Control method of composite copper foil thickness control apparatus

Abstract

The application provides a control method of a composite copper foil thickness control device, comprising the following steps: calculating the deposition thickness of each abnormal concentration group at a future time according to a deposition rate change rule diagram, obtaining the copper layer thickness distribution of the abnormal area, comparing the copper layer thickness distribution of the abnormal area with the thickness distribution of a normal area, and calculating the thickness deviation value of the abnormal area; for the thickness deviation value, a deviation compensation algorithm is adopted, the electroplating process parameters of the abnormal area are adjusted according to the size and sign of the deviation value, the electroplating process parameters include current density and electrolyte flow rate; and the power supply system and the electrolyte circulation system of the composite copper foil thickness control device are parameter set and adjusted according to the optimized electroplating process parameters, so that the composite copper foil thickness control device produces according to the optimized electroplating process parameters.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of information technology, in particular to a control method of a composite copper foil thickness control device. BACKGROUND

[0002] There is a key technical problem in the production process of composite copper foil electroplating: how to accurately control the uniformity of copper layer thickness. This problem is caused by the complex coupling relationship between the three physical fields of electric field distribution, electrolyte flow and copper ion concentration distribution. Specifically, the non-uniformity of the electric field distribution will cause local distortion of the electric field, which will affect the deposition rate of copper ions; at the same time, the flow disturbance of the electrolyte will also cause the non-uniformity of the copper ion concentration distribution, thereby causing local deviation of the deposition thickness. These factors interact and restrict each other, making it extremely difficult to accurately control the thickness of the copper layer. More specifically, the morphology and thickness parameters of the interface layer will directly affect the electric field distribution and the flow state of the electrolyte, and then affect the mass transfer process of copper ions. Different shaped interface layers will produce different electric field distribution and flow field characteristics, causing local abnormalities in copper ion concentration. The deposition rate of these abnormal areas is significantly different from that of normal areas, ultimately causing the non-uniformity of the copper layer thickness. In addition, the time evolution of the copper ion concentration is also an important problem. Since the electroplating process is a dynamic process, the concentration change trend of the abnormal area will directly affect the future deposition thickness. How to accurately predict this dynamic change and adjust the electroplating process parameters is the key to realizing accurate thickness control. In summary, the multi-physical field coupling effect, interface layer influence, abnormal area identification, dynamic concentration prediction and a series of problems in the production process of composite copper foil electroplating, together constitute the core technical problem of copper layer thickness uniformity control. SUMMARY

[0003] The present application provides a control method of a composite copper foil thickness control device, mainly comprising:

[0004] According to the mutual coupling relationship of the electric field, flow field and mass transfer process of the electroplating process, a multi-physical field coupling model describing the production process of the composite copper foil is constructed, the interface layer morphology and thickness parameters of the composite copper foil are integrated into the multi-physical field coupling model, and the electric field distribution, electrolyte flow state and copper ion concentration distribution data on different shaped interface layers are obtained by simulation of the multi-physical field coupling model;

[0005] For the electric field distribution data obtained by simulation, it is judged whether there is a local distortion phenomenon of the electric field, if there is a local distortion of the electric field, the position coordinates of the distortion area are determined, and the electric field data of the distortion area are marked as abnormal data, at the same time, according to the electrolyte flow state data, it is judged whether there is a flow disturbance phenomenon, if there is a disturbance, the position coordinates of the disturbance area are determined, and the flow field data of the disturbance area are marked as abnormal data;

[0006] According to the copper ion concentration distribution data, the copper ion concentration is clustered to obtain copper ion groups at different concentration levels, and the center coordinates of the copper ion groups at each concentration level are calculated, according to the position coordinates of the distortion region and the disturbance region, whether the center coordinates of the copper ion groups at each concentration level are located in the abnormal region is judged, if located in the abnormal region, the copper ion group at the concentration level is marked as an abnormal concentration group;

[0007] For the abnormal concentration group, a time series prediction algorithm is used to predict the concentration change at the future time according to the historical time copper ion concentration change trend, and the concentration change curve of each abnormal concentration group is obtained, and according to the concentration change curve, a curve fitting algorithm is used to obtain the deposition rate change rule diagram of each abnormal concentration group;

[0008] According to the deposition rate change rule diagram, the deposition thickness of each abnormal concentration group at the future time is calculated, and the copper layer thickness distribution of the abnormal region is obtained, and the copper layer thickness distribution of the abnormal region is compared with the thickness distribution of the normal region, and the thickness deviation value of the abnormal region is calculated;

[0009] For the thickness deviation value, a deviation compensation algorithm is used to adjust the electroplating process parameters of the abnormal region according to the size and sign of the deviation value, and the electroplating process parameters include current density and electrolyte flow rate;

[0010] According to the optimized electroplating process parameters, the power system and the electrolyte circulation system of the composite copper foil thickness control equipment are parameter set and adjusted, so that the composite copper foil thickness control equipment produces according to the optimized electroplating process parameters.

[0011] The technical scheme provided by the embodiment of the present application can include the following beneficial effects:

[0012] The present application discloses a control method of a composite copper foil thickness control equipment. The method builds a multi-physical field coupling model to simulate the interaction of electric field, flow field and mass transfer process, and integrates the interface layer morphology and thickness parameters into the model. Through simulation, the electric field distribution, electrolyte flow and copper ion concentration data are obtained, and the abnormal region is identified. For the copper ion concentration group of the abnormal region, time series prediction and curve fitting algorithms are used to predict future deposition rate and thickness distribution. Compare the thickness of the abnormal region with the normal region to calculate the deviation value. According to the deviation value, a compensation algorithm is used to adjust the electroplating process parameters, including current density and electrolyte flow rate. Finally, the composite copper foil thickness control equipment is parameter set and adjusted, the anode plate position angle and electrolyte flow rate pressure are automatically adjusted, the current density distribution is uniform, and the electrolyte flow is stable. The present application realizes the accurate control of the thickness in the production process of the composite copper foil through multi-physical field coupling analysis and intelligent algorithm optimization, improves the product quality and production efficiency. BRIEF DESCRIPTION OF DRAWINGS

[0013] Figure 1 A flow chart of a control method of the composite copper foil thickness control device. DETAILED DESCRIPTION

[0014] The technical solutions of the present application will be described clearly and completely below in conjunction with embodiments. Obviously, the described embodiments are only some of the embodiments of the present application, but not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present application.

[0015] As Figure 1 , the control method of the composite copper foil thickness control device can specifically include:

[0016] In step S101, a multi-physical field coupling model describing the production process of the composite copper foil is constructed according to the mutual coupling relationship of the electric field, the flow field and the mass transfer process of the electroplating process, the interface layer topography and thickness parameters of the composite copper foil are integrated into the multi-physical field coupling model, and the data of the electric field distribution, the electrolyte flow state and the copper ion concentration distribution on the different shaped interface layers are obtained by simulation of the multi-physical field coupling model.

