A method and system for on-line monitoring of an electroplating solution for composite electroplating

Abstract

This invention relates to the field of computer technology, specifically to an online monitoring method and system for electroplating solutions used in composite electroplating. The method includes: acquiring the original time-domain electrical signal of the target ion concentration during the composite electroplating process; identifying and removing first noise generated by collisions of suspended particles and second noise generated by environmental interference based on the waveform characteristics of the original electrical signal to obtain a denoised signal; adaptively compensating for the periodic baseline drift caused by the physical barrier effect of suspended particles in the denoised signal to obtain an actual ion concentration signal; performing trend prediction based on the actual ion concentration signal, and issuing warnings and / or generating control commands based on the prediction results; and accurately identifying and separating interference signals caused by Brownian motion and collisions of suspended particles, thereby improving the purity of the ion concentration signal.

Description

Technical Field

[0001] This invention relates to the field of online monitoring technology for electroplating processes, and particularly to an online monitoring method and system for electroplating solutions used in composite electroplating. Background Technology

[0002] Composite electroplating is a process in which solid suspended particles (such as SiC, Al2O3, nanodiamond, etc.) are added to the electroplating solution for co-deposition with metal ions. Because the composite electroplating solution is a multiphase complex colloidal (solid)-electrolyte (liquid) system, in which metal ions (such as Ni...) are added... 2+ Cu 2+ The stability of online monitoring of the concentrations of (etc.) and suspended particles directly affects the deposition rate of the base metal and the coating quality.

[0003] The ideal state for composite electroplating is that the rate of metal ion reduction and crystallization at the cathode (i.e., the "growth" rate) and the rate at which suspended particles migrate to the cathode surface and are captured (i.e., the "embedding" rate) reach a "solid-liquid ratio" balance. If the monitoring system is interfered with and generates false alarms, for example, if the optical sensor is interfered with by bubbles and falsely reports "particle concentration too low," the system will automatically add a large amount of particles, leading to particle oversaturation and agglomeration, resulting in large particle clumps inside the coating and loss of mechanical strength. If the electrode is interfered with by particles and falsely reports "insufficient metal ion concentration," the system may increase the current density or add metal salts, which will lead to excessive metal reduction polarization ("growth" rate too fast), preventing particles from being embedded into the coating in time, ultimately resulting in a pure metal coating or a defective product with extremely low particle content. Therefore, accurate monitoring of the metal ion concentration in the composite electroplating solution is crucial to ensuring the quality of composite electroplating.

[0004] In the prior art, patent application CN112410864A (applicant: Anhui University of Science and Technology; publication date: 2021-02-26) discloses a related technology, which mainly focuses on monitoring and controlling macroscopic environmental parameters such as temperature and pH value. Its data processing algorithm (such as HMM) aims to handle the nonlinearity and hysteresis problems of sensor signals, but does not address the problem of identifying and removing specific nonlinear and pulse noise generated by the physical movement (such as Brownian motion and collision) of suspended solid particles in composite electroplating solutions on the electrical signal of ion concentration.

[0005] Patent application CN121113922A (Applicant: Zhuzhou Xusen Technology Co., Ltd.; Publication Date: 2025-12-12) discloses an online monitoring process for the composition of lead frame electroplating solution. Its focus is on the stability of spectral signals and the accuracy of models. It does not address the problem of electrical signal noise caused by the physical interference of suspended particles in composite electroplating solution (such as the formation of physical barriers and the generation of collision noise) when the electrochemical sensor measures ion concentration.

[0006] During the online detection of metal ion concentration in composite electroplating solutions, the Brownian motion of suspended solid particles in the complex solid-liquid suspension system not only hinders the free diffusion of ions, but also the movement of suspended particles and the surfactants on the surface of the particles form a physical barrier near the electrode. Ultimately, this results in noise interference in the target ion concentration signal measured by the sensor, which in turn affects the electroplating effect.

[0007] Existing noise reduction methods typically smooth high-frequency noise using low-pass filtering or moving averages. However, noise generated by the motion of suspended particles is often nonlinear and exhibits both instantaneous and periodic pulse noise. Current methods struggle to accurately identify and eliminate this non-stationary noise caused by particle motion without compromising the target ion concentration signal itself. Therefore, there is an urgent need for an online monitoring method that can effectively separate and compensate for signal interference caused by the physical motion of suspended particles, tailored to the characteristics of composite electroplating solution solid-liquid suspension systems. Summary of the Invention

[0008] To address the technical problem that existing noise reduction methods cannot accurately identify and eliminate nonlinear and non-stationary noise caused by the physical movement of suspended particles in composite electroplating solutions without damaging the target ion concentration electrical signal itself, an online monitoring method and system for composite electroplating solutions is provided.

[0009] In a first aspect, embodiments of the present invention provide an online monitoring method for electroplating solutions used in composite electroplating, comprising: Obtain the original time-domain electrical signal of the target ion concentration during the composite electroplating process; Based on the waveform characteristics of the original electrical signal, the first noise generated by the collision of suspended particles and the second noise generated by environmental interference are identified and removed to obtain the denoised signal. Based on the periodic baseline drift caused by the physical barrier effect of suspended particles in the denoised signal, adaptive compensation is performed on the periodic baseline drift to obtain the actual ion concentration signal; Trend prediction is performed based on the actual ion concentration signal, and warnings and / or control commands are issued based on the prediction results.

