CN117721513A - Self-adaptive silver plating method and system based on spectral analysis - Google Patents
Self-adaptive silver plating method and system based on spectral analysis Download PDFInfo
- Publication number
- CN117721513A CN117721513A CN202410180991.XA CN202410180991A CN117721513A CN 117721513 A CN117721513 A CN 117721513A CN 202410180991 A CN202410180991 A CN 202410180991A CN 117721513 A CN117721513 A CN 117721513A
- Authority
- CN
- China
- Prior art keywords
- silver plating
- concentration
- time
- matrix
- soluble reactant
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000007747 plating Methods 0.000 title claims abstract description 210
- BQCADISMDOOEFD-UHFFFAOYSA-N Silver Chemical compound [Ag] BQCADISMDOOEFD-UHFFFAOYSA-N 0.000 title claims abstract description 174
- 229910052709 silver Inorganic materials 0.000 title claims abstract description 174
- 239000004332 silver Substances 0.000 title claims abstract description 174
- 238000000034 method Methods 0.000 title claims abstract description 101
- 238000010183 spectrum analysis Methods 0.000 title claims abstract description 47
- 239000000376 reactant Substances 0.000 claims abstract description 77
- 230000008569 process Effects 0.000 claims abstract description 37
- 238000001228 spectrum Methods 0.000 claims abstract description 32
- 230000005540 biological transmission Effects 0.000 claims abstract description 26
- 238000000411 transmission spectrum Methods 0.000 claims abstract description 19
- 238000013527 convolutional neural network Methods 0.000 claims abstract description 18
- 238000007781 pre-processing Methods 0.000 claims abstract description 17
- 239000011248 coating agent Substances 0.000 claims abstract description 16
- 238000000576 coating method Methods 0.000 claims abstract description 16
- 238000012549 training Methods 0.000 claims abstract description 13
- 238000012360 testing method Methods 0.000 claims abstract description 5
- 239000011159 matrix material Substances 0.000 claims description 52
- 230000003044 adaptive effect Effects 0.000 claims description 33
- 230000006870 function Effects 0.000 claims description 30
- 238000002474 experimental method Methods 0.000 claims description 15
- 238000012544 monitoring process Methods 0.000 claims description 14
- 230000007704 transition Effects 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 7
- 230000009467 reduction Effects 0.000 claims description 7
- 238000004458 analytical method Methods 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 6
- 238000003860 storage Methods 0.000 claims description 6
- 238000004590 computer program Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 4
- 208000032369 Primary transmission Diseases 0.000 claims description 3
- 238000006243 chemical reaction Methods 0.000 claims description 3
- 238000009672 coating analysis Methods 0.000 claims description 3
- 238000013507 mapping Methods 0.000 claims description 3
- 238000011176 pooling Methods 0.000 claims description 3
- 229920006395 saturated elastomer Polymers 0.000 claims description 3
- 230000011218 segmentation Effects 0.000 claims description 3
- 238000012546 transfer Methods 0.000 claims description 3
- 238000001514 detection method Methods 0.000 abstract description 12
- 238000005457 optimization Methods 0.000 abstract description 3
- 239000000463 material Substances 0.000 description 10
- 239000000523 sample Substances 0.000 description 10
- 230000003287 optical effect Effects 0.000 description 9
- 230000000694 effects Effects 0.000 description 6
- 238000010521 absorption reaction Methods 0.000 description 5
- 230000007547 defect Effects 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 4
- 229910052724 xenon Inorganic materials 0.000 description 4
- FHNFHKCVQCLJFQ-UHFFFAOYSA-N xenon atom Chemical compound [Xe] FHNFHKCVQCLJFQ-UHFFFAOYSA-N 0.000 description 4
- 238000003062 neural network model Methods 0.000 description 3
- 210000002569 neuron Anatomy 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000032683 aging Effects 0.000 description 2
- 238000007796 conventional method Methods 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000009713 electroplating Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011065 in-situ storage Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 239000010956 nickel silver Substances 0.000 description 2
- 201000004569 Blindness Diseases 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015556 catabolic process Effects 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000006731 degradation reaction Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 239000012535 impurity Substances 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 239000007769 metal material Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012806 monitoring device Methods 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000000985 reflectance spectrum Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 238000004381 surface treatment Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
- 239000002699 waste material Substances 0.000 description 1
Abstract
The invention relates to the technical field of silver plating spectrum detection, and discloses a self-adaptive silver plating method and system based on spectrum analysis, wherein the method comprises the following steps: collecting a transmission spectrum chart of a silver plating process; preprocessing the transmission spectrogram, and dividing the preprocessed spectrogram into a training set and a testing set; constructing a convolutional neural network model and performing model training and tuning; collecting a transmission spectrogram in the silver plating process in real time, and calculating the real-time concentration of each soluble reactant through a convolutional neural network; comparing the real-time concentration of each soluble reactant with the target concentration of each soluble reactant, and adjusting silver plating parameters according to the comparison result; and obtaining a reflection spectrum of the coating and analyzing the quality of the coating. The invention can realize real-time control of the silver plating process, accurate plating quality prediction and silver plating parameter optimization, and meets the current requirement on higher and higher silver plating quality.
Description
Technical Field
The invention relates to the technical field of silver plating spectrum detection, in particular to a self-adaptive silver plating method and system based on spectrum analysis.