[0017] The quadrilateral grid elements are divided on the surface of the composite copper foil substrate by using the finite element numerical calculation method, the interface layer thickness distribution function is calculated according to the three-dimensional coordinates of the grid element boundary points; the three-dimensional geometric structure digital model of the composite copper foil is established according to the interface layer thickness distribution function, the finite volume method is used to discretely solve the Navier-Stokes equation to obtain the electrolyte velocity field and pressure field distribution data near the interface layer; according to the electrolyte velocity field and pressure field distribution data, the finite element method is used to solve the Poisson equation to obtain the potential distribution, and the current density at the interface layer is calculated in combination with the electrolyte conductivity distribution function; for the current density at the interface layer, the copper ion concentration control equation is established, the velocity field transport term and the diffusion coefficient diffusion term are combined, and the finite difference method is used to solve the mass transfer equation to obtain the ion concentration distribution data.

[0018] Specifically, the finite element numerical calculation is adopted to divide the surface of the composite copper foil substrate into quadrilateral grid units, the interface layer thickness distribution function is calculated according to the three-dimensional coordinates of the grid unit boundary points, and the three-dimensional geometric structure digital model of the composite copper foil is constructed by combining the substrate surface roughness parameters and the interface layer thickness data. The dynamic parameters such as the electrolyte density, viscosity and temperature are imported, the Navier-Stokes equation is discretely solved by using the finite volume method, and the three-dimensional velocity field and pressure field distribution data near the interface layer are obtained by using the three-dimensional geometric structure digital model of the composite copper foil as the boundary condition. Based on the three-dimensional geometric structure of the composite copper foil and the electrolyte flow field data, the finite element method is used to solve the Poisson equation to calculate the potential distribution, the electric field intensity distribution is calculated by combining the electrolyte conductivity distribution function, and the current density calculation expression at the interface layer is established. The current density distribution at the interface layer is used as the electrochemical reaction source term, the copper ion concentration control equation is established, the mass transfer equation is solved by using the finite difference method by combining the velocity field transport term and the diffusion coefficient diffusion term. According to the electric field intensity, fluid velocity and ion concentration distribution data, the Newton iteration method is used to solve the multi-physical field coupling equation set to obtain the steady-state solution as the multi-physical field coupling model of the composite copper foil electroplating process. The multi-physical field coupling model is substituted into different interface layer morphology parameters, and the Runge-Kutta method is used to solve the time advancement of the coupling equation set to obtain the evolution data of each physical field with time. In the production process of the composite copper foil, the key factors affecting the electroplating quality are the interface layer structure and the multi-physical field distribution. Taking the interface layer thickness distribution function h(x) as an example, when the substrate surface of the composite copper foil presents a periodic rough structure, the height distribution can be described by using Fourier series expansion: h(x) = h0 + Σ(Ancos(2πnx / L) + Bnsin(2πnx / L)), wherein h0 is the average thickness, An and Bn are Fourier coefficients, L is the period length, and x is the position parameter of different interface layer morphologies. The first five order Fourier coefficients are obtained by least square fitting of the measured profile data, and an accurate three-dimensional geometric model is established. After obtaining the geometric model, the calculation of the electrolyte flow field is particularly important. For the typical laminar flow state in the electroplating tank, the Reynolds number Re = ρvD / μ, wherein ρ is the electrolyte density, v is the characteristic velocity, D is the hydraulic diameter, and μ is the dynamic viscosity. The Navier-Stokes equation is discretely solved by using the finite volume method, the grid size is taken as 1 / 10 of the interface layer thickness, i.e. 1 micrometer, the pressure-velocity coupling equation is solved by using the SIMPLE algorithm, and the three-dimensional velocity field distribution near the interface layer is obtained. The calculation of the electric field distribution is based on the Poisson equation of the potential: wherein the electrolyte conductivity κ varies with the ion concentration c: κ = κ0(1 + α(c-c0)), wherein κ0 is the standard state conductivity 5.8 S / m, c0 is the standard ion concentration 0.6 mol / L, and α is the concentration coefficient 0.25 L / mol. The potential -0.3 V is applied to the cathode surface, and the potential 0 V is applied to the anode surface. The potential distribution φ is obtained by solving the equation by using the finite element method, and the electric field intensity The mass transfer process is governed by the convection-diffusion equation: where the diffusion coefficient D = 6.5 x 10 -10 m 2 / s. At the interface layer, the current density and ion flux satisfy the relationship: j = nFN, n is the valence number 2, F is the Faraday constant 96485 C / mol, N is the ion flux mol / (m 2 ·s). Using the finite difference format of staggered grid, the time step is taken as 0.001 s, the spatial step is consistent with the flow field calculation, and the concentration field evolution is obtained. When solving the coupled equations of multiple physical fields, the relative error tolerance is selected as 10 -6 , and the Newton iteration method is used for solving. For a given interface layer morphology, such as a periodic groove structure with a 30-degree inclination and a 75-micron spacing, the calculation shows that the current density at the groove bottom is reduced by 25%, while the current density at the edge is increased by 35%, which is consistent with the observed thickness distribution trend of the plated layer. Through the fourth-order Runge-Kutta method for time advancement, the time step is 0.01 s, and the calculation period is 10 s, and the dynamic evolution characteristics of each physical field can be obtained.

[0019] In step S102, for the electric field distribution data obtained by simulation, it is judged whether there is a local distortion phenomenon of electric field. If there is a local distortion of electric field, the position coordinates of the distortion region are determined, and the electric field data of the distortion region are marked as abnormal data. At the same time, according to the electrolyte flow state data, it is judged whether there is a flow disturbance phenomenon. If there is disturbance, the position coordinates of the disturbance region are determined, and the flow field data of the disturbance region are marked as abnormal data.

[0020] According to the grid point electric field distribution data, the electric field intensity gradient value is calculated. If the gradient value exceeds three times the standard deviation of the local average value, the three-dimensional space clustering method is used to obtain the boundary point coordinate set of the distortion region. The velocity gradient tensor and the curl field are calculated from the grid point velocity vector field. If the curl value is greater than twice the laminar reference value, the boundary tracking algorithm is used to obtain the boundary point coordinate set of the disturbance region. The grid point matching is performed for the distortion region boundary point coordinate set and the disturbance region boundary point coordinate set to obtain the volume fraction of the overlapping region. If the volume fraction is greater than a critical value, it is determined that the region is a strong coupling abnormal region. The thermocouple array is used to obtain the temperature field distribution of the strong coupling abnormal region, and the local electrical conductivity is obtained according to the temperature field distribution and the temperature coefficient of electrical conductivity.