[0010] Preferably, the method for identifying and stripping the first noise and the second noise includes: The original electrical signal is decomposed into multiple pulse segments according to the peak and valley values ​​of the waveform; Calculate the asymmetry coefficient and distribution coefficient for each pulse segment; The feature significance of each pulse segment is determined based on the asymmetry coefficient and the distribution coefficient. Based on the comparison results between the feature saliency and the preset threshold, the multiple pulse segments are classified into collision noise segments, white background noise segments, and normal signal segments; Different smoothing methods were used to peel off the collision noise segment and the white background noise segment respectively.

[0011] Preferably, the preset threshold includes a first threshold and a second threshold, wherein the first threshold is greater than the second threshold; classifying the plurality of pulse segments into collision noise segments, white background noise segments, and normal signal segments based on the comparison result of the feature salience and the preset threshold includes: When the significance of the feature is greater than or equal to the first threshold, the pulse segment is determined to be a collision noise segment; When the significance of the feature is less than or equal to the second threshold, the pulse segment is determined to be a white background noise segment; When the saliency of the feature is between the second threshold and the first threshold, the pulse segment is determined to be a normal signal segment.

[0012] Preferably, different smoothing methods are used to peel off the collision noise segment and the white background noise segment, including: For the collision noise segment, a characteristic time sliding window of a preset width is established. The amplitude of the collision noise pulse falling into the characteristic time sliding window is integrated with energy to convert the discrete pulse energy into a continuous analog quantity curve, which serves as the smoothed collision energy index. For the white background noise segment, polynomial smoothing is performed using SG filtering.

[0013] Preferably, the adaptive compensation for periodic baseline drift caused by the physical barrier effect of suspended particles includes: Obtain the cycle start point of periodic baseline drift; The denoised signal is compensated based on the real-time attenuation coefficient within the current cycle. The attenuation coefficient is used to characterize the degree of attenuation of the ion concentration signal by the physical barrier of suspended particles.

[0014] Preferably, the method for obtaining the cycle start point of the periodic baseline drift includes: Acquire the trough moments of the periodic baseline, as well as the moments when the stirring blades are directly facing the sensor probe; The proximity weight is determined based on the liquid flow rate and suspended particle concentration at the sensor location. The cycle start point is determined based on the trough time, the time when the stirring blade is directly facing the sensor probe, and the proximity weight.

[0015] Preferably, the step of predicting trends based on the actual ion concentration signal and issuing early warnings and / or generating control commands based on the prediction results includes: Extract the first and second derivatives of the actual ion concentration signal; Based on the first and second derivatives, a nonlinear Taylor expansion extrapolation model is used to predict the remaining time for the actual ion concentration signal to fall below a preset threshold. When the remaining time is lower than a preset time threshold, an early warning is issued and / or control commands are generated for the linkage actuator to intervene and supplement.

[0016] An online monitoring system for electroplating solutions used in composite electroplating includes: The signal acquisition unit is used to acquire the time-domain raw electrical signal of the target ion concentration during the composite electroplating process; The denoising unit identifies and removes the first noise generated by the collision of suspended particles and the second noise generated by environmental interference based on the waveform characteristics of the original electrical signal, and obtains the denoised signal. The compensation unit adaptively compensates for the periodic baseline drift caused by the physical barrier effect of suspended particles in the denoised signal to obtain the actual ion concentration signal. The prediction unit performs trend prediction based on the actual ion concentration signal, and issues early warnings and / or generates control commands based on the prediction results.

[0017] Thirdly, an electronic device includes: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the online monitoring method for electroplating solutions used in composite electroplating.

[0018] Fourthly, a computer-readable medium storing a computer program that, when executed by a processor, implements the aforementioned method for online monitoring of electroplating solutions for composite electroplating.

[0019] The beneficial effects of this invention are as follows: The motion noise stripping method based on waveform characteristics solves the problem that traditional linear filtering methods cannot handle nonlinear and non-stationary impulse noise without damaging the target ion concentration electrical signal itself. It can accurately identify and separate interference signals caused by Brownian motion and collision of suspended particles, thereby improving the purity of the ion concentration signal.

[0020] The state characteristics of suspended particles under different dimensions were analyzed, and the signal baseline was adaptively compensated based on the adaptive baseline of multidimensional state characteristics. Correction was performed for the periodic baseline drift caused by ion concentration, effectively eliminating the continuous influence of physical barrier effects on sensor measurements and improving the monitoring stability and accuracy of the signal in complex solid-liquid systems.

[0021] Finally, through linkage early warning and adaptive control technology, control commands are automatically generated to link the actuators for external intervention. This fundamentally avoids false alarms of "insufficient metal ion concentration" caused by noise misjudgment, prevents the system from issuing incorrect supplementation commands, and thus ensures the balance of the "solid-liquid ratio" and the coating quality during the composite electroplating process. Attached Figure Description

[0022] Figure 1 A schematic flowchart of an online monitoring method for composite electroplating solution provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the original electrical signal of metal ion concentration in the electroplating solution in the time domain, collected according to an embodiment of the present invention; Figure 3 This is a schematic diagram of the original electrical signal after noise stripping in an embodiment of the present invention; Figure 4 This is a structural block diagram of an online monitoring system for composite electroplating solution in an embodiment of the present invention; Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0023] To enable those skilled in the art to better understand the technical solutions of the present invention, exemplary embodiments of the present invention are described below in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to aid understanding. These should be considered merely exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0024] Where there is no conflict, the various embodiments of the present invention and the features thereof may be combined with each other.

[0025] As used herein, the term “and / or” includes any and all combinations of one or more related enumerated entries.

