Electroplating bath production operation and maintenance big data analysis and optimization system and method

Abstract

The invention discloses an electroplating bath production operation and maintenance big data analysis and optimization system and method, and the method comprises the steps: building an operation and maintenance state mapping table through electroplating bath process parameter data collection and key control node sequence determination; generating a process probability distribution tree by adopting a probability backtracking processing technology, obtaining a quality achievement probability value, determining a main process family and an alternative process family, and constructing a probability optimization chain; a stirring flow velocity distribution coefficient and a plating solution circulation complexity index are extracted from the probability optimization chain, mass transfer efficiency distribution is analyzed and predicted through an eddy current effect, and process optimization configuration is generated; identifying a high-risk process section and a stable process section by using a risk energy release analysis technology, and forming a mass balance process pair through hedging combination; hot spot distribution is determined by combining temperature field analysis, an area with an overlarge temperature gradient is identified, and a heat dissipation optimization factor is extracted; and finally, generating a dynamic production unit and an elastic production process by matching a beat control parameter with a probability optimization chain, so as to realize big data analysis and optimization of production operation and maintenance of the electroplating bath.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of electroplating tank production operation and maintenance management, in particular to an electroplating tank production operation and maintenance big data analysis and optimization system and method. BACKGROUND

[0002] As a key equipment in the field of precision manufacturing, the production operation and maintenance management level of electroplating tank directly affects the product quality stability and production efficiency. Traditional electroplating tank operation and maintenance management mainly relies on manual experience and simple data recording, lacking systematic data analysis and intelligent decision support.

[0003] The existing electroplating tank operation and maintenance management has many technical bottlenecks. The current data collection means is relatively single, mainly focusing on basic current, voltage and other parameter monitoring, lacking in-depth data mining and correlation analysis of complex physical and chemical changes in the process. The production control strategy mainly uses fixed mode parameter setting, which is difficult to adapt to the actual production situation, leading to resource waste and quality instability problems. At the same time, the traditional fault diagnosis and maintenance management mainly based on periodic maintenance and post-fault processing, lacking predictive maintenance capability, unable to effectively prevent production interruption and equipment damage. In addition, the existing system has obvious shortcomings in multi-source heterogeneous data fusion, production process optimization and real-time decision support, which is difficult to meet the requirements of modern intelligent manufacturing for fine management and flexible production. SUMMARY

[0004] The present application discloses an electroplating tank production operation and maintenance big data analysis and optimization system and method, aiming to realize comprehensive digital management of electroplating tank production operation and maintenance through multi-dimensional data collection, intelligent analysis and processing and optimization decision. The system builds a process parameter monitoring network and a state mapping mechanism, combines probability analysis and risk assessment technology to establish an intelligent control system of production process, and uses big data analysis technology to realize continuous optimization of production process and intelligent support of operation and maintenance decision.

[0005] The present application discloses an electroplating tank production operation and maintenance big data analysis and optimization system and method, aiming to realize comprehensive digital management of electroplating tank production operation and maintenance through multi-dimensional data collection, intelligent analysis and processing and optimization decision. The system builds a process parameter monitoring network and a state mapping mechanism, combines probability analysis and risk assessment technology to establish an intelligent control system of production process, and uses big data analysis technology to realize continuous optimization of production process and intelligent support of operation and maintenance decision.

[0006] Acquire electroplating tank process parameter data, determine key control node sequence according to the process parameter data, and establish operation and maintenance state mapping table according to the key control node sequence;

[0007] Perform probability backtracking processing on the operation and maintenance state mapping table to generate a process probability distribution tree, perform branch analysis on the process probability distribution tree to obtain a quality achievement probability value, determine a main process family and an alternative process family using the quality achievement probability value, and generate a probability optimization chain according to the main process family and the alternative process family;

[0008] obtain a stirring flow rate distribution coefficient and a plating solution circulation complexity index from the probability optimization chain, predict a mass transfer efficiency distribution through the stirring flow rate distribution coefficient, determine a flow field regulation rhythm using the plating solution circulation complexity index, and generate a process optimization configuration in combination with the mass transfer efficiency distribution and the flow field regulation rhythm;

[0009] determine a high-risk process section and a stable process section using the probability optimization chain, hedge and combine the high-risk process section and the stable process section to form a quality balance process pair, and generate a process coordination rule according to the quality balance process pair;

[0010] determine a hot spot distribution through temperature field analysis according to the process coordination rule and the process optimization configuration, identify a temperature gradient excessive area through the hot spot distribution, extract a heat dissipation optimization factor from the temperature gradient excessive area, and generate a rhythm control parameter using the heat dissipation optimization factor;

[0011] generate a dynamic production unit by matching the rhythm control parameter with the probability optimization chain, generate an elastic production process using the dynamic production unit, and complete electroplating tank production operation and maintenance big data analysis and optimization.

[0012] The second aspect of the present application proposes an electroplating tank production operation and maintenance big data analysis and optimization system, comprising:

[0013] a data acquisition module configured to obtain electroplating tank process parameter data, determine a key control node sequence according to the process parameter data, and establish an operation and maintenance state mapping table according to the key control node sequence;

[0014] a probability analysis module configured to perform probability backtracking processing on the operation and maintenance state mapping table to generate a process probability distribution tree, perform branch analysis on the process probability distribution tree to obtain a quality achievement probability value, determine a main process family and an alternative process family using the quality achievement probability value, and generate a probability optimization chain according to the main process family and the alternative process family;

[0015] a flow field optimization module configured to obtain a stirring flow rate distribution coefficient and a plating solution circulation complexity index from the probability optimization chain, predict a mass transfer efficiency distribution through the stirring flow rate distribution coefficient, determine a flow field regulation rhythm using the plating solution circulation complexity index, and generate a process optimization configuration in combination with the mass transfer efficiency distribution and the flow field regulation rhythm;

[0016] a risk balance module configured to determine a high-risk process section and a stable process section using the probability optimization chain, hedge and combine the high-risk process section and the stable process section to form a quality balance process pair, and generate a process coordination rule according to the quality balance process pair;

[0017] A temperature control optimization module is configured to determine a hot spot distribution by performing temperature field analysis according to the process coordination rules and the process optimization configuration, identify a temperature gradient excessive area by using the hot spot distribution, extract a heat dissipation optimization factor from the temperature gradient excessive area, and generate a beat control parameter by using the heat dissipation optimization factor.

[0018] A dynamic execution module is configured to generate a dynamic production unit by matching the beat control parameter with the probability optimization chain, generate an elastic production flow by using the dynamic production unit, and complete the electroplating tank production operation and maintenance big data analysis and optimization.

[0019] The beneficial effects of the present application are embodied in the following aspects: first, by establishing a process parameter data acquisition and key control node identification mechanism, combined with the construction of the operation and maintenance state mapping table, the overall monitoring and data management of the electroplating tank operation state are realized, the traditional manual inspection mode is changed into an automatic real-time monitoring system, the accuracy and timeliness of operation and maintenance management are improved, and a complete data tracing and state analysis capability is established. Secondly, the probability analysis and risk assessment technology is used to construct the process probability distribution tree and the probability optimization chain, the identification and balancing mechanism of high-risk process section and stable process section is established, the prediction control of production risk and the stable guarantee of process quality are realized through the quantitative evaluation of quality achievement probability and the formulation of process coordination rules, the reliability and consistency of electroplating production are improved. Finally, through the construction of temperature field analysis, mass transfer efficiency prediction and dynamic production unit, combined with the design of elastic production flow, the flexible scheduling and intelligent optimization of the electroplating tank production process are realized, a continuous improvement mechanism based on big data analysis is established, the production efficiency and resource utilization rate are improved, and the rapid response capability of the production system to market demand changes and abnormal conditions is enhanced.

[0020] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. BRIEF DESCRIPTION OF DRAWINGS

[0021] The drawings herein show specific examples of the technical solutions described in the present application, and constitute part of the specification together with the specific embodiments, for explaining the technical solutions, principles and effects of the present application.

[0022] Unless specifically stated, the same reference signs in different drawings represent the same or similar technical features, and different reference signs may also be used to represent the same or similar technical features.

[0023] Figure 1 is a flowchart of a method for electroplating tank production operation and maintenance big data analysis and optimization.

[0024] Figure 2It is a structural block diagram of a plating tank production operation and maintenance big data analysis and optimization system. DETAILED DESCRIPTION

[0025] In the following description, for purposes of explanation and not limitation, specific details are set forth such as particular architectures, techniques, etc. in order to provide a thorough understanding of the embodiments of the present application. However, it will be apparent to those skilled in the art that the present application can be practiced in other embodiments that depart from these specific details. In other instances, detailed descriptions of well-known methods, devices, and circuits are omitted so as not to obscure the description of the present application with unnecessary detail.

[0026] It is to be understood that the terminology“including,”“comprising,”“consisting of,” and the like as used herein is meant to be open-ended. For example, where the specification states that a process, method, article, or apparatus includes, or consists of, certain components, it is meant that at least those components are present, but it does not exclude the presence of other components.

[0027] Reference throughout this specification to“one embodiment” or“an embodiment” or“some embodiments” means that a particular feature, structure, or characteristic described in connection with the embodiment is included in at least one embodiment of the application. Thus, the appearances of the phrases“in one embodiment” or“in some embodiments” or“in other embodiments” or“in other alternate embodiments” or the like in various places throughout this specification are not necessarily referring to the same embodiment, unless otherwise noted. Furthermore, the terms“comprising,”“including,”“containing,” and“having,” and the like, as used herein, are meant to be equivalent in that they encompass the presence of stated features, integers, steps, operations, elements, and / or components, but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0028] The technical solutions of the embodiments of the present application are introduced as follows.