[0021] Specifically, according to the electric field distribution data, the electric field intensity gradient value g(x, y, z) of each grid point is calculated, if the gradient value g(x, y, z) exceeds three times of the local average value, the three-dimensional space clustering method is used to determine the boundary point coordinate set P(x, y, z) of the distortion region, and the position marker matrix M1 of the distortion region is generated. From the fluid grid point velocity vector field v(x, y, z), the velocity gradient tensor and the vorticity field ω(x, y, z) are calculated, if the vorticity value is greater than twice the laminar reference value or the velocity gradient exceeds the Reynolds stress threshold, the boundary tracking algorithm is used to determine the boundary point coordinate set Q(x, y, z) of the disturbance region, and the position marker matrix M2 of the disturbance region is generated. The grid point matching is performed on the electric field distortion region marker matrix M1 and the flow field disturbance region marker matrix M2, the volume fraction α(x, y, z) of the overlapping region is calculated, if the volume fraction α(x, y, z) is greater than the critical value, the region is marked as a strong coupling abnormal region. The thermocouple array is used to obtain the temperature field distribution T(x, y, z) of the abnormal region, and the local conductivity κ(x, y, z) is calculated combined with the temperature coefficient of conductivity to generate the temperature-compensated electric field intensity distribution E(x, y, z). According to the distortion region position marker matrix M1 and the disturbance region position marker matrix M2, the electric field intensity E(x, y, z), the fluid velocity v(x, y, z) and the temperature T(x, y, z) data of the abnormal region are extracted, and the Hilbert transform is used to calculate the instantaneous frequency characteristics of the abnormal data. The wavelet decomposition is performed on the abnormal region data, the abnormal feature coefficients at different scales are extracted, the multi-scale abnormal feature vector is generated, and the abnormal region feature database is established. In the electric field distortion detection process, the calculation of the electric field intensity gradient uses the central difference format:

[0022] g(x, y, z) = [(E(x+h, y, z)-E(x-h, y, z)) / 2h, (E(x, y+h, z)-E(x, y-h, z)) / 2h, (E(x, y, z+h)-E(x, y, z-h)) / 2h], wherein h is the grid spacing value 0.1 mm. For a typical electroplating tank, when the gradient value of the local region exceeds the mean value plus 3 times of the standard deviation, i.e. |g(x, y, z)| > μg+3σg, the point is marked as a distortion point, wherein μg is the gradient field mean value 0.5 kV / m 2 , σg is the standard deviation 0.15 kV / m 2 . In the flow field disturbance judgment, the velocity gradient tensor is also calculated by the central difference method, and the vorticity field The local vortex intensity is characterized. Under the laminar flow condition, the reference vorticity value is 10 s-1, and when the actual vorticity exceeds 20 s-1, it indicates that there is significant disturbance. The Reynolds stress threshold is set to 0.1 Pa, corresponding to a turbulent intensity of 5%. The grid overlap analysis is used in the region matching process, and the volume fraction a(x, y, z) represents the proportion of the unit volume in which the electric field distortion and the flow field disturbance exist simultaneously. When the a value exceeds 0.3, it indicates that there is a strong coupling anomaly in the region. In the actual electroplating process, a typical strong coupling anomaly region has a size of about 5 mm x 5 mm x 2 mm. The influence of temperature field distribution on electrical conductivity is described by a linear relationship: K(T) = K0[1 + β(T - T0)], where K0 is the electrical conductivity at the reference temperature T0 = 298 K, 5.8 S / m, and β is the temperature coefficient, 0.02 K-1. When the temperature rises to 308 K, the electrical conductivity increases by about 20%, resulting in a corresponding decrease in the local electric field intensity. The Hilbert transform is used to extract the instantaneous frequency characteristics of the abnormal signal. After transforming the electric field intensity data E(t), the analytical signal z(t) = E(t) + jH[E(t)] is obtained, where H[E(t)] is the Hilbert transform result. The instantaneous frequency f(t) = (1 / 2π)·d(argz(t)) / dt reflects the frequency characteristics of the electric field fluctuation in the abnormal region, and the typical value is in the range of 0.1-10 Hz. Wavelet decomposition uses db4 wavelet basis to decompose the abnormal region data into 4 layers to obtain different scale coefficients. The first layer coefficient reflects the high-frequency disturbance (>5 Hz), and the fourth layer coefficient reflects the low-frequency change (<0.5 Hz). By analyzing the energy distribution of each scale coefficient, the dominant frequency component of the disturbance can be identified. When the energy proportion of the high-frequency component exceeds 50%, it indicates that there is severe local fluctuation. This multi-scale feature extraction method can comprehensively characterize the dynamic characteristics of the abnormal region and provide a basis for subsequent process parameter optimization.

[0023] In step S103, the copper ion concentration is grouped according to the copper ion concentration distribution data to obtain copper ion groups of different concentration levels, and the center coordinates of the copper ion groups of each concentration level are calculated. According to the position coordinates of the distortion region and the disturbance region, it is judged whether the center coordinates of the copper ion groups of each concentration level are located in the abnormal region. If it is located in the abnormal region, the copper ion group of the concentration level is marked as an abnormal concentration group.

[0024] According to the copper ion concentration distribution data, a three-dimensional concentration field is constructed, and three concentration level regions of high, medium and low are obtained by a density clustering algorithm; the centroid coordinates and point number density of the concentration level regions are obtained, and the centroid coordinates and point number density are used to obtain the population spatial distribution function and population characteristic radius by a Gaussian kernel function; an abnormal region spatial envelope surface is constructed by using the boundary point set of the electric field distortion region and the fluid disturbance region, and the shortest distance value is obtained from the abnormal region spatial envelope surface and the concentration group centroid; whether the ratio of the shortest distance value and the population characteristic radius is less than a critical value is judged, and if the ratio is less than the critical value, the concentration group is marked as an abnormal group and an abnormal group spatial distribution feature matrix is generated.