Background
Silver plating is a common surface treatment technology and is widely applied to the fields of optical devices, display screens, solar cells and the like. Conventional silver plating methods typically employ fixed parameters, which are determined empirically or by trial and error, such as bath composition, temperature, current density, etc. However, the conventional method has some drawbacks that limit uniformity, adhesion, and stability of effects of silver plating.
Firstly, the optical properties and the shape of different materials are greatly different, and the special requirements of different materials cannot be met by the traditional fixed-parameter silver plating method. For example, glass and metallic materials have large differences in optical properties and require personalized adjustments for their specific reflection, transmission and scattering spectra. In addition, with the development of nanotechnology, materials with fine structures and complex shapes are more and more common, and the requirements on silver plating are higher. Secondly, the traditional method mainly relies on experience or trial and error to determine silver plating parameters, and has certain subjectivity and blindness. This approach often achieves satisfactory results by continually trying different silver plating parameters, lacking in science and accuracy. Meanwhile, the traditional method cannot fully consider the influence of factors such as the change of the material and the aging of the plating solution on the silver plating effect, so that the instability of the plating quality is caused. In addition, the silver plating method with fixed parameters is difficult to realize real-time monitoring and adjustment. Since reflection, transmission and scattering spectra are important indicators of the optical properties of materials, the conventional methods cannot acquire and accurately analyze in real time. This makes it impossible to control the variations in the silver plating process in time, resulting in instability of the plating quality and degradation of the production efficiency.
For example, chinese patent with publication number CN113604857B discloses a nano-plating ultrafast laser strengthening and in-situ on-line monitoring device, which comprises an industrial control desk, an ultrafast laser, a spectrometer, a laser probe, a holographic camera, a pulse generator, a mass spectrometer and an optical path system; the laser probe, the holographic camera and the pulse generator are all arranged in the vicinity of the cathode target of the micro-nano device; the industrial personal computer is respectively connected with the ultrafast laser, the spectrometer, the pulse generator, the mass spectrometer and the pulse electroplating power supply; the holographic camera is connected with the spectrometer. The invention can carry out in-situ on-line monitoring on laser electroplating.
The utility model discloses a plating metal surface film thickness spectrum detection device as granted to the chinese patent with bulletin number CN216847450U, belongs to spectrum detection device technical field, including the base, the last fixed surface of base is connected with the detection case, the inner wall fixedly connected with motor of detection case, the front end fixedly connected with half gear of motor, the outside cover of half gear is equipped with the removal frame, the inner wall fixedly connected with tooth of removal frame, half gear and tooth meshing are connected, the lower fixed surface of removal frame is connected with the spectrum appearance, the below of spectrum appearance is equipped with fixture, the last fixed surface of detection case is connected with the controller. The detection result of the spectrum detection device is prevented from being influenced by some impurities in the air, and the detected substances can be detected twice, so that two groups of detection data can be obtained, and then the two groups of data are integrated, so that the detection data can be obtained more accurately, and the error of the detection data is further reduced.
The problems presented in the background art exist in the above patents: the influence of factors such as the change of the material and the aging of the plating solution on the silver plating effect cannot be fully considered, so that the instability of the plating quality is caused, and the real-time monitoring and adjustment cannot be realized.
The information disclosed in this background section is only for enhancement of understanding of the general background of the invention and should not be taken as an acknowledgement or any form of suggestion that this information forms the prior art already known to a person of ordinary skill in the art.
Disclosure of Invention
The invention aims to overcome the defects of the prior art, and provides a self-adaptive silver plating method and system based on spectral analysis, which realize real-time feedback control of a silver plating process, realize accurate plating quality prediction and silver plating parameter optimization, and meet the requirement of higher and higher silver plating quality requirements.
In order to solve the technical problems, the invention provides the following technical scheme:
in one aspect, the invention provides a spectral analysis-based adaptive silver plating method, comprising the steps of:
s1: collecting a transmission spectrum chart of a silver plating process;
s2: preprocessing the transmission spectrogram, and dividing the preprocessed spectrogram into a training set and a testing set;
s3: constructing a convolutional neural network model and performing model training and tuning;
s4: collecting a transmission spectrogram in the silver plating process in real time, and calculating the real-time concentration of each soluble reactant through a convolutional neural network;
s5: comparing the real-time concentration of each soluble reactant with the target concentration of each soluble reactant, and adjusting silver plating parameters according to the comparison result;
s6: and obtaining a reflection spectrum of the coating and analyzing the quality of the coating.
As a preferred embodiment of the spectral analysis-based adaptive silver plating method of the present invention, wherein:
the transmission spectrum is accompanied by concentration information including the concentration value of each soluble reactant during silver plating.
As a preferred embodiment of the spectral analysis-based adaptive silver plating method of the present invention, wherein:
the preprocessing comprises image calibration and image adjustment; wherein the image calibration includes baseline correction, smooth noise reduction, and wavelength calibration; the image adjustment includes image size adjustment, data enhancement, and lot segmentation.
As a preferred embodiment of the spectral analysis-based adaptive silver plating method of the present invention, wherein:
the input of the convolutional neural network model is a transmission spectrogram, and the output is the concentration of each soluble reactant; the structure is as follows:
an input layer for receiving the transmission spectrum as input;
the convolution layer combination is used for extracting the characteristics of the transmission spectrogram, and each convolution layer has independent convolution kernel size and quantity;
the batch normalization layer is used for normalizing the output of the convolution layer;
the pooling layer is used for carrying out characteristic dimension reduction on the output of the convolution layer;
and the full connection layer is used for mapping the output flattening of the convolution layer to the output layer. An output layer for predicting the concentration of each soluble reactant.