[0026] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used herein, the singular forms “a” and “the” are intended to include the plural forms as well, unless the context clearly indicates otherwise. It will also be understood that when the terms “comprising” and / or “made of” are used in this specification, the presence of the stated feature, integral, step, operation, element, and / or component is specified, but the presence or addition of one or more other features, integrals, steps, operations, elements, components, and / or groups thereof is not excluded. Terms such as “connected” or “linked” are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0027] Unless otherwise specified, all terms used herein (including technical and scientific terms) have the same meaning as commonly understood by one of ordinary skill in the art. It will also be understood that terms such as those defined in commonly used dictionaries should be interpreted as having the meaning consistent with their meaning in the context of the relevant art and the invention, and will not be interpreted as having an idealized or overly formal meaning unless expressly so defined herein.

[0028] In the technical solution of this invention, the collection, storage, use, processing, transmission, provision, and disclosure of user personal information all comply with relevant laws and regulations and do not violate public order and good morals. The use of user data in this technical solution follows relevant national laws and regulations (e.g., the "Information Security Technology - Personal Information Security Specification"). For example: appropriate measures are taken for personal information access control; restrictions are imposed on the display of personal information; the purpose of using personal information does not exceed the scope of direct or reasonable association; and explicit identity targeting is eliminated when using personal information to avoid precisely locating a specific individual.

[0029] In this embodiment, the image data involved was acquired from a real electronic waste physical dismantling production line and the publicly available PCB_FULL dataset, totaling over 5000 high-resolution RGB images. The experimental platform and computing equipment involved in this invention have the following specific hardware and software configurations: the operating system is Windows 11; the central processing unit (CPU) is an Intel(R) CORE(TM) i5-14600KF with 32GB of RAM; the graphics processing unit (GPU) is an NVIDIA GeForce RTX 5060 Ti; the software algorithm architecture is based on Python 3.10.19 compiled language and the PyTorch 2.9.0.dev20250909+cu128 deep learning framework, with a parallel computing architecture of CUDA Version: 12.8. Furthermore, Automatic Mixed Precision (AMP) technology was enabled during model training to effectively reduce GPU memory usage and accelerate gradient calculation while maintaining computational accuracy. All other components and image acquisition hardware, unless otherwise specified, conform to relevant commercially available standards in the industrial vision field. Example 1

[0030] like Figure 1 As shown, this embodiment provides an online monitoring method for electroplating solutions used in composite electroplating, including: S1. Obtain the original time-domain electrical signal of the target ion concentration during the composite electroplating process; During electrical signal acquisition, the target ion concentration electrical signal in the time domain of the composite electroplating solution is obtained through an electrochemical sensor. Specifically, an ion-selective electrode (ISE) or a three-electrode system is used for data acquisition. The sensor is installed at the reflux port of the circulating flow in the electroplating tank. The sensor housing is designed with a hydrodynamically optimized cover to ensure smooth liquid flow. At the same time, a sampling frequency of at least 8 kHz is set to acquire the raw electrical signal.

[0031] In the composite electroplating process, the original electrical signal of the metal ion concentration in the electroplating solution in the time domain is interfered with by the Brownian motion of suspended particles due to the complex solid-liquid suspension system in the electroplating solution. That is, the waveform of the original electrical signal is the result of the superposition of collision noise, interference caused by barrier offset, and the actual ion concentration.

[0032] S2. Based on the waveform characteristics of the original electrical signal, identify and remove the first noise generated by the collision of suspended particles and the second noise generated by environmental interference to obtain the denoised signal; In one embodiment, noise identification and removal are performed by analyzing the waveform characteristics corresponding to white background noise and particle collision noise present in the target ion concentration signal; specifically including: (1) Pulse segment decomposition: Based on the fluctuation peak and valley values ​​of the original concentration signal, the original electrical signal is decomposed into multiple different pulse segments.

[0033] (2) Calculate the asymmetry coefficient: For any pulse segment, calculate its asymmetry coefficient according to the following formula. : ; In the formula, For any pulse segment, the asymmetry coefficient is denoted as . and These are the end and start times of this pulse segment, respectively. This corresponds to the peak point of the pulse segment at that time. and These represent the area values ​​of the left and right regions, respectively, with the peak point as the axis of symmetry. This represents the fall time of this pulse segment. This represents the rise time of this pulse segment. The larger the pulse, the more pronounced the rapid rise and slow fall characteristics of this pulse segment, which is also related to the difference in its left and right areas. Right now The smaller the value, the more obvious the asymmetry of the waveform in this pulse segment, and vice versa.

[0034] (3) For a certain pulse segment The relative peak value is calculated according to the following formula. : ; In the formula, pulse segment The relative peak value pulse segment peak value It represents the average peak value of all pulse segments in the original signal.

[0035] Then calculate the distribution coefficient according to the following formula. : ; In the formula, For this pulse segment The distribution coefficient, and pulse segments The relative peak value between the preceding and following pulse segments. The larger the relative peak value of the collision noise, and the greater the difference between its relative peak value and that of the adjacent pulse segments due to its discreteness, the higher the corresponding distribution coefficient. The larger the value, the higher the value; due to its density, the smaller the relative peak difference between adjacent pulse segments, the higher the corresponding distribution coefficient. The smaller it is.

[0036] (4) For any pulse segment The significance of its features is calculated according to the following formula. : ; In the formula, The characteristic significance of pulse segment x, This is the distribution coefficient of the pulse segment. For this pulse segment asymmetry coefficient, pulse segment The relative peak value. Then, if the pulse segment... The more a pulse segment matches the characteristics of collision noise, the greater its feature significance. Conversely, as the feature significance decreases, it indicates that the pulse segment is more likely to be white background noise.