[0025] Specifically, according to the copper ion concentration distribution data c(x, y, z), a three-dimensional concentration field is constructed, and a density clustering algorithm based on Euclidean distance is used to divide the concentration field into three concentration level regions C1, C2 and C3 by using a concentration gradient threshold value and a minimum point number threshold value n. The centroid coordinates r=(x, y, z) of each concentration level region C are calculated, the point number density p(r) in the region is counted, and the population spatial distribution function f(r) and the population characteristic radius R are calculated by using a Gaussian kernel function. Based on the boundary point set of the electric field distortion region D1 and the fluid disturbance region D2, an abnormal region spatial envelope surface S(x, y, z) is constructed, and the shortest distance d from the concentration group centroid to the envelope surface is calculated. The ratio l=d / R of the shortest distance d from the concentration group centroid to the abnormal region and the population characteristic radius R is used as a judgment parameter, and if the ratio l is less than a critical value lc, the concentration group is marked as an abnormal group. For the concentration groups that overlap in space, the overlap volume ratio g j is calculated, and if the overlap volume ratio g j is greater than an overlap degree threshold value gc, the overlapping regions are merged to form a new concentration group, and the population characteristic parameters are updated. An encoding mapping is established for the marked abnormal concentration group, and an abnormal group spatial distribution feature matrix M(x, y, z) is generated, and the matrix elements include concentration values, centroid positions, population radii and other characteristic quantities. In the copper ion concentration distribution analysis, the density clustering algorithm distinguishes different concentration level regions by setting a concentration gradient threshold value and a minimum point number. Taking an actual electroplating tank as an example, when the concentration gradient threshold value and the minimum point number threshold value n=100 are set, the concentration field can be divided into a high concentration region (c>0.8 mol / L), a medium concentration region (0.4-0.8 mol / L) and a low concentration region (c<0.4 mol / L). For the divided concentration groups, the centroid coordinates are calculated by using a weighted average method: r=Σ(c j r j ) / Σc j , where c j is the concentration value at point j, and r jis the position vector of the group. The group spatial distribution function takes the form of a Gaussian kernel: φ(r) = exp[-(r-r 2 / 2σ 2 ], where σ is the characteristic width with a value of 2 mm. The group characteristic radius R is calculated by integrating ∫φ(r)d 3 r, and the typical value is in the range of 3-5 mm. In the abnormal region determination, if the shortest distance d of the centroid of a concentration group to the boundary of the abnormal region is 2 mm, and the characteristic radius R of the concentration group is 4 mm, then the determination parameter λ = 0.5 is determined. Set the critical value λc = 0.8, and the concentration group is marked as an abnormal group. This distance ratio-based determination method considers the spatial scale of the concentration group, avoiding the deviation that may be caused by a simple distance threshold. For the case of spatial overlap, the overlap volume ratio γj is calculated by Monte Carlo integration: randomly scatter points in the overlap region, and calculate the proportion of points located in both concentration groups. When γ j > 0.3, it indicates that there is significant overlap, and the group merging needs to be performed. The characteristic parameters of the merged new concentration group are obtained by weighted averaging, and the weight is proportional to the number of points of the original group. The abnormal group feature matrix M(x, y, z) adopts a multi-dimensional data structure, and each matrix element contains information such as concentration value c(x, y, z), centroid position r = (x, y, z), group radius R, and point density ρ. Through the matrix, the evolution process of the abnormal concentration group can be tracked, and it is observed that in the electric field distortion region, the ion concentration often presents a local accumulation phenomenon, and the concentration gradient can reach 0.2 mol / (L·mm); while in the fluid disturbance region, the ion distribution tends to be uniform, and the concentration gradient is reduced to below 0.02 mol / (L·mm). This multi-scale feature analysis method can comprehensively characterize the concentration distribution characteristics of the abnormal region, and provides an important basis for process parameter optimization.

[0026] In step S104, for the abnormal concentration group, a time series prediction algorithm is used to predict the concentration change at the future time according to the historical copper ion concentration change trend, to obtain the concentration change curve of each abnormal concentration group, and simultaneously, a curve fitting algorithm is used to obtain the deposition rate variation law diagram of each abnormal concentration group according to the concentration change curve.

[0027] Resample the historical data of the abnormal concentration group with a fixed sampling interval to obtain equidistant time series data and concentration change rate sequence; train a long short-term memory network model using the concentration change rate sequence, and output a next-time concentration prediction value through the network model; calculate the adjacent-time concentration difference value for the concentration prediction value, and calculate the theoretical deposition thickness increment sequence according to the concentration difference value and Faraday's law; perform Fourier transform on the theoretical deposition thickness increment sequence to extract frequency components, and perform polynomial fitting on the frequency components to obtain a deposition rate variation law function, and if the root mean square error of the polynomial fitting is greater than a preset threshold, increase the polynomial order and repeat the fitting process.

[0028] Specifically, the time series {t1, c1} is constructed according to the historical data of the abnormal concentration group, the concentration data is resampled with a fixed sampling interval δt to generate the equidistant time series data {t, c}, and the first-order concentration change rate sequence {t, Δc / Δt} is obtained through a difference operation. The time window width τ is set, the long short-term memory network is used to model the concentration change rate sequence, the network input is the concentration change rate value at τ consecutive time points, the output is the concentration prediction value at the next time point, and the network parameters are optimized through a back propagation algorithm. The concentration difference between adjacent time points is calculated for the predicted concentration sequence, the theoretical deposition thickness increment Δh at each time point is calculated based on Faraday's law and the current density distribution, and the cumulative sequence {t, h} of the thickness over time is generated. The cumulative thickness sequence is subjected to Fourier transform, the main frequency component ω and its corresponding amplitude A in the amplitude spectrum are extracted, and the deposition rate time function v(t) is obtained through inverse transform reconstruction. The deposition rate v(t) is least square fitted by using an n-order polynomial function, the polynomial coefficients {a} are obtained, and the root mean square error σ of the fitting curve and the measured data is calculated. According to the comparison result of the root mean square error σ and the preset threshold, the polynomial order n is adjusted and the fitting process is repeated until the error meets the requirements, and the final deposition rate change law function f(t) is obtained. In the prediction of copper ion concentration, the construction of the time series is crucial to the prediction accuracy. Taking an actual electroplating process as an example, when the sampling interval δt is set to 0.5 seconds, the equidistant sequence {0s, 0.6mol / L; 0.5s, 0.59mol / L; 1.0s, 0.55mol / L} is obtained after resampling the original concentration data points {0s, 0.6mol / L; 0.47s, 0.58mol / L; 0.91s, 0.55mol / L}, and the concentration change rates are -0.02mol / (L·s) and -0.08mol / (L·s) through difference calculation. The time window width τ of the long short-term memory network is set to 10 sampling points (i.e. 5 seconds), the input layer contains 10 neurons corresponding to the concentration change rate values within the window, the hidden layer is set to 64 neurons, and the output layer has 1 neuron to predict the concentration value at the next time point. Through 500 iterations of training, the prediction error decreases from the initial 0.15mol / L to 0.01mol / L. Based on Faraday's law, the copper ion deposition thickness increment calculation formula is: Δh = (M / nF·ρ)·j·Δt, where M is the molar mass of copper 63.5g / mol, n is the valence number 2, F is the Faraday constant 96485C / mol, ρ is the copper density 8.9g / cm 3 , j is the local current density. When the current density is 50mA / cm 2The theoretical deposition thickness increment within 1 second is 0.017 μm. The fast Fourier transform is performed on the cumulative thickness sequence with a sampling point number of 1024, and the main fluctuation periods in the spectrum are 20 seconds and 60 seconds, and the corresponding amplitudes are 0.05 μm and 0.03 μm, respectively. This indicates that there are two main components of short-period fluctuations and long-period drifts in the deposition process. The deposition rate function v(t) = 0.017 + 0.005 sin(0.314t) + 0.003 sin(0.105t) μm / s is reconstructed by inverse transformation. A fourth-order polynomial is used for least squares fitting, and the initial coefficient values are {0.017, -0.001, 0.0002, -0.00001, 0.0000002}. After least squares iteration optimization, the modified coefficients are {0.0172, -0.00095, 0.00018, -0.000012, 0.00000018}, and the root mean square error of the fitting curve and the measured data is 0.002 μm / s. When the polynomial order is increased to 5, the error is only reduced to 0.0019 μm / s, and the improvement is not significant, so a fourth-order polynomial is determined as the final deposition rate variation function. The function reveals the nonlinear variation characteristics of the deposition rate with time: the rate rapidly decreases in the initial stage (0-20 seconds), tends to be stable in the middle stage (20-60 seconds), and slowly increases in the later stage (>60 seconds), which is closely related to the evolution of the electrode surface state and the dynamic change of ion concentration distribution. Through this multi-time scale analysis method, both the rapid fluctuation characteristics and the long-term change trend are captured.