[0029] As shown in the figure, the embodiments of the present application provide a plating tank production operation and maintenance big data analysis and optimization method, including the following steps S110-S160: Figure 1

[0030] Step S110, acquire plating tank process parameter data, determine the key control node sequence according to the process parameter data, and establish the operation and maintenance state mapping table according to the key control node sequence.

[0031] ​Specifically, the electroplating tank process parameter data is acquired. Through the multi-parameter sensor array arranged at the key positions of the electroplating tank, the process parameter change information in the electroplating process is collected in real time. The sensor system includes a current density sensor, a voltage monitor, a temperature sensor, a pH detector and an electrolyte flowmeter, covering the anode area, the cathode area, the electrolyte circulation area and the temperature control area of the electroplating tank. The current density sensor adopts the Hall effect principle, the measurement range covers 0-500 A / dm2, the accuracy reaches ±0.1 A / dm2, and 25 measuring points are installed on the anode surface according to a 5x5 grid layout. The voltage monitoring adopts a high-precision digital multimeter to record the changes of tank voltage, anode potential and cathode potential in real time. The temperature sensor uses a Pt100 platinum resistor, which is distributed in the upper, middle and lower three layers of the electrolyte to monitor the temperature distribution uniformity. The pH detector adopts the glass electrode method, equipped with automatic calibration and temperature compensation functions. The electrolyte flow is measured by an electromagnetic flowmeter to monitor the flow stability of the circulation system. All sensor data are transmitted to the central control system through the field bus network, and the acquisition frequency is set to 1 Hz to ensure the integrity and timeliness of the process parameter data.

[0032] In some embodiments, the key control node sequence is determined according to the process parameter data, including: acquiring a current density map by measuring the current distribution on the anode surface using the process parameter data; determining the current distribution deviation by evaluating the uniformity of the current density map; generating a distribution compensation coefficient by analyzing the influence of the node using the current distribution deviation; and determining the key control node sequence by optimization using the distribution compensation coefficient.

[0033] The current density map is acquired by measuring the current distribution on the anode surface using the process parameter data. Based on the acquired electroplating tank process parameter data, the current and voltage measurement information is analyzed, and the current distribution model on the anode surface is constructed by numerical calculation method. The current density sensor measurement value in the process parameter data is arranged in two dimensions according to the spatial coordinates of the anode surface to form a discrete current density matrix. The spatial interpolation of the current density between the discrete measuring points is performed using the bicubic spline interpolation algorithm to generate a continuous current density distribution function I(x,y)=Σᵢⱼaᵢⱼx i y j , where I is the current density, x and y are the coordinates of the anode surface, and aᵢⱼ is the interpolation coefficient. The environmental influence factors in the current density calculation are corrected in combination with the electrolyte conductivity and temperature distribution information in the process parameter data. The continuous current density distribution calculated is visualized in the form of color contour map to construct a complete current density map. The map uses pseudo-color coding, and the color depth represents the current density size, and the contour interval is determined according to the accuracy requirement of the process parameter data.

[0034] The uniformity of the current density spectrum is assessed to determine the deviation in current distribution. Based on the constructed current density spectrum, statistical analysis methods are used to quantify the uniformity of the current distribution on the anode surface. The average current density μ = (1 / n)Σ at all measuring points in the current density spectrum is calculated. i=1 n Iᵢ, where n is the total number of measuring points, and Iᵢ is the current density at the i-th measuring point. The standard deviation of the current density is calculated as σ = √[(1 / n)Σ i=1 n [(Iᵢ-μ)²], where standard deviation reflects the dispersion of values ​​in the current density spectrum. A uniformity coefficient U = 1 - σ / μ is defined, ranging from 0 to 1; a value closer to 1 indicates a more uniform distribution of the current density spectrum. Skewness and kurtosis parameters of the current density spectrum are calculated to analyze the symmetry and sharpness characteristics of the distribution pattern. The coefficient of variation CV = σ / μ is used to quantify the relative variability of the current density spectrum; a smaller coefficient of variation indicates a more uniform distribution. A spatial analysis model of current distribution deviation is established, dividing the current density spectrum into several regions, and calculating the local deviation for each region. Regional deviation is defined as the relative difference between the local average current density and the global average; a positive deviation indicates a higher current density in that region, and a negative deviation indicates a lower current density.

[0035] Nodal influence analysis is performed using current distribution deviation to generate distribution compensation coefficients. Based on the determined current distribution deviation data, the influence of each location on the anode surface on the overall current distribution uniformity is analyzed. A nodal influence model is established, dividing the anode surface into several control nodes, each corresponding to a specific spatial region. The influence weight of each node is calculated as Wᵢ=|Dᵢ| / Σⱼ|Dⱼ|, where Dᵢ is the current distribution deviation of the i-th node, and the weight reflects the node's contribution to the global distribution non-uniformity. The mutual influence relationships between nodes are analyzed, and an influence transmission matrix is ​​established, where the matrix elements represent the influence intensity of a change in one node on other nodes. Sensitivity analysis is used to assess the influence of changes in node parameters on the current distribution deviation, identifying key and secondary influencing nodes. Based on the nodal influence analysis results, a distribution compensation strategy is designed, assigning a corresponding compensation coefficient to each node. The compensation coefficient Cᵢ=-αDᵢ / |Dᵢ,max|, where α is the compensation intensity parameter, Dᵢ,max is the maximum deviation value, and the negative sign indicates that the compensation direction is opposite to the deviation direction.

[0036] The key control node sequence was determined through distribution compensation coefficient optimization. The optimization objectives included maximizing current distribution uniformity, minimizing control cost, and maximizing system stability. All candidate nodes were sorted according to the absolute value of their distribution compensation coefficients, with nodes having higher absolute values ​​having higher priority. Constraints were established for node selection, including node quantity limits, spatial distribution requirements, and control capability constraints. A genetic algorithm was used to solve the node sequence optimization problem, with chromosome encoding representing node selection and sorting. The fitness function comprehensively considered the uniformity improvement effect and implementation complexity. The distribution compensation coefficient was introduced as a weighting factor for node importance during the optimization process to ensure that selected nodes had the greatest compensation potential. Through multiple generations of evolution and selection operations, the node combination was gradually optimized until the optimal key control node sequence was found. The final determined key control node sequence was arranged in priority order, with the node positions corresponding to specific coordinates on the anode surface. Each node was equipped with a corresponding distribution compensation coefficient and control strategy.

[0037] An operation and maintenance status mapping table is established based on the sequence of key control nodes. A two-dimensional matrix structure is used, with rows representing the node number and position in the sequence, and columns representing the corresponding operation and maintenance status category and parameter thresholds. Status categories cover four levels: normal operation, early warning, abnormal, and fault. Normal operation corresponds to all node parameters in the sequence being within design limits. Early warning indicates that some node parameters in the sequence are approaching threshold boundaries. Abnormal status reflects that the parameters of major nodes in the sequence exceed normal ranges. Fault status represents a serious deviation in a key node of the sequence. The mapping table establishes a quantitative relationship between node parameters and status levels, and sets triggering conditions and response mechanisms for status transitions. Through parameter monitoring of the key control node sequence, the current operation and maintenance status is determined in real time, and corresponding control strategies are triggered.

[0038] Step S120: Perform probabilistic backtracking processing on the operation and maintenance status mapping table to generate a process probability distribution tree, perform branch analysis on the process probability distribution tree to obtain the quality achievement probability value, use the quality achievement probability value to determine the main process family and alternative process families, and generate a probabilistic optimization chain based on the main process family and alternative process families.

[0039] Specifically, a probabilistic backtracking process is applied to the operation and maintenance state mapping table to generate a process probability distribution tree. The state transition records in the operation and maintenance state mapping table are arranged according to time series to identify the causal relationships and transition probabilities. The probabilistic backtracking process starts from the current state and traces back to all preceding state paths that could have led to that state, calculating the probability of each path occurring P(Sᵢ|S0)=∏ⱼP(S j+1 |Sⱼ), where Sᵢ is the target state, S0 is the initial state, and P(S j+1|Sⱼ) represents the state transition probability. A state transition probability matrix is ​​established by statistically analyzing the frequency of each state transition using historical data from the operation and maintenance state mapping table. Bayesian inference is used to process the uncertainty information in the operation and maintenance state mapping table, combining prior probabilities with observed data. During probability backtracking, a tree structure is constructed to represent the evolution path of the process state. The root node represents the initial process state, branch nodes represent intermediate states, and leaf nodes represent the final process result. Each tree node records the corresponding combination of process parameters and state probability, and the connections between nodes represent the state transition path and transition probability.

[0040] In some embodiments, branch analysis is performed on the process probability distribution tree to obtain the quality achievement probability value, including: extracting probability jump nodes from the process probability distribution tree; performing chain reaction analysis on the probability jump nodes to form a probability propagation chain; performing probability attenuation compensation processing along the probability propagation chain to generate a compensation probability; and reconstructing the quality achievement probability value using the compensation probability.

[0041] Probability jump nodes are extracted from the process probability distribution tree. Based on the generated process probability distribution tree, key nodes with abrupt probability changes are identified by analyzing the probability value change characteristics of each node. In the branching structure of the process probability distribution tree, the probability difference ΔP = |P(N) between adjacent nodes is calculated. i+1 )-P(N i )|, where P(N) i Let be the probability value of the i-th node. When the probability difference exceeds a set threshold, the corresponding node is marked as a probability jump node. An adaptive threshold method is used to identify probability jump nodes, with the threshold dynamically adjusted based on the overall distribution characteristics of probability values ​​in the process probability distribution tree. The spatial distribution pattern of probability jump nodes in the process probability distribution tree is analyzed to identify the hierarchical position and branching relationship of jump nodes in the tree structure. The jump amplitude and jump direction of probability jump nodes are calculated; a positive jump indicates an increase in probability, and a negative jump indicates a decrease in probability. A feature vector for probability jump nodes is established, containing attributes such as node position, jump amplitude, jump direction, and influence range. Cluster analysis is used to classify probability jump nodes in the process probability distribution tree, identifying different types of jump patterns and their causal mechanisms. The process parameter combinations and operating conditions corresponding to each probability jump node are recorded, establishing the correlation between jump nodes and process factors. The extracted set of probability jump nodes constitutes the key control points in the process probability distribution tree.