[0037] (5) Threshold classification: Calculate the feature significance of all decomposed pulse segments, normalize the results to the range [0,1], and set a double threshold: when When ≥ 0.65, the corresponding pulse segment is considered as the collision noise segment; when ≥ 0.35 When the value is greater than 0.65, the corresponding pulse segment is considered as a white background noise segment; the rest are normal baseline signal segments.

[0038] (6) Layered smooth peeling: For the collision noise segment, a characteristic time sliding window of preset width is established. The instantaneous amplitude of the effective particle collision pulse falling within the characteristic time sliding window is squared to calculate the cumulative energy integral or energy variance of the discrete pulse within the window. As the characteristic time sliding window moves continuously in the time domain, the discrete pulse energy integral is transformed into a continuous analog curve, and the output is a smooth solid-phase particle collision energy index. This index is positively correlated with the particle suspension concentration and agglomeration degree in the composite electroplating solution.

[0039] For the white background noise segment, SG filtering (Savitzky-Golay filtering) is used for polynomial smoothing, thereby filtering out the white background noise while preserving the baseline characteristics; S3. Based on the periodic baseline drift caused by the physical barrier effect of suspended particles in the denoised signal, adaptive compensation is performed on the periodic baseline drift to obtain the actual ion concentration signal; In one embodiment, after stripping away high-frequency particulate noise, what remains is a low-frequency baseline that appears smooth but exhibits periodic distortion. The drop caused by the physical barrier formed near the electrode due to the movement of suspended particles and the presence of surfactants on the particle surface is "periodic." This is because in industrial electroplating baths, the rotation of mechanical agitation and the pulses of the circulating pump create a dynamic cycle of "accumulation-washing-re-accumulation" of particles and surfactants on the electrode surface. This dynamic cycle curve does not represent the actual concentration of relevant ions in the electroplating solution, thus requiring adaptive compensation for this periodic curve.

[0040] Specifically, this step includes the following sub-steps: (1) Determine the starting point of the loop: For a periodic baseline, obtain all valley moments. The moment when the nearest stirring blade is directly aligned with the sensor probe For any sampling time between the two periods, the proximity weight is calculated according to the following formula. : ; In the formula: To be closer to the weight, The liquid flow velocity at the sensor location. This represents the concentration of suspended particles at the sensor location.

[0041] Then calculate the starting time according to the following formula. : ; In the formula, The starting time is [time]. This represents the physical mask removal rate near the sensor. A higher value indicates a stronger removal rate of particles by agitation, and a faster influx of ions into the electrode, meaning the cycle starts closer to the start time. Conversely, the closer to .by We weight the two time points using a weighted average (normalized to the range [0,1]) to determine a more realistic starting time for the cycle. We then select... The time point serves as the starting point of the current cycle and also the ending point of the previous cycle.

[0042] (2) Calculate the real-time attenuation coefficient: For any given period, the accuracy of ion concentration monitoring changes as the barrier of suspended particles gradually alters. Compensation for the concentration signal requires a comprehensive analysis based on both the electroplating solution itself and the state of the particles.

[0043] The signal attenuation coefficient at time t within a certain period can be calculated using the following formula. : ; In the formula, The signal attenuation coefficient at time t within a certain period; The peak factor at time t in the particle concentration signal; and and They are respectively The concentration of surfactant in the electroplating solution and the kinematic viscosity of the electroplating solution at any given time; This represents the liquid phase promotion degree at the corresponding moment. The larger the value, the more obvious the promotion effect of the electroplating solution on the barrier, and the deeper the barrier. Then, it is combined with the peak factor to obtain the real-time attenuation coefficient.

[0044] (3) Execution of compensation: The real-time attenuation coefficient is calculated for each cycle. The attenuation coefficient represents the barrier effect of the particles on the electrode; a larger attenuation coefficient corresponds to a deeper barrier, requiring a greater compensation force at that moment. Compensation is performed based on the normalized attenuation coefficient at any moment within the cycle. ; In the formula, Within a certain period The actual ion concentration signal value at time [time]. Before compensation The ion concentration signal value at time [time]. for The signal attenuation coefficient at a given time; that is, using the real-time signal attenuation coefficient as the compensation weight, the real-time ion concentration signal after stripping is compensated and adjusted, effectively eliminating the periodic influence of the physical barrier effect on the sensor measurement, and improving the accuracy of target metal ion concentration monitoring.

[0045] S4. Perform trend prediction based on the actual ion concentration signal, and issue early warnings and / or generate control commands based on the prediction results.

[0046] Specifically, for the optimized ion concentration signal monitored and reconstructed in real time, the system continuously extracts its first derivative (instantaneous consumption rate) and second derivative (consumption acceleration). Using a nonlinear Taylor expansion extrapolation model, it calculates in real time the remaining absolute safety time before the ion concentration falls below the process red line under the current dynamic consumption trend. When an anomaly occurs or the remaining absolute safety time falls below a preset threshold, the control system issues targeted intervention commands to the underlying actuators, including physical field modulation and precise quantitative chemical replenishment, to maintain the concentration balance of relevant metal ions in the electroplating solution and ensure the quality of composite electroplating.

[0047] Example 2

[0048] This embodiment provides an online monitoring method for electroplating solutions used in composite electroplating, the method comprising the following steps: Step S1: Obtain the electrical signal of the target ion concentration in the composite electroplating solution in the time domain during the electroplating process using an electrochemical sensor.