[0029] In step S105, the deposition thickness of each abnormal concentration group at a future time is calculated according to the deposition rate variation law diagram, and the copper layer thickness distribution of the abnormal region is obtained. The copper layer thickness distribution of the abnormal region is compared with the thickness distribution of the normal region, and the thickness deviation value of the abnormal region is calculated.

[0030] According to the deposition rate variation law function, a fourth-order Runge-Kutta integral equation is constructed, and an abnormal concentration group cumulative deposition thickness distribution function is obtained by a weighted average method. A radial basis function is used to interpolate and fit the copper layer thickness of the normal region, and a reference thickness distribution function is obtained by three-dimensional grid division. The thickness deviation function is obtained by difference calculation of the copper layer thickness distribution function of the abnormal region and the reference thickness distribution function. The probability density of the thickness deviation function is estimated by using a Gaussian kernel function, and the deviation distribution map is obtained according to the kernel function weight.

[0031] Specifically, a fourth-order Runge-Kutta integral equation is constructed according to the deposition rate variation law function v(t), time step h is set to calculate state quantities k1, k2, k3, and k4, and a weighted average method is used to obtain the cumulative deposition thickness H(x, y, z, t) of each abnormal concentration group. A three-dimensional grid division is used to obtain a set of discrete point coordinates P(x, y, z) of the normal area, a radial basis function is used to interpolate and fit the normal area copper layer thickness H0(x, y, z) to generate a continuous reference thickness distribution function B(x, y, z). The mean value μ and the standard deviation σ are calculated according to the normal area reference thickness distribution function B(x, y, z), and the three standard deviation principle is used to establish the normal area thickness fluctuation range [μ-3σ, μ+3σ] as the reference standard. The difference between the abnormal area copper layer thickness distribution H(x, y, z, t) and the reference thickness distribution function B(x, y, z) is calculated to generate a thickness deviation function D(x, y, z, t) = H(x, y, z, t)-B(x, y, z). The probability density of the thickness deviation value is estimated by using a Gaussian kernel function, the bandwidth parameter w is selected to calculate the kernel function weight, and the deviation value probability density function P(D) is generated. According to the probability density function P(D), the deviation value interval is divided, the spatial distribution characteristics of the deviation points in each interval are counted, and the deviation distribution map G(x, y, z) is generated by using the contour method. In the process of calculating the deposition thickness of the copper layer, the fourth-order Runge-Kutta method is used to numerically integrate the deposition rate function. Taking an actual case as an example, the deposition rate function v(t) = 0.015 + 0.002 sin(0.1t) μm / s is given, the time step h = 0.1 s, then the state quantity calculation formula is: k1 = h-v(t), k2 = h-v(t+h / 2), k3 = h-v(t+h / 2), k4 = h-v(t+h), and the final integral result H = (k1+2k2+2k3+k4) / 6. The calculated cumulative thickness at t = 10s is 0.152 μm. The reference thickness distribution of the normal area is interpolated by using a multi-quadratic radial basis function, and the function form is B(r) = (1+(εr) 2 )^(1 / 2), where r is the spatial distance, and ε is the shape parameter taking 0.1. For typical normal area measurement data, the mean value μ = 0.15 μm and the standard deviation σ = 0.005 μm are calculated, and the reference range [0.135 μm, 0.165 μm] is established. In the thickness deviation calculation, local thickness mutation occurs in the abnormal concentration group. Taking a certain abnormal concentration group as an example, the center position thickness reaches 0.18 μm, which is 0.03 μm different from the reference value, far exceeding the normal fluctuation range. The Gaussian kernel function K(x) = exp(-x 2The probability density estimation is performed, and the kernel function bandwidth parameter h is determined as 0.0075 μm by using the Silverman criterion. The deviation value probability density distribution presents obvious bimodal characteristics, the main peak value is located at 0.02 μm, and the secondary peak value is located at-0.015 μm, indicating that the abnormal area simultaneously exists thickening and thinning phenomena. Spatially, the thickening area is mainly concentrated in the edge of the electric field distortion area, and the maximum deviation reaches 0.035 μm; while the thinning area appears in the center of the fluid disturbance, and the maximum deviation is-0.025 μm. The deviation distribution map is drawn by the contour method, and the contour interval is set as 0.005 μm. The projection in the x-y plane shows that the thickening area presents an elliptical distribution, the long axis is about 5 mm, and the short axis is about 3 mm; the thinning area presents an irregular circle, and the diameter is about 4 mm. There is a transition zone between the two areas, the width is about 1 mm, and the deviation value gradually transitions from-0.01 μm to 0.01 μm. The z-direction profile shows that the deviation value decays exponentially with the depth, and tends to be stable at the surface of 50 μm, reaching the reference thickness level. This three-dimensional distribution characteristic reflects the superposition result of the electric field effect and the fluid action, and provides a quantitative basis for subsequent process optimization.

[0032] In step S106, for the thickness deviation value, a deviation compensation algorithm is used to adjust the electroplating process parameters of the abnormal area according to the size and sign of the deviation value, and the electroplating process parameters include current density and electrolyte flow rate.

[0033] According to the thickness deviation value, a proportional integral compensation equation is established, and the current density compensation coefficient and the flow rate compensation coefficient are obtained by integrating the deviation between the target value and the actual value; a segmented linear feedback function is established for the current density compensation coefficient, and if the absolute value of the compensation coefficient is less than a preset threshold, a linear compensation is used to obtain a modified current density distribution; an exponential nonlinear mapping function is used to transform the flow rate compensation coefficient, and a modified flow rate distribution is obtained by operating a preset mapping coefficient and an original flow rate value; a compensation parameter evaluation function containing a thickness deviation square term, a current density gradient term and a flow rate gradient term is constructed, and an optimized compensation coefficient is obtained by using a deep neural network to perform an optimized operation on the compensation parameter.