[0042] Chain reaction analysis is performed on probabilistic hop nodes to form a probabilistic propagation chain. A connectivity matrix is ​​established between probabilistic hop nodes, with matrix elements representing whether there are direct or indirect influence relationships between nodes. Graph theory algorithms are used to search for paths connecting different probabilistic hop nodes in the process probability distribution tree, identifying possible paths for probability propagation. The time delay effect of probability propagation is analyzed, and the probability propagation time from the source node to the target node is calculated as τ = Σᵢtᵢ, where tᵢ is the propagation time of the i-th path segment. A probability propagation strength model is established, where propagation strength is related to the distance and connection strength between nodes. Through the influence analysis of probabilistic hop nodes, the main probability propagation sources and directions are determined. The topology of the probability propagation chain is constructed, consisting of multiple probabilistic hop nodes connected in series according to their influence relationships. In the probability propagation chain, the probability change of each node affects the probability distribution of downstream nodes, forming a cascading effect. The generated probability propagation chain network reveals the complex relationships between probabilistic hop nodes in the process probability distribution tree.

[0043] Compensated probabilities are generated by performing probability attenuation compensation along the probability propagation chain. Based on the formed probability propagation chain, the attenuation law and loss mechanism of probability during propagation are analyzed. A probability attenuation model is established: P(d) = P0·e^(-λd), where P(d) is the probability value at propagation distance d, P0 is the initial probability, and λ is the attenuation coefficient. The attenuation loss of each node on the probability propagation chain is calculated, reflecting the increased uncertainty of probability information during propagation. A probability attenuation compensation strategy is designed to reduce probability loss by strengthening key nodes on the propagation path. The compensation process adopts a feedback control mechanism, adjusting the node probabilities on the propagation chain in reverse according to the probability requirements of the target node. The compensation probability calculation formula is established: Pᶜ = P + α·ΔP, where Pᶜ is the compensated probability, P is the original probability, α is the compensation coefficient, and ΔP is the probability loss. Local compensation is implemented in each propagation segment of the probability propagation chain, and the cumulative effect produces the overall compensated probability. The impact of different compensation strategies on the probability propagation effect is analyzed, and the compensation parameters and compensation positions are optimized. Through probability attenuation compensation, the information lost during probability propagation is recovered, and the accuracy of probability estimation is improved.

[0044] The probability of quality achievement is obtained by reconstructing the compensation probability. The compensation probabilities are categorized according to process quality levels, establishing a mapping relationship between quality levels and compensation probabilities. A probability density function fitting method is used to fit the distribution of the compensation probability data, determining the most suitable probability distribution type. The cumulative probability distribution for each quality level is calculated to determine the probability threshold and confidence interval for quality achievement. A model for calculating the probability of quality achievement is then established. Here, Q(x) represents the probability of achieving quality level x, and f(t) is the compensation probability density function. A large number of process quality simulation samples are generated by randomly sampling the compensation probabilities using the Monte Carlo sampling method. The statistical characteristics of the simulation samples are analyzed, and the mean, variance, and quantiles of the quality achievement probability are calculated. A multidimensional quality achievement probability matrix is ​​generated, where the row dimension represents different process schemes, and the column dimension represents different quality indicators.

[0045] The primary and alternative process families are determined using quality achievement probability values. All process schemes are sorted according to their quality achievement probability values, with a probability threshold of 0.8 or higher for the primary process family and between 0.6 and 0.8 for the alternative process family. The primary process family contains the combination of process schemes with the highest quality achievement probability values, offering the best coating quality assurance capability and electroplating process stability. The process characteristics of the primary process family include high current density control precision, small temperature fluctuation range, and stable plating solution composition. The process schemes within the primary process family are categorized into different types, such as constant current primary processes, isothermal control primary processes, and cycle-stable primary processes. The alternative process family contains schemes with a high quality achievement probability value but lower than that of the primary process family, serving as alternatives in case the primary process fails. The alternative process family can provide acceptable coating quality under specific conditions, but its stability is slightly lower than that of the primary process family. A classification of the alternative process family is established, including emergency alternative processes, economical alternative processes, and special requirement alternative processes. Analyze the differences between the main process family and the alternative process families, and determine the switching conditions and conversion mechanisms between the two process families.

[0046] A probabilistic optimization chain is generated based on the main process family and alternative process families. The probabilistic optimization chain uses the main process family as its core backbone and alternative process families as supporting branches, forming a multi-level process optimization network. The process schemes of the main process family are arranged from high to low according to their quality achievement probability values, forming the main path of the probabilistic optimization chain. Corresponding schemes from alternative process families are introduced at key nodes of the main path, establishing switching branches between the main and alternative processes. The node structure of the probabilistic optimization chain is designed, with each node containing information such as process parameter combinations, quality achievement probability values, applicable conditions, and switching logic. The main process family occupies the high-probability region at the front of the probabilistic chain, while alternative process families are distributed in the medium-probability region, forming a gradient probability distribution structure. A probability transfer relationship is established between process schemes in the probabilistic optimization chain, with high-probability main process schemes transferring optimization information to adjacent alternative process schemes. An adjustment mechanism for the probabilistic optimization chain is designed, enabling flexible switching between the main and alternative process families when electroplating production conditions change.

[0047] Step S130: Obtain the stirring flow rate distribution coefficient and the plating solution circulation complexity index from the probabilistic optimization chain, predict the mass transfer efficiency distribution through the stirring flow rate distribution coefficient, determine the flow field control cycle using the plating solution circulation complexity index, and generate the process optimization configuration by combining the mass transfer efficiency distribution and the flow field control cycle.

[0048] Specifically, the stirring velocity distribution coefficient and plating solution circulation complexity index are obtained from the probabilistic optimization chain. The probabilistic optimization chain contains multiple process scheme nodes, each recording the corresponding stirring parameters and circulation system configuration information. All nodes in the main process family and alternative process families in the probabilistic optimization chain are traversed to collect basic data such as stirring speed, impeller geometry parameters, and stirrer position. The stirring Reynolds number Re=ρND² / μ corresponding to each node is calculated using fluid dynamics theory, where Re is the stirring Reynolds number, ρ is the plating solution density, N is the stirring speed, D is the stirrer diameter, and μ is the plating solution dynamic viscosity. A stirring velocity distribution model v(r)=v0(r / R)^α is established, where v(r) is the velocity at a radial distance r, v0 is the stirrer tip velocity, R is the stirrer radius, and α is the distribution index. The stirring velocity distribution coefficient is defined as a composite function of the distribution index α and the stirring intensity, comprehensively reflecting the spatial non-uniformity of the stirring effect. Simultaneously, the complexity of the plating solution circulation system in the probabilistic optimization chain is analyzed. The plating solution circulation complexity index is calculated by weighted summation and includes parameters such as the number of circulation paths, circulation resistance coefficient, and flow distribution ratio.

[0049] In some embodiments, predicting the mass transfer efficiency distribution using the stirring velocity distribution coefficient includes: generating a vortex field intensity distribution by stimulating the plating solution vortex effect using the stirring velocity distribution coefficient; capturing the mass transfer enhancement vortex center in the vortex field intensity distribution; performing energy focusing analysis on the mass transfer enhancement vortex center to form a mass transfer energy density map; and deducing the mass transfer efficiency distribution based on the mass transfer energy density map.

[0050] For example, the step of generating a vortex field strength distribution by stimulating the plating solution vortex effect through the stirring velocity distribution coefficient includes: initiating a spiral flow mode of the plating solution using the stirring velocity distribution coefficient; performing vortex amplification processing on the spiral flow mode to form an enhanced vortex; generating a superimposed field strength through field strength superposition of the enhanced vortex; and generating a vortex field strength distribution by spatial distribution mapping of the superimposed field strength.

[0051] A helical flow mode for the plating solution was initiated using a stirring velocity distribution coefficient. This coefficient provides the distribution patterns of radial and tangential velocities, laying the kinetic foundation for the helical flow. A mathematical description of the helical flow was established, with the velocity field comprising radial, tangential, and axial components. The driving mode for the helical flow was designed using the stirring velocity distribution coefficient, controlling the helical angle and pitch by adjusting the geometric and operational parameters of the stirrer. The characteristic parameters of the helical flow mode are determined by the ratio of tangential to axial velocities to define the helical angle, and by the angular and axial velocities to define the pitch. The influence of the helical flow mode on plating solution mixing and mass transfer was analyzed, revealing that the helical motion enhances axial and radial mass exchange. A stability analysis of the helical flow mode was conducted to determine its stable range and instability conditions. Streamline visualization technology was used to display the three-dimensional structure of the helical flow mode, observing its spatial characteristics and evolution. The influence of different stirring velocity distribution coefficients on the helical flow mode was analyzed, and the coefficients were optimized to achieve the best helical flow effect.