[0049] In this embodiment, the composite electroplating process involves adding solid suspended particles (such as SiC, Al2O3, nanodiamond, etc.) to the electroplating solution for co-deposition with metal ions. Since the composite electroplating solution is a multiphase complex colloidal (solid)-electrolyte (liquid) system, metal ions (such as Ni...)... 2+ Cu 2+ The stability of online monitoring of the concentrations of (etc.) and suspended particles directly affects the deposition rate of the matrix metal.

[0050] The ideal state of composite electroplating is that the rate at which metal ions are reduced and crystallized at the cathode (the "growth" rate) and the rate at which suspended particles move to the cathode surface and are captured (the "embedding" rate) reach a "solid-liquid ratio" balance.

[0051] This embodiment focuses on the accuracy of online detection of metal ion concentration in electroplating solutions. First, a contamination-resistant ion-selective electrode (ISE) or a three-electrode system is used for data acquisition. The sensor is installed at the return port of the circulating flow within the electroplating tank. The sensor housing is designed with a hydrodynamically optimized cover to ensure smooth liquid flow, and a frequency of at least 8 kHz is set to acquire the raw electrical signal.

[0052] In the composite electroplating process, the original electrical signal of the metal ion concentration in the electroplating solution in the time domain is as follows: Figure 3 As shown, due to the complex solid-liquid suspension system in the electroplating solution, the Brownian motion of suspended particles directly interferes with the ion concentration electrical signal. That is, the waveform of the original ion concentration electrical signal is the result of the combined superposition of collision noise, barrier shift interference, and the actual ion concentration. Traditional methods for smoothing high-frequency noise generally cannot accurately extract the actual ion concentration value without damage. Therefore, this scheme, based on the two interference effects of suspended particles on metal ions, improves the accuracy of online monitoring of metal ion concentration in the time domain by first stripping away collision noise and then compensating for barrier shift.

[0053] Step S2: Identify and remove the white background noise and particle collision noise present in the target ion concentration signal by analyzing the waveform features.

[0054] In online monitoring of composite electroplating, electrochemical sensors rely on the "electric double layer" structure on the electrode surface to establish a stable potential, thereby inferring ion concentration using the Nernst equation. However, when solid-phase suspended particles are introduced, these particles undergo Brownian motion in the electroplating solution or violently tumble due to mechanical stirring, constantly impacting the sensing interface of the electrochemical probe. The frequent entry and exit of particles cause severe oscillations in the local dielectric environment of the electrode surface, resulting in strong noise in the output electrical signal.

[0055] Therefore, this step requires stripping and smoothing the nonlinear high-frequency noise caused by the frequent collisions and sliding of suspended particles. It's important to note that not all nonlinear high-frequency noise in the original signal is collision noise; there is also white background noise generated by electromagnetic interference from the electronic components themselves and the environment. Since this noise is not generated by particle collisions, it requires a different smoothing method than the collision noise and therefore needs to be stripped away as well.

[0056] Here, we perform a stripping process based on the waveform characteristics of different noises. First, we decompose the original concentration signal into multiple pulse segments based on the peak and trough values ​​of the fluctuations. The waveform characteristics of different noises can then be quantified using the pulse segment characteristics. The waveform of white background noise exhibits a normal distribution, and is continuous, dense, and its amplitude fluctuates uniformly within a fixed variance range; while the waveform of particle collisions shows a "fast rise and slow fall" and is irregular.

[0057] Since collision noise has a significantly greater asymmetry in waveform compared to white background noise, in order to distinguish different noises, the symmetry factor must first be determined based on the waveform; the calculation method of the asymmetry coefficient P for any pulse segment has been described in detail in Example 1. Furthermore, collision noise exhibits a significant and dramatic amplitude with a discrete distribution, while white background noise has a relatively weak and densely distributed amplitude. This characteristic allows for further differentiation between collision noise and white background noise. For a given pulse segment... Its relative peak and distribution coefficient The calculation is performed.

[0058] Because collision noise exhibits greater asymmetry and a larger distribution coefficient compared to white background noise, features of both can be combined to perform feature filtering on any pulse segment. This completes the feature filtering for any pulse segment. Its feature significance The calculation.

[0059] For each decomposed pulse segment, its corresponding feature significance is calculated, and the result is normalized to the range [0,1]. A double threshold is set here: when... When, the corresponding pulse segment is used as the collision noise segment, when At that time, the corresponding pulse segment was designated as the white background noise segment, while the rest was the normal baseline signal. The two noise segments were then marked on the original electrical signal using different methods.

[0060] This completes the identification of the noise signal. The purpose of identifying the noise signal is to strip it to restore the original trend of the electrical signal. Because the two types of noise have different formation mechanisms, the same smoothing method cannot be used for stripping. For collision noise, since traditional smoothing is essentially "peak shaving and valley filling," it directly kills the sharpness and energy of the particle collision waveform, thus affecting the accuracy of ion concentration detection. Therefore, smoothing is needed at the energy characteristic layer of the particles. A characteristic time sliding window of a preset width (3 seconds in this embodiment) is established for the collision noise. The instantaneous amplitude of the effective particle collision pulse falling within the characteristic time sliding window is squared to calculate the cumulative energy integral or energy variance of the discrete pulses within the window. As the characteristic time sliding window continuously moves in the time domain, the discrete pulse energy integral is converted into a continuous analog curve, and the output is a smooth solid-phase particle collision energy index. This index is positively correlated with the particle suspension concentration and agglomeration degree in the composite electroplating solution.