[0034] Specifically, a proportional-integral compensation equation is established according to the thickness deviation value Δh(x, y, z): u(t) = Kp·e(t) + Ki∫e(t)dt, wherein u(t) is the output of the controller, Kp and Ki are compensation coefficients, e(t) is the deviation between the target value and the actual value, and t is time. The current density compensation coefficient α(x, y, z) and the flow rate compensation coefficient β(x, y, z) are calculated. A piecewise linear feedback function is established for the current density compensation coefficient α(x, y, z). If |α| is less than a threshold value α0, linear compensation is adopted; if |α| is greater than α0, saturation compensation is adopted, and a modified current density distribution j'(x, y, z) is generated. The flow rate compensation coefficient β(x, y, z) is converted by using an exponential nonlinear mapping function: v' = v·exp(λβ), wherein λ is a mapping coefficient, and a modified flow rate distribution v'(x, y, z) is calculated. A compensation parameter evaluation function is constructed:

[0035] where J(a, b) is the optimized objective value of the compensation parameter, Ah is the thickness deviation, j is the current density distribution, v is the flow rate distribution, w1, w2, w3 are weight coefficients, and the optimized objective value of the compensation parameter is calculated. A five-layer deep neural network is used to optimize the compensation parameter, the input layer receives the deviation value and the original parameter, the hidden layer uses the ReLU activation function, and the output layer is the optimized compensation coefficient. According to the optimized compensation coefficient, the process parameters are recalculated, the compensation effect is verified by numerical simulation, and the compensation parameter verification index R(x, y, z) is generated. In the process of compensating the electroplating process parameters, the deviation compensation is calculated using a proportional-integral algorithm. When the thickness deviation Ah of a certain point is measured to be 5 pm, the proportional coefficient Kp is set to 0.2, the integral coefficient Ki is set to 0.05, and the corresponding compensation output is u(t) = 0.2 x 5 + 0.05 x 5dt pm. The integral term accumulates over time, and the compensation output gradually increases from 1 pm to 2.5 pm within 10 seconds. The current density compensation adopts a piecewise function processing, and the threshold value a0 is set to 0.3. When the compensation coefficient a = 0.2, it is located in the linear interval, at this time the corrected current density j' = j x (1 + 0.2); when a = 0.4, it exceeds the threshold value and enters the saturation zone, at this time the saturation value j' = j x 1.3 is taken. This piecewise processing method avoids overcompensation. The flow rate compensation adopts an exponential mapping relationship, and the mapping coefficient l = 0.5. When b = 0.4, the corrected flow rate v' = v x exp(0.5 x 0.4) = 1.22v, showing a nonlinear growth characteristic. This exponential relationship responds gently in the small compensation coefficient area and responds rapidly in the large compensation coefficient area. The weight coefficients in the evaluation function reflect the importance of different optimization objectives, and the typical values are w1 = 0.5, w2 = 0.3, and w3 = 0.2. Taking an abnormal area as an example, the evaluation value J before compensation is 2.5, of which the thickness deviation contributes 1.5, the current gradient contributes 0.6, and the flow rate gradient contributes 0.4. The deep neural network adopts a five-layer structure, the input layer of 8 neurons receives the deviation value and the original parameter, the three hidden layers contain 32, 16, and 8 neurons respectively, and the output layer of 2 neurons outputs the optimized compensation coefficient. The ReLU activation function f(x) = max(0, x) is used to process the hidden layer output to avoid the gradient vanishing problem. In the verification stage, the effect of the compensation parameter is evaluated by the index R(x, y, z). The index R of a test point before compensation is 0.85, and after one round of optimization, it is improved to R = 0.92. Analysis of the optimization process shows that the current density compensation dominates the improvement of thickness uniformity, while the flow rate compensation significantly reduces local disturbance. The synergistic effect of the two compensation mechanisms is reflected in multiple components of the evaluation function: the thickness deviation square sum is reduced by 45%, the current gradient square sum is reduced by 35%, and the flow rate gradient square sum is reduced by 30%. This multi-objective optimization strategy not only improves the deposition uniformity, but also maintains the smooth distribution of process parameters.

[0036] In step S107, the power system and the electrolyte circulation system of the composite copper foil thickness control equipment are parameter set and adjusted according to the optimized electroplating process parameters, so that the composite copper foil thickness control equipment produces according to the optimized electroplating process parameters.

[0037] The target region anode plate theoretical position coordinates are calculated according to the current density distribution function, the anode plate actual position is measured through the spatial distance sensor array, and the position deviation compensation amount is obtained by using a proportional integral controller; the laser displacement sensor array is used to obtain a distance change rate matrix between the anode plate and the cathode plate, the anode plate inclination angle is calculated according to the distance change rate matrix, and an angle adjustment signal is output through an angle closed-loop controller; if the current density uniformity index is less than a preset threshold value, the anode plate position compensation mechanism is driven to adjust until the uniformity index is greater than the preset threshold value; the electrolyte flow rate set value is calculated according to the flow field distribution function, the actual flow rate value is measured through a flow sensor, the valve opening is adjusted by using a pressure closed-loop controller, and the liquid inlet pipeline and the liquid return pipeline flow rate ratio is adjusted according to the valve opening.