As a preferred embodiment of the spectral analysis-based adaptive silver plating method of the present invention, wherein:
the method for calculating the real-time concentration of each soluble reactant is as follows:
setting initial silver plating parameters, starting silver plating, and making silver plating every fixed time intervalCollecting a primary transmission spectrogram and marking the collecting time; and (3) inputting the transmission spectrogram collected each time into the convolutional neural network model after image calibration and size adjustment to obtain the concentration of each soluble reactant during each collection.
As a preferred embodiment of the spectral analysis-based adaptive silver plating method of the present invention, wherein:
the target concentration of each soluble reactant is expressed as a reference concentration of each soluble reactant at each time point in the silver plating process, and the acquisition method is as follows:
silver plating experiments were performed at intervals from the beginning of the silver plating reactionThe concentration of each soluble reactant was measured and recorded as the target concentration for each soluble reactant.
As a preferred embodiment of the spectral analysis-based adaptive silver plating method of the present invention, wherein:
the silver plating parameters include temperature, pH value and current density.
As a preferred embodiment of the spectral analysis-based adaptive silver plating method of the present invention, wherein:
the method for adjusting the silver plating parameters comprises the following steps:
a concentration prediction model of the silver plating process is established, and the model equation is as follows:
；
wherein,representing a concentration matrix at time t, the concentration matrix consisting of concentration data for each soluble reactant during silver plating; a represents a state transition matrix; />For the output of the model, t + is expressed>Predicted values of the concentration matrix at the moment;
a parameter matrix representing the time t, wherein the parameter matrix is composed of silver plating parameters; b represents a parameter transfer matrix;
representing a residual matrix, representing the difference between a concentration prediction model and a real silver plating process; k represents a gain matrix; residual matrix->From the parameter momentMatrix->Concentration matrix->The basis functions of (a) are formed as follows:
；
wherein,represents the i-th basis function,/->A weight parameter representing an ith basis function; the value range of i is 1,2, … …, n, n is the total number of the basis functions.
Collecting target concentration of each soluble reactant in N silver plating experiments and target concentration of each time point, and carrying out parameter identification on silver plating parameters to obtain a concentration prediction model; the formula involved is as follows:
；
wherein θ represents a parameter set of the concentration prediction model, and comprises an element in a state transition matrix A, an element in a parameter transition matrix B, an element in a gain matrix K and a residual matrixThe weight parameters in (a) and the parameters in the basis functions; />Representing the optimal solution of θ, preserve->And applied to the concentration prediction model;
expressed in j-th silver plating experiment, +.>Predicted values of the concentration matrix at the moment; m is the number of time points in each silver plating experiment;
representation in the j-th silver plating experiment, when t= = ->A target concentration matrix at the time; the target concentration matrix is composed of target concentration data of each soluble reactant;
the objective function is calculated as follows:
；
wherein,an objective function representing the time t; />、/>Representing the weight coefficient; />Represents t + and +>A target concentration of the kth soluble reactant at a time; />Represents t + and +>A concentration predicted value of a kth soluble reactant at the moment is obtained based on a concentration matrix output by the concentration prediction model; />The value of the v-th silver plating parameter at the moment t is represented; p is the number of soluble reactant species; q is the number of silver plating parameters;
if objective functionLess than or equal to a preset threshold->The silver plating process is not adjusted; otherwise, linear quadratic programming is used to minimize the objective function +.>Obtaining the optimal value of the silver plating parameter +.>V has a value of 1,2, … …, q; according to->Carrying out feedback adjustment on silver plating parameters;
and iteratively performing the method step of adjusting the silver plating parameters until silver plating is completed.
As a preferred embodiment of the spectral analysis-based adaptive silver plating method of the present invention, wherein:
the linear quadratic programming follows the following constraints:
；
wherein,indicating the pH value of the plating solution at time t, +.>Represents the lowest threshold value of the pH value of the plating solution, +.>Maximum threshold value representing pH value of plating solution；
Indicating the temperature of the plating solution at time t +.>A highest threshold value representing the bath temperature;
indicating the current density at time t +.>A highest threshold representing current density;
represents t + and +>Predicted value of the concentration of the kth soluble reactant at time instant,/->Representing the saturated concentration of the kth soluble reactant.
In a second aspect, the invention provides a self-adaptive silver plating system based on spectrum analysis, which comprises a spectrum acquisition module, a preprocessing module, a spectrum analysis module, a feedback control module, a plating layer analysis module and a monitoring feedback module; wherein:
the spectrum acquisition module is used for acquiring a transmission spectrum chart and a reflection spectrum chart of the plating layer in the silver plating process;
the preprocessing module is used for preprocessing the transmission spectrogram;
the spectrum analysis module is used for calculating the concentration of each soluble reactant in the silver plating process;
the feedback control module is used for comparing the real-time concentration of each soluble reactant with the target concentration and adjusting silver plating parameters according to the comparison result;
the coating analysis module is used for analyzing and evaluating the coating quality;
the monitoring feedback module is used for monitoring the concentration change of each soluble reactant in real time, and sending alarm information to operators when the concentration of any reactant exceeds a specified threshold range.
In a third aspect, the present invention provides an electronic device comprising: a memory for storing instructions; and the processor is used for executing the instructions to enable the equipment to execute the operation of the adaptive silver plating method based on the spectrum analysis.