[0052] Enhanced vortices are formed by amplifying the vortex in a spiral flow mode. Based on the initiated spiral flow mode of the plating solution, vortex amplification technology is used to enhance the strength and stability of the vortex structure. The spiral flow mode contains abundant vortex components, and vortex amplification can be achieved through stretching and folding mechanisms. A dynamic equation for vortex amplification is established, including vortex stretching terms, volume change terms, and viscous diffusion terms. The velocity gradient distribution in the spiral flow mode is analyzed to identify regions and conditions favorable for vortex amplification. A control strategy for vortex amplification is designed to achieve effective vortex enhancement by adjusting the intensity and direction of the spiral flow. Vortex dynamics analysis is used to track the transport and amplification process of vortex in the spiral flow. Criteria for enhancing vortices are established, including vortex intensity exceeding a threshold, vortex size meeting requirements, and duration meeting conditions. The spatial distribution characteristics of the enhanced vortices are analyzed to identify concentrated and dispersed vortex regions. The resulting enhanced vortices exhibit higher vortex intensity and a more stable spatial structure.

[0053] A superimposed field strength is generated by superimposing field strengths through enhanced eddies. Enhanced eddies form localized high-intensity field regions in space, and the interaction of multiple eddies produces a superposition effect. A mathematical model for field strength superposition is established, where the total field strength equals the vector sum of the contributions of each enhanced eddy. The interaction mechanisms between enhanced eddies are analyzed, including phenomena such as eddy merging, eddy splitting, and eddy pairing. The influence factors of eddy interaction are calculated, and these factors are related to eddy spacing, intensity ratio, and relative position. An optimization strategy for field strength superposition is established, achieving the best superposition effect by adjusting the position and intensity of the enhanced eddies. The spatial distribution of the superimposed field strength is calculated using numerical integration methods, considering the different contributions from the near and far fields. The stability and persistence of the superimposed field strength are analyzed, and the temporal evolution law of the superposition effect is determined. A method for identifying local enhanced regions of the superimposed field strength is established to locate the positions and ranges of significant field strength enhancement. Through field strength superposition processing, the synergistic enhancement of the eddy effect is achieved, producing a comprehensive field strength effect stronger than that of a single eddy.

[0054] A spatial distribution mapping method is used to generate eddy current field strength distributions from the superimposed field strengths. Based on the generated superimposed field strengths, a spatial distribution mapping algorithm is established to convert the field strength information into a visualized distribution map. The superimposed field strengths contain complex spatial variation information, requiring effective spatial representation through mapping techniques. A spatial coordinate system is established, discretizing the three-dimensional space of the plating tank into a regular grid structure, with each grid node corresponding to a field strength value. An interpolation algorithm is used to process the discrete data of the superimposed field strengths, generating a continuous field strength distribution function. A multi-resolution mapping strategy is adopted, using fine grids in areas of drastic field strength changes and coarse grids in areas of gradual changes. A color mapping scheme for the field strength distribution is established, using rainbow color charts or heatmaps to visually display the magnitude and distribution of the field strength. The spatial characteristics of the eddy current field strength distribution are analyzed to identify high field strength regions, low field strength regions, and regions with gradient changes in field strength. Statistical parameters of the field strength distribution are calculated, including average field strength, maximum field strength, field strength variance, and distribution uniformity index.

[0055] Mass transfer-enhancing vortex centers are captured within the eddy current intensity distribution. The Q criterion is used to identify vortex cores: Q = (1 / 2)(|Ω|² - |S|²), where Q is the vortex core determination parameter, Ω is the rotation tensor, and S is the strain rate tensor; Q > 0 corresponds to a vortex core region. All regions satisfying the Q criterion are marked within the eddy current intensity distribution, forming a set of candidate vortex centers. The mass transfer enhancement potential of each vortex center is analyzed, and a correlation model between vortex intensity and mass transfer coefficient is established. The criteria for determining mass transfer-enhancing vortex centers include vortex intensity exceeding a threshold, duration meeting requirements, and suitable spatial size. The circulation intensity of the vortex center is calculated using line integrals along the vortex boundary; vortices with higher circulation intensity have stronger mass transfer enhancement capabilities. The Lagrange coherent structure analysis method is used to track the motion trajectory and evolution process of the vortex centers. A classification system for mass transfer-enhancing vortex centers is established, classifying vortices into different types based on size, intensity, and location characteristics. The spatial distribution patterns of vortex centers are analyzed to identify the regions and locations with the best mass transfer enhancement effects. Record the characteristic parameters of each mass transfer enhanced vortex center, including center coordinates, intensity value, influence radius, and duration.

[0056] Energy focusing analysis is performed on the centers of mass transfer-enhanced vortices to generate mass transfer energy density maps. The kinetic energy density of each vortex center is calculated to reflect the energy intensity of the vortex motion. An energy focusing model is established to describe the process and intensity distribution of vortex energy concentration towards the central region. The energy dissipation mechanism of the vortex centers is analyzed, and the viscous dissipation rate is calculated to quantify energy loss. The definition and calculation method of mass transfer energy are established; mass transfer energy reflects the contribution of vortex motion to the mass transfer process. The energy balance equation is used to analyze the energy input, transfer, and dissipation processes of the vortex centers, determining the key parameters of energy focusing. The mass transfer energy density is calculated, comprehensively considering the contributions of kinetic energy and concentration gradient. A mass transfer energy density map is plotted, using contour lines and color mapping to display the spatial distribution of energy density. High-energy and low-energy regions in the mass transfer energy density map are analyzed to identify spatial variation patterns of mass transfer efficiency.

[0057] The distribution of mass transfer efficiency is deduced from the mass transfer energy density map. Based on the generated mass transfer energy density map, a quantitative relationship model between mass transfer energy and mass transfer efficiency is established. Using mass transfer theory, the correlation between the mass transfer coefficient and local energy density is established: k = aRe^bSc^c, where k is the mass transfer coefficient, a, b, and c are empirical constants, Re is the Reynolds number, and Sc is the Schmidt number. The energy distribution characteristics of different regions in the mass transfer energy density map are analyzed, and the average energy density and energy density gradient of each region are calculated. A spatial distribution function of mass transfer efficiency is established, transforming the continuous distribution of the mass transfer energy density map into a continuous distribution of mass transfer efficiency. Numerical interpolation methods are used to process the discrete data in the mass transfer energy density map, generating a high-resolution mass transfer efficiency distribution map. The statistical characteristics of the mass transfer efficiency distribution are analyzed, and the average mass transfer efficiency, efficiency variance, and efficiency distribution uniformity index are calculated. High-efficiency and low-efficiency regions in the mass transfer efficiency distribution are identified, and the causes of uneven efficiency distribution and directions for improvement are analyzed.

[0058] The flow field control cycle time is determined using the plating solution circulation complexity index. Based on the obtained plating solution circulation complexity index, a time-series control strategy for the flow field control system is established. The plating solution circulation complexity index reflects the operational difficulty and control precision requirements of the circulation system; higher complexity necessitates a more refined control cycle time. The relationship between the plating solution circulation complexity index and the system response time is analyzed, and a calculation model for the control cycle time is established. The control period is determined through a linear relationship. According to the numerical range of the plating solution circulation complexity index, the flow field control cycle time is divided into three levels: rapid control, standard control, and fine control. Rapid control corresponds to low-complexity operating conditions, with a short control period, suitable for simple circulation systems. Standard control corresponds to medium-complexity operating conditions, with a moderate control period, balancing response speed and control precision. Fine control corresponds to high-complexity operating conditions, with a longer control period, emphasizing control precision and system stability.

[0059] This paper combines mass transfer efficiency distribution and flow field control timing to generate a process optimization configuration. Based on the predicted mass transfer efficiency distribution and the determined flow field control timing, a comprehensive process optimization configuration scheme is constructed. Mass transfer efficiency distribution data and flow field control timing information are matched spatially and temporally to establish a multi-dimensional optimization parameter matrix. The mass transfer efficiency distribution provides guidance for spatial optimization, indicating the key areas and intensity requirements for optimization. The flow field control timing provides a control strategy for temporal optimization, determining the timing and frequency of control actions. An objective function for the process optimization configuration is established, comprehensively considering multiple objectives such as maximizing mass transfer efficiency, minimizing energy consumption, and maximizing equipment utilization. A multi-objective optimization algorithm is used to solve for the optimal configuration parameters, including agitator speed distribution, circulating pump flow rate regulation, and valve opening control. The process optimization configuration adopts a regional control strategy, dividing the electroplating tank into several control regions according to the mass transfer efficiency distribution, with each region implementing independent optimization control.

[0060] Step S140: Use the probabilistic optimization chain to determine the high-risk process segment and the stable process segment, combine the high-risk process segment and the stable process segment to form a quality-balanced process pair, and generate process coordination rules based on the quality-balanced process pair.

[0061] Specifically, a probabilistic optimization chain is used to identify high-risk and stable process segments. A risk assessment model is established, defining process segments with a quality achievement probability value below 0.6 as high-risk segments and those above 0.8 as stable segments. The probability fluctuation characteristics of each process segment in the probabilistic optimization chain are analyzed, and the probability variance σ²=Σ(pi-μ)² / n is calculated, where σ² is the probability variance, pi is the probability value of the i-th node, μ is the average probability, and n is the total number of nodes. High-risk process segments typically exhibit low probability values ​​and drastic fluctuations, while stable process segments show high probability values ​​and gradual changes. Typical high-risk process segments include the rapid current density adjustment stage, the period of plating solution temperature change, and the additive replenishment process. Stable process segments mainly include the constant current plating stage, the steady-state temperature maintenance period, and the stable operation period of the plating solution circulation. A sliding window technique is used to identify continuous risk segments on the probabilistic optimization chain, with the window length determined based on the characteristics of the plating process cycle. A classification standard for process segments is established, comprehensively considering factors such as probability value, probability gradient, and duration. The reasons for the formation of high-risk process sections include high current density sensitivity, difficulty in temperature control, and strong fluctuations in plating solution composition.