[0038] Specifically, the target region anode plate theoretical position coordinates (x0, y0, z0) are calculated according to the optimized current density distribution function j(x, y, z), the anode plate actual position (x, y, z) is measured by the spatial distance sensor array, and the position deviation compensation amount Ar is calculated by using the proportional integral controller. The laser displacement sensor array is used to measure the distance change rate between the anode plate and the cathode plate in real time, and the optimal anode plate inclination angle θ(x, y) is calculated according to the distance change rate matrix D(x, y, z), and the adjustment signal is output by the angle closed-loop controller. The anode plate fine tuning parameters are calculated according to the current density uniformity index, and if the uniformity index is less than the preset threshold η0, the fine position compensation mechanism is activated for fine adjustment until the uniformity index is greater than the threshold η0. The optimized flow field distribution function v(x, y, z) is used to calculate the electrolyte flow rate set value v0, the actual flow rate v is measured by the flow sensor, and the circulating pump speed n is adjusted according to the deviation Δv=v0-v. The electrolyte pipeline pressure p(t) is collected in real time, the pressure signal is filtered by a third-order Butterworth low-pass filter, and the valve opening compensation amount φ is calculated by the pressure closed-loop controller. The flow ratio γ of the liquid inlet pipeline and the liquid return pipeline is adjusted according to the valve opening compensation amount φ, and the liquid level h(t) is monitored by the liquid level sensor to establish a liquid level stable control loop. During the adjustment of the parameters of the electroplating equipment, the anode plate position control adopts a multi-stage closed-loop feedback mechanism. When the theoretically calculated optimal position coordinates of the anode plate are (150mm, 200mm, 80mm), the actual position measured by the spatial distance sensor is (152mm, 198mm, 82mm), and the position deviation vector Ar=(2, -2, 2)mm. By using the proportional integral controller, the proportional coefficient Kp=0.8 and the integral coefficient Ki=0.2 are set, and the position compensation amount is calculated to drive the servo motor for fine tuning. In the anode plate inclination angle adjustment, the laser displacement sensor array is arranged around the cathode plate, and the sampling frequency is 100Hz. When the measured distance change rate matrix shows that the right side distance increase rate is 0.5mm / s and the left side distance decrease rate is -0.4mm / s, it indicates that the anode plate has a tendency to tilt to the right. The optimal inclination angle θ=2.5 degrees is obtained by matrix operation, and the corresponding step motor pulse signal is output by the angle controller. The current density uniformity index η is calculated by the ratio of the standard deviation to the average value, and the preset threshold η0=0.05. The measured data shows that the uniformity index η=0.08 of a certain region exceeds the threshold range. At this time, the fine position compensation mechanism is activated, and the fine tuning is performed with a step size of 0.1mm. After 3 iterations, the uniformity index decreases to 0.045. In terms of flow field control, the optimized flow rate distribution function gives the set value v0=0.15m / s, the actual flow rate v=0.12m / s is measured by the flow sensor. The circulating pump speed increment Δn=300rpm is calculated by the PI controller, and the frequency converter adjusts the output frequency accordingly. At the same time, the pressure sensor monitors the pipeline pressure fluctuation, which contains 50Hz power frequency interference.The pressure signal is processed by a third-order Butterworth filter with a cutoff frequency of 10 Hz, and the stable pressure reading p = 0.2 MPa is obtained after filtering out the high-frequency noise. The liquid level control adopts a double-pipe balance strategy, and the flow ratio γ of the inlet pipe and the return pipe is realized by adjusting the valve opening. When the inlet valve opening is 60%, the return valve opening should be adjusted to 55% to maintain the liquid level at the target height h0= 500 mm. If the liquid level fluctuates ±10 mm, the liquid level controller will automatically adjust the valve opening, and the compensation amount φ is determined by the deviation value and the change rate. This multi-parameter coupling control scheme realizes the dynamic balance of the process parameters, and each control loop cooperates with each other to maintain the stable operation of the electroplating process.

[0039] The preferred embodiments of the application disclosed above are only used to help explain the application. The preferred embodiments do not describe all the details and limit the application to the specific embodiments. Obviously, many modifications and variations can be made according to the content of the specification. The specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the application, so that those skilled in the art can well understand and utilize the application. The application is limited only by the claims and their full scope and equivalents.

Claims

1. A control method of a composite copper foil thickness control apparatus, characterized by, The method comprises: According to the mutual coupling relationship of the electric field, flow field and mass transfer process of the electroplating process, a multi-physical field coupling model describing the production process of the composite copper foil is constructed, the interface layer morphology and thickness parameters of the composite copper foil are integrated into the multi-physical field coupling model, and the electric field distribution, electrolyte flow state and copper ion concentration distribution data on different shaped interface layers are obtained through simulation of the multi-physical field coupling model; For the electric field distribution data obtained by simulation, it is judged whether there is a local distortion phenomenon of the electric field, if there is a local distortion of the electric field, the position coordinates of the distortion region are determined, and the electric field data of the distortion region are marked as abnormal data, at the same time, according to the electrolyte flow state data, it is judged whether there is a flow disturbance phenomenon, if there is a disturbance, the position coordinates of the disturbance region are determined, and the flow field data of the disturbance region are marked as abnormal data; According to the copper ion concentration distribution data, the copper ion concentration is grouped to obtain copper ion groups of different concentration levels, and the center coordinates of the copper ion groups of different concentration levels are calculated, according to the position coordinates of the distortion region and the disturbance region, it is judged whether the center coordinates of the copper ion groups of different concentration levels are located in the abnormal region, if located in the abnormal region, the copper ion group of the concentration level is marked as an abnormal concentration group; For the abnormal concentration group, a time series prediction algorithm is used to predict the concentration change at future time according to the historical copper ion concentration change trend, and the concentration change curve of each abnormal concentration group is obtained, and at the same time, according to the concentration change curve, a curve fitting algorithm is used to obtain the deposition rate change rule diagram of each abnormal concentration group; According to the deposition rate change rule diagram, the deposition thickness of each abnormal concentration group at future time is calculated to obtain the copper layer thickness distribution of the abnormal region, and the copper layer thickness distribution of the abnormal region is compared with the thickness distribution of the normal region to calculate the thickness deviation value of the abnormal region; For the thickness deviation value, a deviation compensation algorithm is used to adjust the electroplating process parameters of the abnormal region according to the size and sign of the deviation value, the electroplating process parameters including current density and electrolyte flow rate; According to the optimized electroplating process parameters, the power system and electrolyte circulation system of the composite copper foil thickness control equipment are parameter set and adjusted, so that the composite copper foil thickness control equipment produces according to the optimized electroplating process parameters.

2. The method of claim 1, wherein, According to the mutual coupling relationship of the electric field, flow field and mass transfer process of the electroplating process, a multi-physical field coupling model describing the production process of the composite copper foil is constructed, the interface layer morphology and thickness parameters of the composite copper foil are integrated into the multi-physical field coupling model, and the electric field distribution, electrolyte flow state and copper ion concentration distribution data on different shaped interface layers are obtained through simulation of the multi-physical field coupling model, including: The quadrilateral grid elements of the composite copper foil substrate surface are divided by using the finite element numerical calculation method, and the interface layer thickness distribution function is calculated according to the three-dimensional coordinates of the grid element boundary points; According to the three-dimensional geometric structure digital model of the composite copper foil established by the interface layer thickness distribution function, the finite volume method is used to discretely solve the Navier-Stokes equation to obtain the electrolyte velocity field and pressure field distribution data near the interface layer. According to the electrolyte velocity field and pressure field distribution data, the potential distribution is obtained by solving Poisson equation by using finite element method, and the current density at the interface layer is calculated by combining the electrolyte conductivity distribution function; For the current density at the interface layer, the copper ion concentration control equation is established, combined with the velocity field transport term and the diffusion coefficient diffusion term, and the ion concentration distribution data is obtained by solving the mass transfer equation by using finite difference method.