In a fourth aspect, the present invention provides a computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the spectral analysis based adaptive silvering method of the present invention.
Compared with the prior art, the invention has the following beneficial effects:
and monitoring the reflection spectrum of the material to be silvered in real time by a spectrometer, and obtaining a characteristic curve or a spectral image of the material. By establishing a correlation model between the reflection spectrum and the silver plating parameters, the silver plating parameters can be adjusted in real time according to actual spectrum data, so that the real-time monitoring and control of the silver plating process are realized, problems can be found and adjusted in time in the silver plating process, and unnecessary waste and loss are avoided.
The invention can carry out personalized adjustment according to the optical characteristics of different materials, realize accurate control of silver plating process and obtain a plating layer with more uniformity, stability and high quality. When measuring the concentration of various reactants in the silver plating process, adopting a transmission spectrum; when the quality of the plating layer is analyzed, the advantages of spectrum analysis can be fully exerted by adopting the reflection spectrum, and the regulation and control level of the silver plating process is improved.
Realizing the real-time feedback control of the silver plating process. The method utilizes machine learning to perform feature extraction and pattern recognition on the optical data so as to realize more accurate plating quality prediction and silver plating parameter optimization, has wide application prospect in the fields of photoelectrons, optical devices and the like, and can meet the requirement of higher and higher silver plating quality requirements.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the description of the embodiments will be briefly described below, it being obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art. Wherein:
FIG. 1 is a flow chart of a spectral analysis-based adaptive silver plating method provided by the invention;
fig. 2 is a schematic structural diagram of a spectral analysis-based adaptive silver plating system provided by the invention;
FIG. 3 is a flow chart of a method for adjusting silver plating parameters provided by the invention;
fig. 4 is a schematic diagram of a method for feedback control of silver plating parameters according to the present invention.
Detailed Description
The following detailed description of the present invention is made with reference to the accompanying drawings and specific embodiments, and it is to be understood that the specific features of the embodiments and the embodiments of the present invention are detailed description of the technical solutions of the present invention, and not limited to the technical solutions of the present invention, and that the embodiments and the technical features of the embodiments of the present invention may be combined with each other without conflict.
Example 1
This embodiment describes an adaptive silver plating method based on spectral analysis, referring to fig. 1, the method comprising the steps of:
s1: collecting a transmission spectrum chart of a silver plating process;
the transmission spectrum provides information about the interior of the sample by measuring the change in intensity of the light after it has passed through the sample, thereby analyzing the composition of the sample. In a transmission spectrum image, different elements may exhibit different absorption peaks or bands. The position, shape and intensity of the absorption bands of these absorption peaks are related to the element concentration. By analyzing the characteristics of these peaks in the transmission spectrum image, the concentration of each element can be deduced.
A xenon lamp was used as a light source for collecting the transmission spectrum. Xenon lamps can provide a broad spectrum of light, including the ultraviolet and visible regions. Meanwhile, the intensity of the xenon lamp is stable, and the xenon lamp is suitable for concentration analysis of various elements.
The transmission spectrum is accompanied by concentration information including the concentration value of each soluble reactant during silver plating.
S2: preprocessing the transmission spectrogram, and dividing the preprocessed spectrogram into a training set and a testing set;
the preprocessing comprises image calibration and image adjustment; wherein the image calibration includes baseline correction, smooth noise reduction, and wavelength calibration; the image adjustment includes image size adjustment, data enhancement, and lot segmentation.
First, a baseline correction is performed on a spectral image using piecewise linear correction, and the baseline of the spectral image is adjusted to zero to reduce the effects of baseline shifts. After the baseline is removed, the spectral image is subjected to gaussian filtering to preserve the shape of the absorption peak, absorption band, and reduce the effect of noise. And finally, using characteristic peaks with known wavelengths, and adopting a linear interpolation method to calibrate the wavelength axis of the transmission spectrogram. Through the image calibration processing, the quality of the spectrum image can be gradually improved, and the accuracy of subsequent analysis is ensured.
Cutting the transmission spectrogram after image calibration into a fixed size, and then converting the transmission spectrogram through a data enhancement technology including random rotation, translation and overturn to expand a training set; and finally dividing the transmission spectrogram into different training batches so as to facilitate the input of a convolutional neural network model for batch processing.
S3: constructing a convolutional neural network model and performing model training and tuning;
the input of the convolutional neural network model is a transmission spectrogram, and the output is the concentration of each soluble reactant; the structure is as follows:
an input layer for receiving the transmission spectrum as input;
the convolution layer combination is used for extracting the characteristics of the transmission spectrogram, each convolution layer uses a plurality of convolution kernels to extract the characteristics, and each convolution kernel can learn different characteristics; each convolution layer has independent convolution kernel sizes and numbers to capture features of different scales;
the batch normalization layer is used for carrying out normalization processing on the output of the convolution layer, accelerating training and improving the stability of the model;
the pooling layer is used for carrying out characteristic dimension reduction on the output of the convolution layer;
and the full connection layer is used for mapping the output flattening of the convolution layer to the output layer. Dropout regularization is used to randomly set a proportion of the neuron outputs to 0 to reduce the risk of overfitting.
An output layer for predicting the concentration of each soluble reactant. The output layer has a plurality of neurons, each neuron corresponding to a soluble reactant.