[0061] For white background noise, the goal is to smooth out the white noise without causing phase lag in the low-frequency baseline. Therefore, this embodiment uses SG filtering to perform polynomial smoothing on the white background noise, thereby filtering out the white background noise while preserving the baseline characteristics.

[0062] Step S3: Based on the stripped signal, periodically adaptively compensate the baseline by analyzing the state changes of the particles in different phase dimensions.

[0063] After identifying and removing noise from the original ion concentration signal using the above method, the remaining electrical signal characteristics are the time-domain signal characteristics of the monitored ion concentration. However, these monitored time-domain signals do not necessarily represent the actual ion concentration changes. After removing high-frequency particle noise, what remains is a... Figure 3 The low-frequency baseline shown appears smooth but is distorted periodically.

[0064] The movement of suspended particles and the surfactants coating their surfaces create a physical barrier near the electrode. The drop caused by this physical barrier is "periodic" because the fluid environment in a real industrial electroplating bath is not absolutely static. The rotation of mechanical agitation and the pulses of the circulating pump cause a dynamic cycle of "accumulation-washing-re-accumulation" of particles and surfactants on the electrode surface. This dynamic cycle curve does not represent the actual concentration of the relevant ions in the electroplating solution. Therefore, adaptive compensation is needed for this periodic curve.

[0065] Data compensation first requires determining the starting point of each barrier cycle. It's important to note that in harsh conditions like composite electroplating—characterized by high noise, strong drift, and non-stationarity—simply determining the starting point by analyzing the trend of periodic curves can lead to phase lag. This phase lag causes the compensation curve and the actual attenuation curve to misalign, affecting the compensation effect. Therefore, this embodiment uses a hardware and software linkage approach to determine the cycle starting point.

[0066] Since circulation is generally caused by mechanical stirring, the scouring of the barrier only occurs the instant the stirring blades are directly facing the sensor probe. When the physical barrier is broken down by the fluid, the blocked ions gradually flow into the electrode, causing the signal to bounce. That is, when the signal bounces, ions have already flowed into the electrode, and circulation has begun. The start time of circulation is generally between these two points.

[0067] For a periodic baseline, obtain all valley moments. The moment when the nearest stirring blade is directly aligned with the sensor probe For any sampling time between the two periods, obtain the one closest to the weight. and starting time .

[0068] This represents the physical mask removal rate near the sensor. A higher value indicates a stronger removal rate of particles by agitation, and a faster influx of ions into the electrode, meaning the cycle starts closer to the start time. Conversely, the closer to Therefore, with We weight the two time points using a weighted average (normalized to the range [0,1]) to determine a more realistic starting time for the cycle. We then select... The time point serves as the starting point of this cycle and also the ending point of the previous cycle.

[0069] Thus, the start and end points of different periods were obtained for this periodic signal. For any given period, as the barrier of suspended particles gradually changes, the accuracy of ion concentration monitoring will change accordingly. When compensating for the concentration signal, a comprehensive analysis based on the electroplating solution itself and the state of the particles is required.

[0070] The main reason physical barriers are generally formed is that when particles aggregate (i.e., multiple small particles clump together to form larger particles), their individual volume and mass increase dramatically, leading to their adhesion to the electrode surface. Simultaneously, the surfactants and kinematic viscosity of the electroplating solution in the liquid phase also promote adhesion. Therefore, the real-time attenuation coefficient within a certain period can be determined by combining these factors. .

[0071] The above method calculates the real-time attenuation coefficient for each period of the periodic ion concentration signal. Since the attenuation coefficient represents the barrier effect of the particles on the electrode, a larger attenuation coefficient corresponds to a deeper barrier, requiring a greater compensation force at that moment. Therefore, the attenuation coefficient (normalized) at any moment within the period can be used as the compensation coefficient, and compensation is performed according to the following formula: ; This method uses the real-time signal attenuation coefficient as a compensation weight to adjust the real-time ion concentration signal after stripping. This effectively eliminates the periodic influence of the physical barrier effect on sensor measurements and improves the accuracy of target metal ion concentration monitoring.

[0072] The above method is used to perform noise stripping and defect compensation on the ion concentration signals collected throughout the monitoring period to obtain the preferred ion concentration signal.

[0073] Step S4: Based on the preferred ion concentration signal, perform trend prediction and issue a fluctuation warning, and then link the underlying execution mechanism to perform manual intervention and chemical supplementation.

[0074] For the optimized ion concentration signal detected and reconstructed in real time, the system continuously extracts its first derivative (instantaneous consumption rate v) and second derivative (consumption acceleration a). Using a nonlinear Taylor expansion extrapolation model, it calculates in real time the remaining absolute safety time before the ion concentration falls below the process limit under the current dynamic consumption trend. This represents a leap from post-event alarm to pre-event zero-delay intervention. When an anomaly occurs or the remaining absolute safety time falls below a preset threshold, the control system issues targeted intervention commands to the underlying actuators, including physical field modulation and precise quantitative chemical replenishment, to maintain the concentration balance of relevant metal ions in the electroplating solution and ensure the quality of composite electroplating.