[0062] In some embodiments, the step of combining the high-risk process segment with the stable process segment to form a quality-balanced process pair includes: performing risk energy release analysis on the high-risk process segment to generate a risk release spectrum; extracting risk absorption capacity from the stable process segment to construct an absorption capacity pool; performing energy balance matching between the risk release spectrum and the absorption capacity pool to obtain a balance threshold; and performing a hedging combination to form a quality-balanced process pair according to the balance threshold.

[0063] For example, the step of performing risk energy release analysis on the high-risk process segment to generate a risk release spectrum includes: performing cross-cycle energy tracing analysis on the high-risk process segment to obtain energy accumulation data; using the energy accumulation data to analyze the energy impact effect of each process element to generate an energy release intensity distribution, wherein each process element includes current density fluctuation, temperature gradient change and plating solution composition shift; and performing spectrum conversion processing on the energy release intensity distribution to generate a risk release spectrum.

[0064] A cross-cycle energy tracing analysis was conducted on high-risk process sections to obtain energy accumulation data. High-risk process sections exhibit energy accumulation effects during continuous production cycles; the overshoot energy during the rapid current density adjustment phase affects coating uniformity in subsequent cycles. An energy accumulation function E(t) = ΣᵢαᵢE(ti)e^(-λi) was established, where E(t) is the accumulated energy at the current moment, αᵢ is the weighting coefficient of the i-th historical cycle, λ is the time decay factor, i is the cycle interval, and e is the natural constant. The runaway energy caused by uneven electrochemical reactions, the compositional deviation energy caused by anode dissolution fluctuations, and the mass transfer instability energy caused by turbulent stirring in high-risk process sections were analyzed. Energy accumulation during current density adjustment mainly originates from the transient responses of ohmic polarization and concentration polarization, while energy accumulation during the plating solution temperature change period stems from changes in heat capacity and reaction kinetics fluctuations. The energy input for each cycle was quantified using an electrical charge calculation method, determined through the integral relationship between the square of the current, resistance, and time. The spatial distribution of energy accumulation was analyzed to identify the energy concentration characteristics of the high current density region at the anode edge and the low flow velocity region at the bottom of the tank. By performing cross-cycle analysis, a complete dataset containing the accumulation of electrochemical energy, thermal energy, and kinetic energy is obtained.

[0065] The energy impact of various process elements was analyzed using energy accumulation data to generate an energy release intensity distribution. Current density fluctuations directly affect the deposition rate of metal ions by altering the local electric field strength; the fluctuation amplitude has a quadratic relationship with the energy change. Temperature gradient changes affect the conductivity and ion mobility of the plating solution; a 10°C change in temperature leads to approximately a 30% change in conductivity, corresponding to a 15-25% increase in energy release intensity. The composition shift of the plating solution is mainly manifested in changes in the main salt concentration and additive consumption; a 5 g / L decrease in main salt concentration corresponds to an approximately 8% increase in resistance, and a 20% shift in additive concentration causes changes in surface tension, thus affecting bubble adhesion and mass transfer. An energy impact model of process elements was established, expressing the coupling effect of current density, temperature, and composition through a weighted combination. The differences in energy distribution of each process element at the anode surface, the plating solution body, and the cathode interface were analyzed. The anode region is mainly affected by current density fluctuations, the plating solution body is most sensitive to temperature changes, and the cathode interface has the strongest response to composition shifts. The energy release intensity of each process element is calculated. The release intensity in the current density fluctuation region can reach 1.2-1.8 times the reference value, the temperature gradient region is 1.1-1.5 times, and the composition offset region is 1.05-1.3 times.

[0066] The energy release intensity distribution is processed by spectral transformation to generate a risk release spectrum. The energy release intensity distribution reflects the difference in risk levels at different locations within the electroplating tank. Spectral analysis can identify periodic variations and spatial frequency characteristics. A two-dimensional fast Fourier transform is used to perform spectral analysis on the intensity distribution on the anode surface: F(kx,ky)=∫∫I(x,y)e^(-i2π(kxx+kyy))dxdy, where F(kx,ky) is the spectral function, I(x,y) is the intensity distribution function, kx and ky are the spatial frequency components in the x and y directions, i is the imaginary unit, and π is pi. Analyzing the main frequency components in the spectrum, low-frequency components correspond to large-scale current distribution non-uniformity, while high-frequency components reflect local microscopic non-uniformity. The main characteristic frequencies of the electroplating tank include the agitator blade passage frequency, the temperature control period frequency, and the power supply ripple frequency. The power spectral density is calculated; the power concentration in the low-frequency band indicates that large-scale non-uniformity is dominant, while the large high-frequency power indicates severe local fluctuations. Establish spectral characteristic parameters, including dominant frequency, spectral width, spectral centroid, and spectral asymmetry. Analyze the directional characteristics of the spectrum; the radial frequency components reflect the non-uniformity from the center to the edge, while the tangential frequency components correspond to the circumferential changes caused by stirring.

[0067] This study explores risk absorption capacity in stable process sections to construct an absorption capacity pool. The constant current electroplating stage, characterized by stable current and reaction equilibrium, can absorb disturbances from current fluctuations. The thermal buffer system during the steady-state temperature maintenance period can absorb temperature disturbances, and the thermal inertia of the large-capacity plating solution provides a stable temperature reference. The flow field homogenization effect during the stable operation of the plating solution circulation can absorb local concentration disturbances and mass transfer inhomogeneities. A definitional model for risk absorption capacity is established: current absorption capacity is calculated by dividing the difference between the maximum withstand current and the nominal current by the current standard deviation. Temperature absorption capacity is determined by the product of density, specific heat capacity, volume, and temperature difference. Concentration absorption capacity is expressed as the product of volume and concentration range. The absorption mechanism in stable process sections is analyzed. The proportional-integral-derivative (PID) control system provides rapid response and steady-state accuracy; the acid-base balance of the plating solution buffer system maintains pH stability; and the turbulent mixing of the stirring and homogenizing system eliminates local inhomogeneities. The upper limit of absorption capacity for various stable process sections was calculated. The constant current section can withstand current fluctuations of ±10%, the steady-state temperature section can buffer temperature changes of ±5℃, and the cyclic stabilization section can handle concentration deviations of 20%. The spatial distribution of absorption capacity was established, showing that the absorption capacity is strongest in the central area of ​​the tank and relatively weaker in the corner areas. The buffer capacity near the anode is affected by the electrochemical reaction.

[0068] The energy balance threshold is obtained by matching the risk release spectrum with the absorption capacity pool. Based on the generated risk release spectrum and the constructed absorption capacity pool, an energy balance matching algorithm is established to achieve the optimal pairing of risk and capacity in the electroplating process. The frequency characteristic distribution of the risk release spectrum is analyzed to identify the main energy release frequency bands of each high-risk process segment, while the frequency response characteristics of each stable process segment in the absorption capacity pool are extracted. A spectrum matching calculation method is established: R=∫F1(ω)F2*(ω)dω / √(∫|F1(ω)|²dω∫|F2(ω)|²dω), where R is the matching correlation coefficient, F1 is the risk release spectrum, F2 is the absorption capacity spectrum, * denotes complex conjugation, and ω is the angular frequency. By calculating the spectrum matching degree of different process segment combinations, risk-stable pairing relationships with high correlation are identified. Energy balance constraints are established to ensure that the total released energy does not exceed the total absorption capacity, and a safety margin is set to prevent capacity saturation. The process constraints during the matching process are analyzed. Current matching requires release and absorption to be synchronized in time, temperature matching needs to consider the thermal conduction delay, and concentration matching is limited by the diffusion rate. An iterative optimization algorithm is used to determine the critical equilibrium points of each process parameter. Taking into account factors such as spectral matching degree, timing constraints, and safety margins, the equilibrium thresholds for key parameters such as current, temperature, and concentration are obtained.

[0069] Based on the equilibrium threshold, a hedging combination is executed to form a quality-balanced process pair. A specific hedging combination strategy is designed for the high-risk process segment and the stable process segment based on the obtained equilibrium threshold. When the fluctuation amplitude of the current density rapid adjustment segment exceeds the equilibrium threshold ±8A / dm², the compensation mechanism of the constant current electroplating segment is activated, dispersing the impact energy to the stable segment for absorption through preset buffer time and gradual transition. The hedging combination for the temperature change period and the steady-state temperature maintenance period adopts a preheating and precooling strategy. The temperature reserve of the steady-state segment is activated 30 minutes before the temperature change, and the change amplitude is controlled within ±4℃ of the equilibrium threshold. The combination of the additive replenishment process and the stable operation period is achieved through batch addition and enhanced circulation. The single addition amount does not exceed 25% of the total amount, and the circulation flow rate is immediately increased to 150% of the normal value after addition. Timing coordination of the hedging combination is established: the stable process segment enters the preparation state 10 minutes before the start of the high-risk process segment; the stable segment maintains maximum absorption capacity during the execution of the high-risk segment; and the stable segment continues to run for 20 minutes after the high-risk segment ends to ensure complete absorption.