S4: collecting a transmission spectrogram in the silver plating process in real time, and calculating the real-time concentration of each soluble reactant through a convolutional neural network; the method comprises the following steps:
setting initial silver plating parameters, starting silver plating, and making silver plating every fixed time intervalCollecting a primary transmission spectrogram and marking the collecting time; and (3) inputting the transmission spectrogram collected each time into the convolutional neural network model after image calibration and size adjustment to obtain the concentration of each soluble reactant during each collection.
S5: comparing the real-time concentration of each soluble reactant with the target concentration of each soluble reactant, and adjusting silver plating parameters according to the comparison result;
the target concentration of each soluble reactant is expressed as a reference concentration of each soluble reactant at each time point in the silver plating process, and the acquisition method is as follows:
silver plating experiments were performed at intervals from the beginning of the silver plating reactionThe concentration of each soluble reactant was measured and recorded as the target concentration for each soluble reactant.
The silver plating parameters comprise temperature, pH value and current density;
referring to fig. 3, the method for adjusting the silver plating parameters is as follows:
a concentration prediction model of the silver plating process is established, and the model equation is as follows:
；
wherein,representing a concentration matrix at time t, the concentration matrix consisting of concentration data for each soluble reactant during silver plating; a represents a state transition matrix; />For the output of the model, t + is expressed>Predicted values of the concentration matrix at the moment;
a parameter matrix representing the time t, wherein the parameter matrix is composed of silver plating parameters; b represents a parameter transfer matrix;
representing a residual matrix, representing the difference between a concentration prediction model and a real silver plating process; k represents a gain matrix; residual matrix->From the parameter matrix->Concentration matrix->The basis functions of (a) are formed as follows:
；
wherein,represents the i-th basis function,/->A weight parameter representing an ith basis function; the value range of i is 1,2, … …, n, n is the total number of the basis functions. In this embodiment, a polynomial basis function and a wavelet basis function are used, n=2, and the residual matrix is +.>The form of (2) is as follows:
；
wherein,、/>、/>eta represents a weight coefficient, < >>Representing a wavelet function, delta representing a scale parameter; the parameters of the basis function include->、/>、/>、η、δ；
And collecting the target concentration of each soluble reactant in N silver plating experiments and the target concentration of each time point, and carrying out parameter identification on silver plating parameters to obtain a concentration prediction model. The formula involved is as follows:
；
wherein θ represents a parameter set of the concentration prediction model, and comprises an element in a state transition matrix A, an element in a parameter transition matrix B, an element in a gain matrix K and a residual matrixThe weight parameters in (a) and the parameters in the basis functions; />Representing the optimal solution of θ, preserve->And applied to the concentration prediction model;
expressed in j-th silver plating experiment, +.>Predicted values of the concentration matrix at the moment; m is the number of time points in each silver plating experiment;
representation in the j-th silver plating experiment, when t= = ->A target concentration matrix at the time; the target concentration matrix is composed of target concentration data of each soluble reactant;
the objective function is calculated as follows:
；
wherein,an objective function representing the time t; />、/>Representing weight coefficients, and setting and adjusting according to experimental experience; />Represents t + and +>A target concentration of the kth soluble reactant at a time; />Represents t + and +>A concentration predicted value of a kth soluble reactant at the moment is obtained based on a concentration matrix output by the concentration prediction model; />The value of the v-th silver plating parameter at the moment t is represented; p is the number of soluble reactant species; q is the number of silver plating parameters;
setting the threshold of the objective function according to experimental experienceIf the objective function->Less than or equal to->The silver plating process is not adjusted; otherwise, linear quadratic programming is used to minimize the objective function +.>Obtaining the optimal value of the silver plating parametersV has a value of 1,2, … …, q; according to->The silver plating parameters are fed back and adjusted to enable +.>As close to or equal to->。
And iteratively performing the step of adjusting the silver plating parameters until silver plating is completed.
The linear quadratic programming follows the following constraints:
；
wherein,indicating the pH value of the plating solution at time t, +.>Represents the lowest threshold value of the pH value of the plating solution, +.>A highest threshold value representing the pH value of the plating solution;
indicating the temperature of the plating solution at time t +.>A highest threshold value representing the bath temperature;
indicating the current density at time t +.>Representing current densityA highest threshold;
represents t + and +>Predicted value of the concentration of the kth soluble reactant at time instant,/->Representing the saturated concentration of the kth soluble reactant.
The constraint conditions constrain the system state and input, ensure that each silver plating parameter is controlled in a reasonable range in the silver plating process, consider the concentration reasonable range of each soluble reactant in the process of solving the optimal value of the silver plating parameter, and ensure that the optimal value of the silver plating parameter obtained by solving is a feasible solution. The method for feedback control of silver plating parameters is shown in fig. 4.
S6: after silver plating is completed, a reflection spectrum of the plating layer is obtained and the quality of the plating layer is analyzed. The method comprises the following steps:
firstly, collecting reflection spectrograms of different silver plating samples; performing relevant silver plating quality test, and obtaining accurate plating quality through electron microscope observation and thickness measurement; correlating the reflection spectrum with the corresponding coating quality, and establishing a reflection spectrum data set of marked coating quality;
preprocessing the collected reflection spectrogram, including baseline correction, smooth noise reduction and wavelength calibration, so as to improve the accuracy and stability of subsequent modeling;
a neural network model is established, and one of a multi-layer perceptron, a convolutional neural network and a cyclic neural network can be selected; training a neural network model by using a reflection spectrum chart data set with marked coating quality, performing cross verification to evaluate the performance of the model, and performing model tuning;
and predicting a reflection spectrum of the new silver plating sample by using the trained neural network model so as to judge the plating quality of the new silver plating sample.