[0075] In a new embodiment, such as Figure 4 The present invention provides an online monitoring system for electroplating solutions used in composite electroplating, comprising: The signal acquisition unit is used to acquire the time-domain raw electrical signal of the target ion concentration during the composite electroplating process; The denoising unit identifies and removes the first noise generated by the collision of suspended particles and the second noise generated by environmental interference based on the waveform characteristics of the original electrical signal, and obtains the denoised signal. The compensation unit adaptively compensates for the periodic baseline drift caused by the physical barrier effect of suspended particles in the denoised signal to obtain the actual ion concentration signal. The prediction unit performs trend prediction based on the actual ion concentration signal, and issues early warnings and / or generates control commands based on the prediction results.

[0076] Based on the same inventive concept, embodiments of the present invention also provide an electronic device. Figure 5 This is a structural block diagram of an electronic device provided in an embodiment of the present invention. Figure 5 As shown, an embodiment of the present invention provides an electronic device including: one or more processors 101, a memory 102, and one or more I / O interfaces 103. The memory 102 stores one or more programs, which, when executed by the one or more processors, cause the one or more processors to implement any of the online monitoring methods for composite electroplating solutions described in the above embodiments; the one or more I / O interfaces 103 are connected between the processor and the memory, configured to enable information interaction between the processor and the memory.

[0077] The processor 101 is a device with data processing capabilities, including but not limited to a central processing unit (CPU); the memory 102 is a device with data storage capabilities, including but not limited to random access memory (RAM, more specifically SDRAM, DDR, etc.), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), and flash memory (FLASH); the I / O interface (read / write interface) 103 is connected between the processor 101 and the memory 102, and can realize information interaction between the processor 101 and the memory 102, including but not limited to a data bus (Bus).

[0078] In some embodiments, the processor 101, memory 102, and I / O interface 103 are interconnected via bus 104, and thus connected to other components of the computing device.

[0079] In some embodiments, the one or more processors 101 include a field-programmable gate array.

[0080] This invention also provides a computer-readable medium. The computer-readable medium stores a computer program, which, when executed by a processor, implements the steps of any of the online monitoring methods for composite electroplating solutions described in the above embodiments. The computer-readable storage medium can be volatile or non-volatile.

[0081] This invention also provides a computer program product, including computer-readable code, or a non-volatile computer-readable storage medium carrying computer-readable code. When the computer-readable code is run in the processor of an electronic device, the processor in the electronic device executes the above-described online monitoring method for electroplating solution for composite electroplating.

[0082] Those skilled in the art will understand that all or some of the steps, systems, and apparatuses disclosed above, and their functional modules / units, can be implemented as software, firmware, hardware, or suitable combinations thereof. In hardware implementations, the division between functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit (ASIC). Such software can be distributed on a computer-readable storage medium, which may include computer storage media (or non-transitory media) and communication media (or transient media).

[0083] As is known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable program instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), static random access memory (SRAM), flash memory or other memory technologies, portable compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical disc storage, magnetic cartridges, magnetic tape, disk storage or other magnetic storage devices, or any other medium that can be used to store desired information and is accessible to a computer. Furthermore, it is known to those skilled in the art that communication media typically contain computer-readable program instructions, data structures, program modules, or other data in modulated data signals such as carrier waves or other transmission mechanisms, and may include any information delivery medium.

[0084] The computer-readable program instructions described herein can be downloaded from computer-readable storage media to various computing / processing devices, or downloaded via a network, such as the Internet, local area network, wide area network, and / or wireless network, to an external computer or external storage device. The network may include copper transmission cables, fiber optic transmission, wireless transmission, routers, firewalls, switches, gateway computers, and / or edge servers. A network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards them to the computer-readable storage media in the respective computing / processing device.

[0085] The computer program instructions used to perform the operations of this invention may be assembly instructions, instruction set architecture (ISA) instructions, machine instructions, machine-dependent instructions, microcode, firmware instructions, state setting data, or source code or object code written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Smalltalk, C++, etc., and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The computer-readable program instructions may be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or may be connected to an external computer (e.g., via the Internet using an Internet service provider). In some embodiments, electronic circuitry, such as programmable logic circuitry, field-programmable gate arrays (FPGAs), or programmable logic arrays (PLAs), is personalized by utilizing state information from the computer-readable program instructions. This electronic circuitry can execute the computer-readable program instructions to implement various aspects of the invention.

[0086] The computer program product described herein can be implemented specifically through hardware, software, or a combination thereof. In one alternative embodiment, the computer program product is specifically embodied in a computer storage medium; in another alternative embodiment, the computer program product is specifically embodied in a software product, such as a software development kit (SDK), etc.

[0087] Various aspects of the present invention are described herein with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer-readable program instructions.

[0088] These computer-readable program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus to produce a machine such that, when executed by the processor of the computer or other programmable data processing apparatus, they create means for implementing the functions / actions specified in one or more blocks of the flowchart and / or block diagram. These computer-readable program instructions can also be stored in a computer-readable storage medium that causes a computer, programmable data processing apparatus, and / or other device to operate in a particular manner; thus, the computer-readable medium storing the instructions comprises an article of manufacture that includes instructions for implementing aspects of the functions / actions specified in one or more blocks of the flowchart and / or block diagram.

[0089] Computer-readable program instructions may also be loaded onto a computer, other programmable data processing apparatus, or other device to cause a series of operational steps to be performed on the computer, other programmable data processing apparatus, or other device to produce a computer-implemented process, thereby causing the instructions executed on the computer, other programmable data processing apparatus, or other device to perform the functions / actions specified in one or more boxes of a flowchart and / or block diagram.

