the temperature in the right edge region also begins to decrease from 90.8 degrees Celsius. The temperature in the central region slightly increases from 85.3 degrees Celsius. The gradient norm of each frame of the thermal image is calculated in real time. Before adjustment, the gradient norm is 3.2 degrees Celsius per millimeter. After adjustment, this value begins to decrease. After approximately 30 seconds of continuous monitoring, the gradient norm decreases and stabilizes at 0.8 degrees Celsius per millimeter. The preset uniformity threshold is set to 1.0 degrees Celsius per millimeter. Since 0.8 is less than 1.0, it is determined that self-healing equilibrium has been reached, and the positions of all cathode segments are then locked. The remaining processing time was completed under this stable electric field environment. The resulting electrode foil, after offline sampling and analysis, showed film thicknesses of 105.2 nm, 103.8 nm, and 104.5 nm at the left, center, and right positions, respectively, with a range of only 1.4 nm. In contrast, the uncontrolled control sample showed a film thickness range of 8.7 nm. This demonstrates the process of successfully generating a finished electrode foil with optimized microstructure consistency by guiding the electrochemical reaction to self-heal equilibrium.

[0075] It should be noted that the formulas described above, through the principle of dimensional consistency and mathematical standardization methods (such as normalization, dimensionless parameter conversion, or unit system unification), can translate physical quantities with different properties into unitless standard values ​​or superimposed parameters of the same dimension. This eliminates the interference of different dimensions on the computational logic, allowing the formulas to retain the original data distribution characteristics while possessing mathematical rationality and adaptability to objective laws. The descriptions are merely exemplary embodiments of the present invention and should not be construed as limiting the scope of the invention.

[0076] Each of the modules can be implemented in whole or in part through software, hardware, or a combination thereof. It supports hardware embedded in or independent of the processor in the computer device, and also supports software stored in the memory of the computer device, so that the processor can call and execute the operations corresponding to each of the above modules.

[0077] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included within the protection scope of the present invention.

Claims

1. A monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil, characterized in that, include: The dynamic thermal field online acquisition module is used to acquire dynamic thermal field data of the moving electrode foil surface and generate a two-dimensional thermal distribution image stream. The reaction non-uniformity feature extraction module is used to analyze the two-dimensional heat distribution image stream, extract indicators that characterize the local differences in electrochemical reaction, and generate a reaction non-uniformity feature vector. The cathode configuration control strategy module, based on the reaction non-uniformity feature vector, performs inversion calculations through an electrochemical-thermal coupling control model and outputs the target electrode spacing compensation value for each independent cathode segment in the segmented cathode array. A multi-channel drive command generation module is used to generate multi-channel mechanical drive commands based on the target pole spacing compensation value; The segmented cathode array reconfiguration module is used to respond to multi-channel mechanical drive commands, drive each independent cathode segment to generate physical displacement, reconfigure the geometric configuration of the segmented cathode array, and construct a spatially adjustable physical electric field. The feedback locking module corrects the local current density flowing through different regions of the electrode foil based on the spatially adjustable physical electric field, continuously monitors the changes in the thermal field, and locks the current cathode array configuration when the thermal distribution gradient norm falls back to the preset uniformity judgment threshold.

2. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 1, characterized in that, The dynamic thermal field online acquisition module is specifically configured to perform the following operations: The infrared thermal imaging acquisition unit is used to capture the infrared radiation signal emitted by the surface of the electrode foil immersed in the electrolyte in real time. The infrared radiation signal is mapped into a two-dimensional temperature matrix arranged in time order to form the raw data of the dynamic thermal field. The original dynamic thermal field data is subjected to joint time-domain and spatial-domain filtering and geometric correction to generate a two-dimensional thermal distribution image stream.

3. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 1, characterized in that, The specific configuration of the reaction unevenness feature extraction module is used to perform the following operations: Each frame of two-dimensional thermal distribution image is divided into multiple virtual monitoring zones along the width direction of the electrode foil; Calculate the average temperature value, temperature gradient norm, and migration rate of hotspot patches within each virtual monitoring zone; Based on the average temperature value, temperature gradient norm, and migration velocity, combined with the temperature difference between the edge and the center and the thermal field gradient dispersion, a response non-uniformity feature vector is generated.

4. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 1, characterized in that, The cathode configuration control strategy module is specifically configured to perform the following operations: Based on the deviation between the average temperature value of each zone in the reaction non-uniformity feature vector and the preset benchmark value, the over-reaction zone and the under-reaction zone are identified. Calculate the positive inter-electrode spacing compensation value for the overreaction zone and the negative inter-electrode spacing compensation value for the underreaction zone to generate the target inter-electrode spacing compensation value; For partitions that are not identified as abnormal, their target interpole spacing compensation value is set to zero.

5. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 1, characterized in that, The multi-channel driver instruction generation module is specifically configured to perform the following operations: Calculate the target rotation angle or linear travel of each drive axis based on the type of servo actuator; Generate independent pulse control signals for each independent cathode segment, and add a unified action synchronization timestamp to all signals; Integrate the pulse control signals from each channel with the synchronization timestamp to generate multi-channel mechanical drive commands.

6. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 1, characterized in that, The segmented cathode array reconfiguration module is specifically configured to perform the following operations: Each independent cathode segment is driven to produce independent displacement in a direction perpendicular to the electrode foil surface by a servo actuator; The electrolyte resistance distribution in the corresponding area is adjusted by changing the local anode-cathode spacing of each partition. Under constant power supply voltage, a spatially adjustable physical electric field is generated based on the adjusted electrolyte resistance distribution.

7. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 1, characterized in that, The feedback locking module is specifically configured to perform the following operations: Continuously acquire new two-dimensional thermal distribution image streams and calculate their thermal distribution gradient norm in real time; Determine whether the gradient norm continues to decrease and falls below a preset uniformity threshold; If so, a locking command is sent to the servo actuator to lock and maintain the current cathode configuration and electric field distribution.

8. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 2, characterized in that, Preprocessing is performed on the raw dynamic thermal field data, including: A Gaussian low-pass filter is used to smooth the data to suppress random temperature fluctuations caused by environmental radiation fluctuations or detector noise. Based on the optical distortion parameters of the infrared thermal imaging acquisition unit lens and its mounting pitch angle relative to the electrode foil plane, the temperature distribution image of each frame is corrected using an image transformation algorithm.

9. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 3, characterized in that, The virtual monitoring zone includes at least a left edge zone, a central main zone, and a right edge zone. The width of each zone is set based on the total width of the electrode foil and engineering experience.

10. The monitoring and control system for the preparation process of aluminum electrolytic capacitor electrode foil according to claim 3, characterized in that, The migration speed of hotspot patches is calculated by identifying the centroid displacement of high-temperature connected regions in consecutive frames and combining it with time intervals.

Images