The reflection spectrum is used for obtaining sample information by measuring the intensity of light reflected back from the surface of the sample, and the method has good effect on analyzing surface defects, morphology and quality of materials. In analyzing surface defects of the coating, the reflectance spectrum may provide information about the surface defects, assisting in detecting and assessing the surface condition of the coating sample.
Example 2
This embodiment is a second embodiment of the present invention; based on the same inventive concept as in example 1, referring to fig. 2, this example describes an adaptive silver plating system based on spectral analysis, comprising:
the device comprises a spectrum acquisition module, a preprocessing module, a spectrum analysis module, a feedback control module, a plating layer analysis module and a monitoring feedback module; wherein:
the spectrum acquisition module is used for acquiring a transmission spectrum chart and a reflection spectrum chart of the plating layer in the silver plating process;
the preprocessing module is used for preprocessing the transmission spectrogram, including background noise removal, wavelength correction and data smoothing, so as to improve the accuracy of subsequent analysis;
the spectrum analysis module is used for training, optimizing and applying a convolutional neural network model, and calculating the concentration of each soluble reactant in the silver plating process by adopting the convolutional neural network through a transmission spectrum image;
the feedback control module is used for comparing the real-time concentration of each soluble reactant with the target concentration and adjusting silver plating parameters according to the comparison result;
the coating analysis module is used for analyzing and evaluating the coating quality according to the reflection spectrogram of the coating;
the monitoring feedback module is used for monitoring the concentration change of each soluble reactant in real time, and sending alarm information to operators when the concentration of any reactant exceeds a specified threshold range and the coating quality is defective.
The specific functions of the above modules are related to the adaptive silver plating method based on spectral analysis provided in reference to embodiment 1, and will not be described in detail.
Example 3
Based on the same inventive concept as the other embodiments, this embodiment introduces an electronic device, including a memory and a processor, where the memory is configured to store instructions, and the processor is configured to execute the instructions, so that the computer device performs the adaptive silver plating method based on spectral analysis provided by the foregoing embodiments.
Since the electronic device described in this embodiment is an electronic device used to implement the adaptive silver plating method based on spectral analysis in this embodiment, based on the adaptive silver plating method based on spectral analysis described in this embodiment, those skilled in the art will be able to understand the specific implementation of the electronic device in this embodiment and various modifications thereof, so how this electronic device is implemented in this embodiment will not be described in detail herein. The electronic device used by those skilled in the art to implement the adaptive silver plating method based on spectral analysis in the embodiments of the present application falls within the scope of protection intended in the present application.
Example 4
Based on the same inventive concept as the other embodiments, this embodiment introduces a computer-readable storage medium on which a computer program is stored, which when executed by a processor, implements the spectral analysis based adaptive silver plating method provided by the above embodiments.
It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as a method, system, or computer program product. Accordingly, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, and the like) having computer-usable program code embodied therein.
The embodiments of the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present invention and the scope of the claims, which are all within the protection of the present invention.
Claims (12)
1. A self-adaptive silver plating method based on spectrum analysis is characterized in that: the method comprises the following steps:
s1: collecting a transmission spectrum chart of a silver plating process;
s2: preprocessing the transmission spectrogram, and dividing the preprocessed spectrogram into a training set and a testing set;
s3: constructing a convolutional neural network model and performing model training and tuning;
s4: collecting a transmission spectrogram in the silver plating process in real time, and calculating the real-time concentration of each soluble reactant through a convolutional neural network;
s5: comparing the real-time concentration of each soluble reactant with the target concentration of each soluble reactant, and adjusting silver plating parameters according to the comparison result;
s6: and obtaining a reflection spectrum of the coating and analyzing the quality of the coating.
2. The spectral analysis-based adaptive silver plating method according to claim 1, wherein: the transmission spectrum is accompanied by concentration information including the concentration value of each soluble reactant during silver plating.
3. The spectral analysis-based adaptive silver plating method according to claim 2, wherein: the preprocessing comprises image calibration and image adjustment; wherein the image calibration includes baseline correction, smooth noise reduction, and wavelength calibration; the image adjustment includes image size adjustment, data enhancement, and lot segmentation.
4. A spectral analysis based adaptive silver plating method according to claim 3, characterized in that: the input of the convolutional neural network model is a transmission spectrogram, and the output is the concentration of each soluble reactant; the structure is as follows:
an input layer for receiving the transmission spectrum as input;
the convolution layer combination is used for extracting the characteristics of the transmission spectrogram, and each convolution layer has independent convolution kernel size and quantity;
the batch normalization layer is used for normalizing the output of the convolution layer;
the pooling layer is used for carrying out characteristic dimension reduction on the output of the convolution layer;
the full connection layer is used for mapping the output flattening of the convolution layer to the output layer;
an output layer for predicting the concentration of each soluble reactant.
5. An adaptive silver plating method based on spectral analysis according to claim 4, characterized in that: the method for calculating the real-time concentration of each soluble reactant is as follows:
setting initial silver plating parameters, starting silver plating, and making silver plating every fixed time intervalCollecting a primary transmission spectrogram and marking the collecting time; and (3) inputting the transmission spectrogram collected each time into the convolutional neural network model after image calibration and size adjustment to obtain the concentration of each soluble reactant during each collection.