[0070] Based on the quality balance process pair, process coordination rules are generated. The quality balance process pair records the matching information between high-risk and stable process segments, including key data such as risk type, absorption capacity, balance threshold, and trigger conditions. According to the risk-stability pairing relationship in the quality balance process pair, corresponding coordination control logic is generated. When the quality balance process pair shows a balance between the "rapid current density adjustment stage" and the "constant current plating stage," a coordination rule is generated: if the current density fluctuation exceeds the balance threshold ±8A / dm², constant current stabilization control is immediately initiated, limiting the current density adjustment rate to within 2A / dm² / min, while simultaneously activating the current pre-compensation algorithm to reduce overshoot. When the "plating bath temperature change period" and the "steady-state temperature maintenance period" are paired in the quality balance process pair, a temperature coordination rule is generated: if the temperature deviation exceeds the balance threshold ±4℃, a temperature buffer mechanism is activated, controlling the temperature change rate to within 1℃ / min through preheating the plating bath and segmented heating, while simultaneously turning on the auxiliary heater to provide temperature reserves. When the "additive replenishment process" and the "cycle stable operation period" form a process pair, a concentration coordination rule is established: before additive addition, an enhanced circulation mode is activated, the flow rate is increased to 150%, the addition is carried out using a multi-point dispersion method, and the amount added at one time is limited to within 25% of the total demand. After addition, the enhanced circulation is maintained for 30 minutes to ensure uniform distribution. Based on the balance threshold data of the quality balance process pair, the trigger conditions for each coordination rule are set as follows: when the deviation reaches 80% of the threshold, an early warning is issued; when it reaches 100%, execution is carried out; and when it exceeds 120%, an emergency response is initiated.

[0071] Step S150: Based on the process coordination rules and process optimization configuration, perform temperature field analysis to determine the hot spot distribution, identify areas with excessive temperature gradients through the hot spot distribution, extract heat dissipation optimization factors from the areas with excessive temperature gradients, and use the heat dissipation optimization factors to generate cycle control parameters.

[0072] Specifically, temperature field analysis was conducted based on process coordination rules and process optimization configuration to determine hotspot distribution. The process coordination rules include triggering conditions and response strategies for temperature control, such as activating a buffer mechanism when the temperature deviation exceeds ±4℃ and limiting the temperature change rate to within 1℃ / min. The process optimization configuration provides key parameters such as stirrer speed distribution, heater power configuration, and circulation flow rate settings. Combining the temperature control requirements in the process coordination rules, the temperature distribution characteristics of different areas within the electroplating tank were analyzed. In the anode region, Joule heating occurs due to electrochemical reactions, resulting in temperatures typically several degrees Celsius higher than the main body of the plating solution. Localized high-temperature zones form near the heaters, reaching the upper limit of the set value. The temperature at the bottom of the tank is lower due to gravity stratification, and the temperature distribution is uneven at the corners due to circulation dead zones. A temperature field calculation equation was established: ∂T / ∂t=α∇²T+S, where T is temperature, t is time, α is the thermal diffusivity, ∇² is the Laplace operator, and S is the heat source term. Based on the stirring parameters and circulation flow rate in the process optimization configuration, the convective heat transfer coefficient and flow field distribution were calculated. The influence of heater layout and power distribution on the temperature field was analyzed. Centralized heating easily leads to hot spots, while distributed heating is beneficial for temperature uniformity. The main heat sources within the electroplating bath were identified, including electrochemical reaction heat, ohmic loss heat, and frictional heat from stirring. The temperature field distribution was solved using the finite element method, with mesh generation considering geometric complexity and temperature gradient changes. The locations of hot spots were determined; hot spots typically appear at the center of the anode surface, near the heater, and in high current density regions, where the temperature is several degrees Celsius higher than the surrounding area.

[0073] In some embodiments, identifying regions with excessive temperature gradients through the hotspot distribution includes: using the hotspot distribution to perform heat source intensity inversion to obtain heat source distribution density; performing gradient field reconstruction processing on the heat source distribution density to generate a temperature gradient field; performing a threshold cutting operation in the temperature gradient field to divide gradient levels; and identifying regions with excessive temperature gradients by filtering the gradient levels.

[0074] The heat source intensity is obtained by inverting the heat source distribution density using hotspot distribution. Based on the determined hotspot distribution, an inversion algorithm is used to deduce the intensity distribution and spatial density of heat sources within the electroplating bath. The hotspot distribution reveals high-temperature regions, and the heat source characteristics generating these hotspots are determined through inversion analysis. A heat source intensity inversion calculation is established, using the observed hotspot temperature values ​​as boundary conditions to infer the internal heat source distribution density. The adjoint method is used to solve the inversion problem, establishing the adjoint equation ∂φ / ∂t+α∇²φ=δ(x-xs), where φ is the adjoint variable, δ is the Dirac function, and xs is the measurement point location. The electrochemical reaction heat source is mainly distributed on the anode surface, with a heat source intensity qe=I²R / A, where I is the current, R is the resistance, and A is the area. Ohmic loss heat sources are distributed throughout the entire conductive path, with intensity proportional to the square of the current density. The frictional heat source is concentrated around the stirrer blades, with an intensity qm = μ(∂v / ∂y)², where μ is the dynamic viscosity and ∂v / ∂y is the velocity gradient. The spatial distribution characteristics of different types of heat sources are analyzed: reaction heat sources are distributed planarly, ohmic heat sources are distributed volumetrically, and frictional heat sources are distributed along the stirring trajectory. The heat source distribution density in each region is calculated; the density is highest in the anode center region, moderate in the edge regions, and low in the bulk of the bath liquid. A heat source distribution density map is established, using contour lines and color mapping to visually display density changes. The non-uniformity of the heat source distribution is analyzed, and the variance and skewness parameters are calculated. High variance indicates high heat source concentration, and positive skewness indicates the existence of significant high-density regions.

[0075] A gradient field reconstruction process is performed on the heat source distribution density to generate a temperature gradient field. Based on the obtained heat source distribution density, the temperature gradient distribution within the electroplating tank is calculated using a gradient field reconstruction algorithm. The heat source distribution density provides spatial distribution information of heat generation, which needs to be combined with the heat transfer mechanism to calculate the corresponding temperature gradient field. The gradient field reconstruction calculation is established using Fourier's law of heat conduction: q = -k∇T, where q is the heat flux density vector, k is the thermal conductivity, and ∇T is the temperature gradient. The heat source distribution density is used as the source term of the heat flux density, and the temperature field is reconstructed by solving the Poisson equation ∇²T = -S / k, thereby calculating the temperature gradient. The temperature gradient field contains components in three directions: x, y, and z, ∇T = (∂T / ∂x, ∂T / ∂y, ∂T / ∂z). The physical meaning of the gradients in each direction is analyzed: the horizontal gradient reflects the transverse temperature difference within the tank, while the vertical gradient reflects the stratification effect and buoyancy influence. The magnitude of the temperature gradient is calculated as |∇T| = √[(∂T / ∂x)² + (∂T / ∂y)² + (∂T / ∂z)²]. The gradient magnitude reflects the drasticness of temperature change. The temperature gradient is largest near the anode surface, relatively smaller in the middle of the bath, and larger in the boundary layer region due to convective heat transfer. The directional characteristics of the temperature gradient are analyzed; the gradient vector points in the direction of the fastest temperature increase and is radially distributed near the heat source. A method for representing the gradient field is established, using a vector diagram to display the gradient direction and color intensity to represent the gradient magnitude.

[0076] A threshold-cutting operation is performed in the temperature gradient field to classify gradient levels. Based on the generated temperature gradient field, a threshold-cutting technique is used to divide the continuous gradient distribution into different level intervals. The temperature gradient field contains continuous changes from low to high gradients, and threshold-cutting is needed to identify gradient regions of different degrees. Gradient level classification criteria are established, and gradient thresholds are set according to the requirements of temperature uniformity in the electroplating process. Slight gradient levels correspond to small temperature change rates, moderate gradient levels represent moderate temperature changes, severe gradient levels represent significant temperature changes, and extremely severe gradient levels reflect drastic temperature changes. The threshold-cutting algorithm is used to process the temperature gradient field data, converting continuous gradient values ​​into discrete level identifiers. The spatial distribution characteristics of each gradient level are analyzed. Slight gradient regions are mainly located in the center of the bath, moderate gradient regions are distributed in the transition zone, and severe gradient regions are concentrated near the heat source and the boundary layer. The area proportion of each level region is calculated. Under normal electroplating conditions, slight gradient regions should occupy the main part of the total area, and severe gradient regions should be controlled within a small range.

[0077] Regions with excessive temperature gradients are identified through gradient level screening. Based on the defined gradient levels, a screening algorithm is used to identify regions where the temperature gradient exceeds the acceptable range. Regions with severe and extremely severe gradient levels are defined as regions with excessive temperature gradients. These regions experience drastic temperature changes, which may affect electroplating quality and coating uniformity. Identification criteria for regions with excessive gradients are established, considering both gradient magnitude and continuity. The gradient magnitude criterion requires the temperature gradient to exceed the process allowable range, while the continuity criterion requires the area of ​​the excessive gradient region to reach a certain scale. Typical locations of regions with excessive temperature gradients are analyzed, mainly including the anode surface center, the heater direct heating zone, the stirrer high-shear zone, and the thermally conductive boundary of the tank wall. The anode center region has the highest current density and the most intense electrochemical reaction, resulting in a significant temperature gradient. Near the heater, due to direct heat input, a distinct thermal boundary layer is formed, resulting in a large gradient. Around the stirrer blades, mechanical energy is converted into heat energy, causing localized temperature rises and gradient concentration. The geometric characteristics of regions with excessive gradients are calculated, including parameters such as region area, perimeter, aspect ratio, and shape factor. The influence mechanism of regions with excessive gradient on the electroplating process was analyzed. The ion migration rate in high gradient regions is uneven, which leads to differences in coating thickness and uneven distribution of internal stress.

[0078] Heat dissipation optimization factors are extracted from regions with excessive temperature gradients. These regions require enhanced heat dissipation to reduce the temperature gradient, and the heat dissipation optimization factors reflect the direction and intensity of heat dissipation improvement. A definition method for the heat dissipation optimization factors is established, including parameters such as the heat dissipation area enhancement coefficient, the convective heat transfer coefficient enhancement factor, and the reduction in thermal resistance. The heat dissipation limiting factors in regions with excessive temperature gradients are analyzed: the anode surface area is blocked by electrochemical reaction products, there is a lack of effective convection channels near the heater, and thermal resistance concentration exists in the boundary layer region. The heat dissipation area enhancement coefficient Af = A_enhanced / A_original is calculated, where Af is the area enhancement coefficient, A_enhanced is the enhanced heat dissipation area, and A_original is the original heat dissipation area. The heat dissipation area can be increased by adding heat dissipation fins, improving surface roughness, and optimizing the geometry. The potential for improving the convective heat transfer coefficient is analyzed; the heat transfer coefficient can be improved by increasing the stirring intensity, optimizing the flow field distribution, and increasing the fluid velocity. The convective heat transfer coefficient enhancement factor hf = h_enhanced / h_original is calculated, where hf is the heat transfer coefficient enhancement factor, h_enhanced is the enhanced heat transfer coefficient, and h_original is the original heat transfer coefficient.

[0079] Cyclic control parameters are generated using heat dissipation optimization factors. Based on the extracted heat dissipation optimization factors, an algorithm for generating cycle control parameters is established to achieve optimized temperature field control. The heat dissipation optimization factors provide information on the direction and intensity of heat dissipation improvement, which needs to be converted into specific control parameters to guide equipment operation. A method for generating cycle control parameters is established, including heater power adjustment cycle, stirrer speed variation cycle, and circulating pump flow control cycle. The heater power adjustment cycle is calculated, and the amplitude and frequency of power adjustment are determined based on the heat dissipation area enhancement coefficient: T_heat = T_base × (1 + α_f × Af), where T_heat is the heating cycle period, T_base is the standard period, α_f is the adjustment coefficient, and Af is the area enhancement coefficient. The stirrer speed variation cycle is designed, and the speed adjustment strategy is determined based on the heat transfer coefficient enhancement factor: N_adjust = N_base × (1 + β_f × hf), where N_adjust is the adjusted speed, N_base is the standard speed, β_f is the speed adjustment coefficient, and hf is the heat transfer coefficient enhancement factor. A circulating pump flow control cycle is established, which enhances heat transfer and reduces dead zones through flow pulsation. The generated cycle control parameters include complete information such as cycle time, adjustment range, phase relationship, and execution priority, providing precise control parameter guidance for the temperature field of the electroplating tank.

[0080] Step S160: Match the cycle control parameters with the probability optimization chain to generate dynamic production units, use the dynamic production units to generate flexible production processes, and complete the big data analysis and optimization of electroplating tank production and maintenance.

[0081] In some embodiments, the step of matching the cycle control parameters with the probabilistic optimization chain to generate dynamic production units includes: constructing a coating growth rate monitoring network using the cycle control parameters; mapping the probabilistic optimization chain to the coating growth rate monitoring network to obtain thickness monitoring points; performing uniformity analysis on the thickness monitoring points to determine thickness distribution clusters; and adjusting current allocation according to the thickness distribution clusters to generate dynamic production units.

[0082] A plating growth rate monitoring network was constructed using cycle time control parameters. Based on the generated cycle time control parameters, a real-time monitoring network covering the entire electroplating tank was established. The sampling period of the temperature monitoring points was determined according to the calculated heater power adjustment cycle time T_heat, and the monitoring frequency was set as a multiple of the cycle time to ensure accurate capture of temperature changes. The response time of the flow field monitoring points was set based on the stirrer speed adjustment value N_adjust, and the data update frequency of the monitoring points was synchronized with the speed change cycle time to achieve real-time tracking of flow field distribution changes. The plating solution circulation mode was adjusted by using the circulation pump flow control cycle time, and concentration monitoring points were configured accordingly in the circulation-affected area, with the monitoring interval matching the flow pulsation period. The node layout of the monitoring network was established, and the location and density of monitoring points were determined according to the specific values ​​of the cycle time control parameters. Dense temperature-sensitive monitoring points were set in the high-temperature area affected by the heater power adjustment cycle time T_heat, flow field monitoring points were arranged in the strong shear area affected by the stirrer speed N_adjust, and concentration monitoring points were configured in the main flow path covered by the circulation pump. A method for measuring the coating growth rate was designed, and electrochemical quartz crystal microbalance technology was used to achieve real-time thickness monitoring with nanometer-level precision. The growth rate calculation formula was established as v = dh / dt, where v is the growth rate, h is the coating thickness, and t is time.

[0083] A probabilistic optimization chain is mapped to a coating growth rate monitoring network to obtain thickness monitoring points. High-probability nodes in the main process family of the probabilistic optimization chain correspond to areas with stable quality within the electroplating bath, where coating growth should maintain good consistency. Medium-probability nodes in the alternative process families correspond to areas with larger quality fluctuations, requiring focused monitoring of coating thickness trends. A mapping algorithm is established to associate the combination of process parameters in the probabilistic optimization chain with the spatial location of the monitoring network. The influence of process parameters on coating thickness distribution is analyzed: current density directly determines the local coating growth rate, temperature affects the crystal morphology and density of the coating, and concentration affects ion supply and coating continuity. Based on the quality achievement probability values ​​of each node in the probabilistic optimization chain, the importance level of the corresponding monitoring point is determined. Monitoring points corresponding to high-probability nodes are subject to continuous high-precision monitoring, those corresponding to medium-probability nodes are subject to periodic focused monitoring, and those corresponding to low-probability nodes are subject to routine monitoring. A labeling system for thickness monitoring points is established, including location coordinates, probability level, monitoring frequency, and alarm threshold information.

[0084] Uniformity analysis of thickness monitoring points was performed to identify thickness distribution clusters. Thickness statistical parameters for each monitoring point were calculated, including average thickness, thickness variance, maximum deviation, and distribution range. A thickness uniformity evaluation system was established, using the coefficient of variation (CV) = σ / μ to quantify the uniformity of thickness distribution, where σ is the thickness standard deviation and μ is the average thickness. Spatial patterns of thickness distribution were analyzed to identify the distribution characteristics of areas with excessive thickness, areas with excessive thickness, and areas with normal thickness. Clustering analysis algorithms were used to group thickness monitoring points, grouping those with similar thickness characteristics into the same cluster. A distance calculation method was established: d = √[(h1-h2)² + (x1-x2)² + (y1-y2)²], where d is the distance, h is the thickness value, and x and y are spatial coordinates. The k-means clustering algorithm was used to divide the monitoring points into several thickness distribution clusters, each cluster representing a region with similar thickness characteristics. Characteristic parameters of each thickness distribution cluster were analyzed, including cluster center thickness, thickness variance within the cluster, cluster coverage area, and cluster boundary shape. The causes of thickness distribution clusters can be identified, including factors such as uneven current distribution, temperature gradient influence, the existence of dead zones in the flow field, and uneven distribution of additives.

[0085] Dynamic production units are generated by adjusting current distribution based on thickness distribution clusters. The current demand characteristics of each thickness distribution cluster are analyzed: clusters with lower thickness require increased current supply, clusters with higher thickness require appropriate reduction in current density, and clusters with normal thickness maintain their current level. A current distribution adjustment algorithm is established, calculating the corresponding current adjustment amount ΔI = k × (h_target - h_actual) based on the thickness deviation of each cluster, where ΔI is the current adjustment amount, k is the proportional coefficient, h_target is the target thickness, and h_actual is the actual thickness. A multi-zone current control system is designed, dividing the electroplating tank into multiple current control zones, each corresponding to a specific thickness distribution cluster. A current distribution matrix is ​​established to describe the current distribution ratio and adjustment range of each control zone. The impact mechanism of current adjustment on coating quality is analyzed: moderate current adjustment can improve thickness uniformity, while excessive adjustment may affect the adhesion and surface quality of the coating. Safety constraints for current adjustment are designed to ensure that the adjusted current density remains within the process allowable range. The organizational structure of the dynamic production unit is established, with each unit containing a specific thickness distribution cluster, a corresponding current distribution strategy, and supporting control parameters.

[0086] Based on the constructed dynamic production units, a flexible production process organizational structure is established, combining multiple dynamic production units in series and parallel according to process logic and quality requirements. A process switching mechanism is designed to enable flexible switching between different dynamic production units based on changes in product specifications, equipment status adjustments, and quality indicator requirements. A coordination and control strategy for production units is established to ensure parameter matching and timing continuity between units, avoiding process fluctuations during switching. Through the modular combination of dynamic production units, a flexible process configuration adaptable to different production needs is formed, improving the responsiveness and adaptability of the production system. Simultaneously, a big data analysis system for electroplating tank production operation and maintenance is established, integrating process parameter data, equipment status data, product quality data, and energy consumption and environmental protection data generated during the operation of dynamic production units. Data correlation analysis identifies key factors affecting product quality and production efficiency, establishing a quantitative relationship between process parameters and quality indicators. A comprehensive evaluation index system for production operation and maintenance is established, including dimensions such as quality stability, production efficiency, equipment reliability, and cost control. Through continuous data analysis and process optimization, big data analysis and optimization of electroplating tank production operation and maintenance are completed.

[0087] To implement the big data analysis and optimization method for electroplating tank production and operation corresponding to the above method embodiments, in order to achieve the corresponding functions and technical effects. See also Figure 2 , Figure 2 This diagram illustrates a structural block diagram of an electroplating tank production operation and maintenance big data analysis and optimization system 200 provided in an embodiment of this application. For ease of explanation, only the parts relevant to this embodiment are shown. The electroplating tank production operation and maintenance big data analysis and optimization system 200 provided in this embodiment includes:

[0088] Data acquisition module 201 is used to acquire electroplating tank process parameter data, determine the key control node sequence based on the process parameter data, and establish an operation and maintenance status mapping table based on the key control node sequence.