[0090] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of an instruction, which contains one or more executable instructions for implementing a specified logical function. In some alternative implementations, the functions marked in the blocks may occur in a different order than those shown in the drawings. For example, two consecutive blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, may be implemented using a dedicated hardware-based system that performs the specified function or action, or using a combination of dedicated hardware and computer instructions.

[0091] Example embodiments have been disclosed herein, and while specific terminology has been used, it is for illustrative purposes only and should be construed as such, and is not intended to be limiting. In some instances, it will be apparent to those skilled in the art that features, characteristics, and / or elements described in conjunction with particular embodiments may be used alone, or in combination with features, characteristics, and / or elements described in conjunction with other embodiments, unless otherwise expressly indicated. Therefore, those skilled in the art will understand that various changes in form and detail may be made without departing from the scope of the invention as set forth in the appended claims.

Claims

1. A method for online monitoring of electroplating solution for composite electroplating, characterized in that, include: Obtain the original time-domain electrical signal of the target ion concentration during the composite electroplating process; Based on the waveform characteristics of the original electrical signal, the first noise generated by the collision of suspended particles and the second noise generated by environmental interference are identified and removed to obtain the denoised signal. Based on the periodic baseline drift caused by the physical barrier effect of suspended particles in the denoised signal, adaptive compensation is performed on the periodic baseline drift to obtain the actual ion concentration signal; Trend prediction is performed based on the actual ion concentration signal, and warnings and / or control commands are issued based on the prediction results.

2. The online monitoring method for electroplating solution for composite electroplating according to claim 1, characterized in that, The method for identifying and stripping the first noise and the second noise includes: The original electrical signal is decomposed into multiple pulse segments according to the peak and valley values ​​of the waveform; Calculate the asymmetry coefficient and distribution coefficient for each pulse segment; The feature significance of each pulse segment is determined based on the asymmetry coefficient and the distribution coefficient. Based on the comparison results between the feature saliency and the preset threshold, the multiple pulse segments are classified into collision noise segments, white background noise segments, and normal signal segments; Different smoothing methods were used to peel off the collision noise segment and the white background noise segment respectively.

3. The online monitoring method for electroplating solution for composite electroplating according to claim 2, characterized in that, The preset threshold includes a first threshold and a second threshold, wherein the first threshold is greater than the second threshold; the classification of the plurality of pulse segments into collision noise segments, white background noise segments, and normal signal segments based on the comparison result of the feature salience and the preset threshold includes: When the significance of the feature is greater than or equal to the first threshold, the pulse segment is determined to be a collision noise segment; When the significance of the feature is less than or equal to the second threshold, the pulse segment is determined to be a white background noise segment; When the saliency of the feature is between the second threshold and the first threshold, the pulse segment is determined to be a normal signal segment.

4. The online monitoring method for electroplating solution for composite electroplating according to claim 2, characterized in that, Different smoothing methods are used to peel off the collision noise segment and the white background noise segment, respectively, including: For the collision noise segment, a characteristic time sliding window of a preset width is established. The amplitude of the collision noise pulse falling into the characteristic time sliding window is integrated with energy to convert the discrete pulse energy into a continuous analog quantity curve, which serves as the smoothed collision energy index. For the white background noise segment, polynomial smoothing is performed using SG filtering.

5. The online monitoring method for electroplating solution for composite electroplating according to claim 1, characterized in that, The adaptive compensation for periodic baseline drift caused by the physical barrier effect of suspended particles includes: Obtain the cycle start point of periodic baseline drift; The denoised signal is compensated based on the real-time attenuation coefficient within the current cycle. The attenuation coefficient is used to characterize the degree of attenuation of the ion concentration signal by the physical barrier of suspended particles.

6. The online monitoring method for electroplating solution for composite electroplating according to claim 5, characterized in that, The method for obtaining the cycle start point of the periodic baseline drift includes: Acquire the trough moments of the periodic baseline, as well as the moments when the stirring blades are directly facing the sensor probe; The proximity weight is determined based on the liquid flow rate and suspended particle concentration at the sensor location. The cycle start point is determined based on the trough time, the time when the stirring blade is directly facing the sensor probe, and the proximity weight.

7. The online monitoring method for electroplating solution for composite electroplating according to claim 1, characterized in that, The step of predicting trends based on the actual ion concentration signal and issuing early warnings and / or generating control commands based on the prediction results includes: Extract the first and second derivatives of the actual ion concentration signal; Based on the first and second derivatives, a nonlinear Taylor expansion extrapolation model is used to predict the remaining time for the actual ion concentration signal to fall below a preset threshold. When the remaining time is lower than a preset time threshold, an early warning is issued and / or control commands are generated for the linkage actuator to intervene and supplement.

8. An online monitoring system for electroplating solutions used in composite electroplating, characterized in that, include: The signal acquisition unit is used to acquire the time-domain raw electrical signal of the target ion concentration during the composite electroplating process; The denoising unit identifies and removes the first noise generated by the collision of suspended particles and the second noise generated by environmental interference based on the waveform characteristics of the original electrical signal, and obtains the denoised signal. The compensation unit adaptively compensates for the periodic baseline drift caused by the physical barrier effect of suspended particles in the denoised signal to obtain the actual ion concentration signal. The prediction unit performs trend prediction based on the actual ion concentration signal, and issues early warnings and / or generates control commands based on the prediction results.

9. An electronic device, characterized in that, include: One or more processors; Memory, used to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the online monitoring method for electroplating solutions for composite electroplating as described in any one of claims 1 to 7.

10. A computer-readable medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements an online monitoring method for electroplating solution for composite electroplating as described in any one of claims 1 to 7.

Images