6. An adaptive silver plating method based on spectral analysis according to claim 5, characterized in that: the target concentration of each soluble reactant is expressed as a reference concentration of each soluble reactant at each time point in the silver plating process, and the acquisition method is as follows:
silver plating experiments were performed at intervals from the beginning of the silver plating reactionThe concentration of each soluble reactant was measured and recorded as the target concentration for each soluble reactant.
7. An adaptive silver plating method based on spectral analysis according to claim 6, characterized in that: the silver plating parameters include temperature, pH value and current density.
8. An adaptive silver plating method based on spectral analysis according to claim 7, characterized in that: the method for adjusting the silver plating parameters comprises the following steps:
a concentration prediction model of the silver plating process is established, and the model equation is as follows:
；
wherein,representing a concentration matrix at time t, the concentration matrix consisting of concentration data for each soluble reactant during silver plating; a represents a state transition matrix; />For the output of the model, t + is expressed>Predicted values of the concentration matrix at the moment;
a parameter matrix representing the time t, wherein the parameter matrix is composed of silver plating parameters; b represents a parameter transfer matrix;
representing a residual matrix, representing the difference between a concentration prediction model and a real silver plating process; k represents a gain matrix; residual matrix->From the parameter matrix->Concentration matrix->The basis functions of (a) are formed as follows:
；
wherein,represents the i-th basis function,/->A weight parameter representing an ith basis function; the value range of i is 1,2, … …, n, n is the total number of the basis functions;
collecting target concentration of each soluble reactant in N silver plating experiments and target concentration of each time point, and carrying out parameter identification on silver plating parameters to obtain a concentration prediction model; the formula involved is as follows:
；
wherein θ represents a parameter set of the concentration prediction model, and comprises an element in a state transition matrix A, an element in a parameter transition matrix B, an element in a gain matrix K and a residual matrixThe weight parameters in (a) and the parameters in the basis functions; />Representing the optimal solution of θ, preserve->And is applied to the concentration prediction modeA shape;
expressed in j-th silver plating experiment, +.>Predicted values of the concentration matrix at the moment; m is the number of time points in each silver plating experiment;
representation in the j-th silver plating experiment, when t= = ->A target concentration matrix at the time; the target concentration matrix is composed of target concentration data of each soluble reactant;
the objective function is calculated as follows:
；
wherein,an objective function representing the time t; />、/>Representing the weight coefficient; />Represents t + and +>A target concentration of the kth soluble reactant at a time; />Represents t + and +>A concentration predicted value of a kth soluble reactant at the moment is obtained based on a concentration matrix output by the concentration prediction model; />The value of the v-th silver plating parameter at the moment t is represented; p is the number of soluble reactant species; q is the number of silver plating parameters;
if objective functionLess than or equal to a preset threshold->The silver plating process is not adjusted; otherwise, linear quadratic programming is used to minimize the objective function +.>Obtaining the optimal value of the silver plating parameter +.>V has a value of 1,2, … …, q; according to->Carrying out feedback adjustment on silver plating parameters;
and iteratively performing the method step of adjusting the silver plating parameters until silver plating is completed.
9. The spectral analysis-based adaptive silver plating method according to claim 8, wherein: the linear quadratic programming follows the following constraints:
；
wherein the method comprises the steps of，Indicating the pH value of the plating solution at time t, +.>Represents the lowest threshold value of the pH value of the plating solution, +.>A highest threshold value representing the pH value of the plating solution;
indicating the temperature of the plating solution at time t +.>A highest threshold value representing the bath temperature;
indicating the current density at time t +.>A highest threshold representing current density;
represents t + and +>Predicted value of the concentration of the kth soluble reactant at time instant,/->Representing the saturated concentration of the kth soluble reactant.
10. An adaptive silver plating system based on spectral analysis, which is realized based on an adaptive silver plating method based on spectral analysis according to any of claims 1-9, characterized in that: the device comprises a spectrum acquisition module, a preprocessing module, a spectrum analysis module, a feedback control module, a plating layer analysis module and a monitoring feedback module; wherein:
the spectrum acquisition module is used for acquiring a transmission spectrum chart and a reflection spectrum chart of the plating layer in the silver plating process;
the preprocessing module is used for preprocessing the transmission spectrogram;
the spectrum analysis module is used for calculating the concentration of each soluble reactant in the silver plating process;
the feedback control module is used for comparing the real-time concentration of each soluble reactant with the target concentration and adjusting silver plating parameters according to the comparison result;
the coating analysis module is used for analyzing and evaluating the coating quality;
the monitoring feedback module is used for monitoring the concentration change of each soluble reactant in real time, and sending alarm information to operators when the concentration of any reactant exceeds a specified threshold range.
11. An electronic device, comprising: a memory for storing instructions; a processor for executing the instructions, causing the device to perform operations to implement the spectral analysis-based adaptive silver plating method according to any of claims 1-9.