3. The method of claim 1, wherein, According to the electric field distribution data obtained by simulation, it is judged whether there is local distortion phenomenon of electric field, if there is local distortion of electric field, the position coordinates of distortion region are determined, and the electric field data of distortion region are marked as abnormal data, at the same time, according to the flow state data of electrolyte, it is judged whether there is flow disturbance phenomenon, if there is disturbance, the position coordinates of disturbance region are determined, and the flow field data of disturbance region are marked as abnormal data, including: According to the electric field distribution data of grid points, the electric field intensity gradient value is calculated, if the gradient value exceeds three times of the standard deviation of local average value, the boundary point coordinate set of distortion region is obtained by using three-dimensional space clustering method; The velocity gradient tensor and vorticity field are calculated from the grid point velocity vector field, if the vorticity value is greater than twice of the laminar reference value, the boundary point coordinate set of disturbance region is obtained by using boundary tracking algorithm; For the boundary point coordinate set of the distortion region and the boundary point coordinate set of the disturbance region, the volume fraction of the overlapping region is obtained by grid point matching, if the volume fraction is greater than the critical value, it is judged that the region is a strong coupling abnormal region; The temperature field distribution of strong coupling abnormal region is obtained by using thermocouple array, and the local conductivity is obtained according to the temperature field distribution and the temperature coefficient of conductivity.

4. The method of claim 1, wherein, According to the copper ion concentration distribution data, the copper ion concentration is divided into different concentration levels, and the center coordinates of each concentration level of copper ion group are calculated, according to the position coordinates of the distortion region and the disturbance region, it is judged whether the center coordinates of each concentration level of copper ion group are located in the abnormal region, if located in the abnormal region, the copper ion group of this concentration level is marked as abnormal concentration group, including: According to the copper ion concentration distribution data, a three-dimensional concentration field is constructed, and the three-dimensional concentration field is obtained by density clustering algorithm to obtain three concentration level regions; For the concentration level region, the centroid coordinates and point number density are obtained, and the centroid coordinates and point number density are obtained by Gaussian kernel function to obtain group space distribution function and group characteristic radius; The boundary point set of electric field distortion region and fluid disturbance region is used to construct abnormal region space envelope surface, and the shortest distance value is obtained by the abnormal region space envelope surface and the concentration group centroid; According to the ratio of the shortest distance value to the group characteristic radius, it is judged whether it is less than the critical value, if the ratio is less than the critical value, the concentration group is marked as abnormal group and the abnormal group space distribution feature matrix is generated.

5. The method of claim 1, wherein, For the abnormal concentration group, a time series prediction algorithm is used to predict the concentration change at future time according to the historical time copper ion concentration change trend, and the concentration change curve of each abnormal concentration group is obtained, at the same time, according to the concentration change curve, a curve fitting algorithm is used to obtain the deposition rate change rule diagram of each abnormal concentration group, including: Resample the historical data of the abnormal concentration group with a fixed sampling interval to obtain equidistant time series data and concentration change rate sequence; Train a long short-term memory network model using the concentration change rate sequence, and output the next time concentration prediction value through the network model; Calculate the concentration difference value of adjacent time for the concentration prediction value, and calculate the theoretical deposition thickness increment sequence according to the concentration difference value and Faraday's law; Perform Fourier transform on the theoretical deposition thickness increment sequence to extract frequency components, and perform polynomial fitting on the frequency components to obtain a deposition rate variation function, and if the root mean square error of the polynomial fitting is greater than a preset threshold, increase the polynomial order and repeat the fitting process.

6. The method of claim 1, wherein, Calculate the deposition thickness of each abnormal concentration group at future time according to the deposition rate variation function, obtain the copper layer thickness distribution of the abnormal area, and compare the copper layer thickness distribution of the abnormal area with the thickness distribution of the normal area to calculate the thickness deviation value of the abnormal area, including: Construct a fourth-order Runge-Kutta integral equation according to the deposition rate variation function, and obtain the cumulative deposition thickness distribution function of the abnormal concentration group by weighted average method; Interpolate and fit the copper layer thickness of the normal area using radial basis function, and obtain the reference thickness distribution function by three-dimensional grid division; Calculate the thickness deviation function by difference calculation between the copper layer thickness distribution function of the abnormal area and the reference thickness distribution function; Estimate the probability density of the thickness deviation function using Gaussian kernel function, and obtain the deviation distribution map according to the kernel function weight.

7. The method of claim 1, wherein, For the thickness deviation value, use a deviation compensation algorithm to adjust the electroplating process parameters of the abnormal area according to the size and sign of the deviation value, including current density and electrolyte flow rate, including: Establish a proportional-integral compensation equation according to the thickness deviation value, and obtain the current density compensation coefficient and flow rate compensation coefficient by integrating the deviation between the target value and the actual value; Establish a segmented linear feedback function for the current density compensation coefficient, and if the absolute value of the compensation coefficient is less than a preset threshold, use linear compensation to obtain the modified current density distribution; Transform the flow rate compensation coefficient using an exponential nonlinear mapping function, and obtain the modified flow rate distribution by operating the preset mapping coefficient and the original flow rate value; Construct a compensation parameter evaluation function containing thickness deviation square term, current density gradient term and flow rate gradient term, and optimize the compensation parameter using deep neural network to obtain the optimized compensation coefficient.

8. The method of claim 7, wherein, The proportional-integral compensation equation is: u(t) = Kp·e(t) + Ki∫e(t)dt Where u(t) is the output of the controller, Kp and Ki are compensation coefficients, e(t) is the deviation between the target value and the actual value, and t is the time.

9. The method of claim 7, wherein, The compensation parameter evaluation function is: Where J(α,β) is the optimization target value of the compensation parameter, Δh is the thickness deviation, j is the current density distribution, v is the flow rate distribution, and w1, w2 and w3 are weight coefficients.

10. The method of claim 1, wherein, The power supply system and the electrolyte circulation system of the composite copper foil thickness control equipment are set and adjusted according to the optimized electroplating process parameters, so that the composite copper foil thickness control equipment produces according to the optimized electroplating process parameters, comprising: According to the current density distribution function, the theoretical position coordinates of the target area anode plate are calculated, the actual position of the anode plate is measured through the spatial distance sensor array, and the position deviation compensation amount is obtained by using the proportional integral controller; The laser displacement sensor array is used to obtain the distance change rate matrix between the anode plate and the cathode plate, the anode plate inclination angle is calculated according to the distance change rate matrix, and the angle adjustment signal is output through the angle closed-loop controller; If the current density uniformity index is less than the preset threshold value, the anode plate position compensation mechanism is driven to adjust until the uniformity index is greater than the preset threshold value; According to the flow field distribution function, the electrolyte flow rate set value is calculated, the actual flow rate value is measured through the flow sensor, the valve opening is adjusted by using the pressure closed-loop controller, and the flow rate ratio of the liquid inlet pipeline and the liquid return pipeline is adjusted according to the valve opening.

Images