[0089] The probability analysis module 202 is used to perform probability backtracking processing on the operation and maintenance status mapping table to generate a process probability distribution tree, perform branch analysis on the process probability distribution tree to obtain a quality achievement probability value, use the quality achievement probability value to determine the main process family and the alternative process family, and generate a probability optimization chain based on the main process family and the alternative process family.

[0090] The flow field optimization module 203 is used to obtain the stirring flow rate distribution coefficient and the plating solution circulation complexity index from the probabilistic optimization chain, predict the mass transfer efficiency distribution through the stirring flow rate distribution coefficient, determine the flow field control cycle using the plating solution circulation complexity index, and generate process optimization configuration by combining the mass transfer efficiency distribution and the flow field control cycle.

[0091] The risk balancing module 204 is used to determine high-risk process segments and stable process segments using the probability optimization chain, to combine the high-risk process segments and the stable process segments in a hedging combination to form a quality-balanced process pair, and to generate process coordination rules based on the quality-balanced process pair.

[0092] Temperature control optimization module 205 is used to perform temperature field analysis to determine hot spot distribution based on the process coordination rules and the process optimization configuration, identify areas with excessive temperature gradients through the hot spot distribution, extract heat dissipation optimization factors from the areas with excessive temperature gradients, and use the heat dissipation optimization factors to generate cycle control parameters.

[0093] The dynamic execution module 206 is used to match the cycle control parameters with the probability optimization chain to generate dynamic production units, and use the dynamic production units to generate flexible production processes to complete big data analysis and optimization of electroplating tank production and maintenance.

[0094] The electroplating tank production operation and maintenance big data analysis and optimization system 200 described above can implement the electroplating tank production operation and maintenance big data analysis and optimization method of the above method embodiments. The options in the above method embodiments are also applicable to this embodiment, and will not be detailed here. The remaining content of this application embodiment can be referred to the content of the above method embodiments, and will not be repeated in this embodiment.

[0095] The purpose of the above embodiments is to reproduce and derive the technical solution of the present invention by way of example, and to fully describe the technical solution, purpose and effect of the present invention. The purpose is to enable the public to have a more thorough and comprehensive understanding of the disclosure of the present invention, and not to limit the scope of protection of the present invention.

[0096] The above embodiments are not an exhaustive list based on the present invention, and there may be many other embodiments not listed. Any substitutions and improvements made without departing from the concept of the present invention are within the protection scope of the present invention.

Claims

1. A method for big data analysis and optimization of electroplating tank production and operation, characterized in that, include: Obtain electroplating tank process parameter data, determine the key control node sequence based on the process parameter data, and establish an operation and maintenance status mapping table based on the key control node sequence. The operation and maintenance status mapping table is subjected to probabilistic backtracking to generate a process probability distribution tree. Branch analysis is performed on the process probability distribution tree to obtain the quality achievement probability value. The quality achievement probability value is used to determine the main process family and the alternative process family. A probabilistic optimization chain is generated based on the main process family and the alternative process family. The stirring flow rate distribution coefficient and the plating solution circulation complexity index are obtained from the probabilistic optimization chain. The mass transfer efficiency distribution is predicted by the stirring flow rate distribution coefficient. The flow field control cycle is determined by the plating solution circulation complexity index. The process optimization configuration is generated by combining the mass transfer efficiency distribution and the flow field control cycle. High-risk process segments and stable process segments are determined using the probabilistic optimization chain. The high-risk process segments and stable process segments are then combined in a hedging manner to form quality-balanced process pairs. Process coordination rules are generated based on the quality-balanced process pairs. Based on the process coordination rules and the process optimization configuration, temperature field analysis is performed to determine the hot spot distribution. The hot spot distribution is used to identify areas with excessive temperature gradients. Heat dissipation optimization factors are extracted from the areas with excessive temperature gradients. The heat dissipation optimization factors are then used to generate cycle control parameters. The cycle control parameters are matched with the probability optimization chain to generate dynamic production units. The dynamic production units are then used to generate flexible production processes, thereby completing big data analysis and optimization of electroplating tank production and maintenance.

2. The method according to claim 1, characterized in that, The step of determining the sequence of key control nodes based on the process parameter data includes: The current density spectrum was obtained by measuring the current distribution on the anode surface using the process parameter data. The uniformity of the current density spectrum is evaluated to determine the current distribution deviation. The current distribution deviation is used to perform node impact analysis and generate distribution compensation coefficients; The sequence of key control nodes is determined by optimizing the distribution compensation coefficient.

3. The method according to claim 1, characterized in that, The step of performing branch analysis on the process probability distribution tree to obtain the quality achievement probability value includes: Extract probability jump nodes from the process probability distribution tree; Chain reaction analysis is performed on the probabilistic jump nodes to form a probability propagation chain; A probability attenuation compensation process is performed along the probability propagation chain to generate a compensation probability. The probability value of achieving quality is obtained by reconstructing the compensation probability.

4. The method according to claim 1, characterized in that, The method of predicting the mass transfer efficiency distribution using the stirring flow rate distribution coefficient includes: The stirring velocity distribution coefficient is used to stimulate the eddy current effect in the plating solution and generate a eddy current field intensity distribution. Capture mass transfer-enhancing vortex centers within the aforementioned eddy field intensity distribution; Energy focusing analysis is performed on the mass transfer enhancement vortex center to generate a mass transfer energy density map; The mass transfer efficiency distribution can be deduced from the mass transfer energy density diagram.

5. The method according to claim 1, characterized in that, The step of combining the high-risk process segment with the stable process segment to form a quality-balanced process pair includes: Risk energy release analysis is performed on the high-risk process section to generate a risk release spectrum; Risk absorption capacity is extracted from the stable process section to construct an absorption capacity pool; The balance threshold is obtained by performing energy balance matching between the risk release spectrum and the absorption capacity pool; The hedging combination is executed according to the balance threshold to form a quality balance process pair.

6. The method according to claim 1, characterized in that, The process of identifying areas with excessive temperature gradients through the hotspot distribution includes: The heat source distribution density is obtained by inverting the heat source intensity using the aforementioned hotspot distribution. A gradient field reconstruction process is performed on the heat source distribution density to generate a temperature gradient field; A threshold cutting operation is performed in the temperature gradient field to divide the gradient levels; The gradient level is used to filter and identify areas with excessive temperature gradients.

7. The method according to claim 1, characterized in that, The step of matching the beat control parameters with the probabilistic optimization chain to generate dynamic production units includes: A coating growth rate monitoring network is constructed using the aforementioned cycle control parameters; The probability optimization chain is mapped to the coating growth rate monitoring network to obtain thickness monitoring points; Uniformity analysis of the thickness monitoring points was performed to determine thickness distribution clusters; Dynamic production units are generated by adjusting the current distribution according to the thickness distribution cluster.

8. The method according to claim 4, characterized in that, The process of generating a vortex field intensity distribution by stimulating the plating solution vortex effect through the stirring flow velocity distribution coefficient includes: The spiral flow mode of the plating solution is initiated using the aforementioned stirring velocity distribution coefficient; The spiral flow pattern is subjected to vortex amplification processing to form enhanced vortices; A superimposed field strength is generated by superimposing the field strength through the enhanced eddy current; The superimposed field strength is spatially distributed and mapped to generate an eddy current field strength distribution.

9. The method according to claim 5, characterized in that, The step of performing risk energy release analysis on the high-risk process section to generate a risk release spectrum includes: Perform cross-cycle energy traceability analysis on the high-risk process section to obtain energy accumulation data; The energy accumulation data is used to analyze the energy impact of each process element and generate an energy release intensity distribution. The process elements include current density fluctuation, temperature gradient change and plating solution composition shift. The energy release intensity distribution is subjected to spectral conversion processing to generate a risk release spectrum.

10. A big data analysis and optimization system for the production and operation of electroplating tanks, characterized in that, include: The data acquisition module is used to acquire process parameter data of the electroplating tank, determine the sequence of key control nodes based on the process parameter data, and establish an operation and maintenance status mapping table based on the sequence of key control nodes. The probability analysis module is used to perform probability backtracking processing on the operation and maintenance status mapping table to generate a process probability distribution tree, perform branch analysis on the process probability distribution tree to obtain a quality achievement probability value, use the quality achievement probability value to determine the main process family and the alternative process family, and generate a probability optimization chain based on the main process family and the alternative process family. The flow field optimization module is used to obtain the stirring flow rate distribution coefficient and the plating solution circulation complexity index from the probabilistic optimization chain, predict the mass transfer efficiency distribution through the stirring flow rate distribution coefficient, determine the flow field control cycle using the plating solution circulation complexity index, and generate process optimization configuration by combining the mass transfer efficiency distribution and the flow field control cycle. The risk balancing module is used to determine high-risk process segments and stable process segments using the probabilistic optimization chain, to combine the high-risk process segments and the stable process segments in a hedging combination to form a quality-balanced process pair, and to generate process coordination rules based on the quality-balanced process pair. The temperature control optimization module is used to perform temperature field analysis to determine hot spot distribution based on the process coordination rules and the process optimization configuration, identify areas with excessive temperature gradients through the hot spot distribution, extract heat dissipation optimization factors from the areas with excessive temperature gradients, and use the heat dissipation optimization factors to generate cycle control parameters. The dynamic execution module is used to match the cycle control parameters with the probability optimization chain to generate dynamic production units, and to use the dynamic production units to generate flexible production processes, thereby completing big data analysis and optimization of electroplating tank production and maintenance.

Images