12. A computer readable storage medium, on which a computer program is stored which, when being executed by a processor, implements an adaptive silver plating method based on spectral analysis according to any of claims 1-9.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410180991.XA CN117721513B (en) | 2024-02-18 | 2024-02-18 | Self-adaptive silver plating method and system based on spectral analysis |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410180991.XA CN117721513B (en) | 2024-02-18 | 2024-02-18 | Self-adaptive silver plating method and system based on spectral analysis |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117721513A true CN117721513A (en) | 2024-03-19 |
| CN117721513B CN117721513B (en) | 2024-04-16 |
Family
ID=90211119
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410180991.XA Active CN117721513B (en) | 2024-02-18 | 2024-02-18 | Self-adaptive silver plating method and system based on spectral analysis |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117721513B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060246591A1 (en) * | 2003-03-29 | 2006-11-02 | Harald Engler | Method and device for controlling at least one operating parameter of an electrolytic bath |
| JP2008032719A (en) * | 2006-07-31 | 2008-02-14 | Applied Materials Inc | Raman spectroscopy as integrated chemical measurement |
| CN101896640A (en) * | 2007-12-12 | 2010-11-24 | 朗姆研究公司 | Method and apparatus for plating solution analysis and control |
| CN102507472A (en) * | 2011-10-25 | 2012-06-20 | 清华大学深圳研究生院 | Electrolyte measurement method and electrolyte measurement device for vanadium redox flow battery |
| CN114438548A (en) * | 2022-01-12 | 2022-05-06 | 杭州三耐环保科技股份有限公司 | Additive abnormity monitoring method and system in electrolytic production |
| CN114836813A (en) * | 2022-05-19 | 2022-08-02 | 武汉光目科技有限公司 | In-situ multifunctional electroplating bath device and working method |
| CN117054352A (en) * | 2022-05-06 | 2023-11-14 | 湖南文理学院 | Electroplating solution concentration detection method based on spectral analysis and scattering correction |
-
2024
- 2024-02-18 CN CN202410180991.XA patent/CN117721513B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060246591A1 (en) * | 2003-03-29 | 2006-11-02 | Harald Engler | Method and device for controlling at least one operating parameter of an electrolytic bath |
| JP2008032719A (en) * | 2006-07-31 | 2008-02-14 | Applied Materials Inc | Raman spectroscopy as integrated chemical measurement |
| CN101896640A (en) * | 2007-12-12 | 2010-11-24 | 朗姆研究公司 | Method and apparatus for plating solution analysis and control |
| CN102507472A (en) * | 2011-10-25 | 2012-06-20 | 清华大学深圳研究生院 | Electrolyte measurement method and electrolyte measurement device for vanadium redox flow battery |
| CN114438548A (en) * | 2022-01-12 | 2022-05-06 | 杭州三耐环保科技股份有限公司 | Additive abnormity monitoring method and system in electrolytic production |
| CN117054352A (en) * | 2022-05-06 | 2023-11-14 | 湖南文理学院 | Electroplating solution concentration detection method based on spectral analysis and scattering correction |
| CN114836813A (en) * | 2022-05-19 | 2022-08-02 | 武汉光目科技有限公司 | In-situ multifunctional electroplating bath device and working method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117721513B (en) | 2024-04-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109253985B (en) | Method for identifying wood grade for koto panel by near infrared spectrum based on neural network | |
| CN108596246A (en) | The method for building up of soil heavy metal content detection model based on deep neural network | |
| WO2018184262A1 (en) | Dynamic calibration method for echelle spectrometer for laser induced breakdown spectrum collection | |
| CN110889442B (en) | Object material classification method for pulse type TOF depth camera | |
| CN104730042A (en) | Method for improving free calibration analysis precision by combining genetic algorithm with laser induced breakdown spectroscopy | |
| CN108573105A (en) | The method for building up of soil heavy metal content detection model based on depth confidence network | |
| CN114619292A (en) | Milling cutter wear monitoring method based on fusion of wavelet denoising and attention mechanism with GRU network | |
| CN108802002B (en) | Silkworm egg Raman spectrum model construction method for rapidly identifying and removing diapause without damage | |
| CN111999258B (en) | Spectral baseline correction-oriented weighting modeling local optimization method | |
| CN114330647A (en) | Model training method and device and silicon rod weight prediction method | |
| CN115112699A (en) | XRF soil heavy metal element quantitative analysis method | |
| CN114216877B (en) | Automatic detection and reconstruction method and system for spectral peak in tea near infrared spectral analysis | |
| CN117721513B (en) | Self-adaptive silver plating method and system based on spectral analysis | |
| CN113033704B (en) | Intelligent judging method and system for copper converter converting copper-making final point based on pattern recognition | |
| CN113567375B (en) | Self-adaptive multi-metal ion concentration regression prediction method and system based on linear feature separation | |
| CN116026780A (en) | Method and system for online detection of coating moisture absorption rate based on series strategy wavelength selection | |
| CN112801936B (en) | Self-adaptive background subtraction method for X-ray fluorescence spectrum | |
| CN115561282A (en) | Detection method and system for heavy metal in underground water | |
| CN114354667A (en) | Element quantitative analysis method and system based on X-ray fluorescence spectrum | |
| CN112697746A (en) | Spectrum information-based sweet potato powder alum content detection method | |
| Lin et al. | A feature domain space transfer method for improving identification of maize haploid seed based on near-infrared spectroscopy | |
| CN111624194A (en) | Blade element nondestructive measurement method based on laser-induced breakdown spectroscopy | |
| CN115901677B (en) | Method for predicting concentration of ammonium nitrate in nitric acid-ammonium nitrate solution with updating mechanism | |
| CN117807497A (en) | Method and system for quantitatively analyzing lithium element in field | |
| Hu et al. | Quantitative accuracy improvement of laser-induced breakdown spectroscopy by a modified corner-cutting method for continuum background removal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |