A management method for electrolytic copper additive production equipment

Abstract

The present invention discloses a management method for electrolytic copper additive production equipment. The electrolytic cell and electrolyte data are collected through a monitoring module, and a prediction model based on random forest is constructed in combination with product detection defect information. The model integrates the linear change characteristics of the electrolytic cell and electrolyte monitoring data, uses the CART algorithm to independently train decision trees, and predicts production defects through a voting mechanism. In real-time production, the pre-processed monitoring data is imported into the model to predict defects, and the defect categories are analyzed through the prediction results and the additive ratio is dynamically adjusted to achieve real-time optimization and adjustment of electrolytic copper production. Through the present invention, multi-dimensional real-time monitoring and analysis of electrolytic copper production and precise quality prediction can be achieved, dynamic analysis and regulation of additive schemes can be achieved, and production efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the field of processing and production data analysis, and more specifically, to a management method for electrolytic copper additive production equipment. Background Art

[0002] In the electrolytic copper production process, precise control of additives is crucial for improving copper quality. Traditional additive control methods suffer from low precision and slow response, making them unable to meet the high-quality product demands of modern electrolytic copper production. Furthermore, existing technologies lack sufficient mining and predictive analysis of electrolytic copper monitoring data, relying primarily on manual experience to control and analyze the production process, which is not very practical. Therefore, it is particularly important to develop a comprehensive management method for electrolytic copper additive production equipment, improve the analytical and predictive capabilities of production monitoring, and achieve high-precision, high-efficiency, and intelligent equipment management. Summary of the Invention

[0003] The present invention overcomes the defects of the prior art and proposes a management method for electrolytic copper additive production equipment.

[0004] A first aspect of the present invention provides a management method for electrolytic copper additive production equipment, comprising:

[0005] In electrolytic copper-based production equipment, set up a production monitoring module;

[0006] During N electrolytic copper production cycles, the production monitoring module collects electrolyte monitoring data and electrolytic cell monitoring data, and the copper product inspection module inspects the product and obtains processing defect information;

[0007] Constructing a prediction model based on random forests, performing linear change analysis on the electrolytic cell monitoring data based on the two dimensions of current density parameters and voltage parameters, and obtaining first change characteristic data; performing linear change analysis on the electrolyte monitoring data based on the two dimensions of electrolyte temperature parameters and electrolyte concentration parameters, and obtaining second change characteristic data;

[0008] In the prediction model, the electrolytic cell monitoring data, the first change characteristic data, and the processing defect information are used as the first sub-sample set. A first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data, and the processing defect information are used as the second sub-sample set. A second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism.

[0009] In a real-time production cycle, real-time monitoring data of the electrolytic cell and electrolyte is obtained and pre-processed and extracted to obtain the data to be predicted. The data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time;

[0010] Through production defect prediction information, the defect category of electrolytic copper is analyzed and the corresponding electrolytic copper additive ratio scheme is adjusted, a real-time additive scheme is generated, and real-time adjustments to electrolytic copper production are made through production equipment.

[0011] In this solution, a production monitoring module is set in the electrolytic copper-based production equipment, specifically:

[0012] The production monitoring module includes a variety of sensors for monitoring the operation of the electrolytic cell and the status of the electrolyte during the electrolytic copper production process.

[0013] In this solution, during the N electrolytic copper production cycles, the production monitoring module collects electrolyte monitoring data and electrolytic cell monitoring data, and the copper product detection module detects the product and obtains processing defect information, specifically:

[0014] During N electrolytic copper production cycles, different batches of electrolytic copper products are monitored in real time through the production monitoring module, and electrolyte monitoring data and electrolytic cell monitoring data are collected;

[0015] The copper product inspection module is used to inspect electrolytic copper products for defects, including surface smoothness, density, uniformity, and surface defects, and to obtain processing defect information.

[0016] In this solution, when constructing the prediction model based on random forest, the linear change analysis of the electrolytic cell monitoring data is performed based on the two dimensions of current density parameter and voltage parameter, and the first change characteristic data is obtained. The linear change analysis of the electrolyte monitoring data is performed based on the two dimensions of electrolyte temperature parameter and electrolyte concentration parameter, and the second change characteristic data is obtained, specifically:

[0017] Perform data cleaning and standardization preprocessing on electrolyzer monitoring data;

[0018] Taking an electrolytic copper production cycle as the analysis unit, multiple monitoring nodes within an electrolytic copper production cycle are obtained;

[0019] Based on the two dimensions of current density parameters and voltage parameters, a linear change analysis of the electrolytic cell monitoring data is performed. The analysis sampling points of the data are based on multiple monitoring nodes, and the first change characteristic data is formed through the linear change analysis;

[0020] Based on the two dimensions of electrolyte temperature parameters and electrolyte concentration parameters, the electrolyte monitoring data is subjected to linear change analysis. The analysis sampling points of the data are based on multiple monitoring nodes, and the second change characteristic data is formed through linear change analysis.

[0021] In this solution, in the prediction model, the electrolytic cell monitoring data, the first change characteristic data and the processing defect information are used as the first sub-sample set, and a first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data and the processing defect information are used as the second sub-sample set, and a second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism, specifically:

[0022] One electrolytic copper production cycle is used as the unit of analysis;

[0023] In the prediction model, the electrolytic cell monitoring data and the first change feature data are used as feature training data, and the processing defect information is used as the true prediction result to form a first sub-sample set;

[0024] Construct the first decision tree and initialize the maximum depth and minimum number of samples. Import the first subsample set into the first decision tree. Use the CART algorithm to perform feature selection and optimal split point calculation for each node of the decision tree. Use a cyclic recursive construction method until the decision tree reaches the maximum depth and obtains the complete first decision tree.

[0025] The electrolyte monitoring data and the second change feature data are used as feature training data, and the processing defect information is used as the true prediction result to form a second sub-sample set;

[0026] Construct a second decision tree and initialize the maximum depth and minimum number of samples. Import the second subsample set into the second decision tree. Use the CART algorithm to perform feature selection and optimal split point calculation for each node of the decision tree. Use a cyclic recursive construction method until the decision tree reaches the maximum depth and obtains a complete second decision tree.

[0027] Based on N electrolytic copper production cycles and corresponding electrolytic cell monitoring data and electrolyte monitoring data, a plurality of first sub-sample sets and a second sub-sample set are generated, and a plurality of first decision trees and a plurality of second decision trees are constructed;

[0028] The plurality of first decision trees are combined with the plurality of second decision trees to form a prediction model based on a random forest, in which prediction results are selected through a voting mechanism.

[0029] In this solution, in a real-time production cycle, real-time monitoring data of the electrolytic cell and electrolyte is obtained and preprocessed and extracted to obtain the data to be predicted. The data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time. Specifically:

[0030] In a real-time production cycle, obtain real-time monitoring data of the electrolytic cell and the electrolyte, and clean and standardize the data;

[0031] Set a time window and divide it into K time nodes based on preset intervals. In the real-time production cycle, based on the most recent time window, extract data from the electrolytic cell real-time monitoring data and the electrolyte real-time monitoring data and perform linear change analysis. The data analysis sampling points are based on K time nodes to obtain real-time detection change data.

[0032] The real-time monitoring data of the electrolytic cell, the real-time monitoring data of the electrolyte and the real-time detection change data are imported into the prediction model as the data to be predicted to predict processing defects and obtain production defect prediction information.

[0033] In this solution, the defect category analysis of electrolytic copper is carried out through production defect prediction information, and the corresponding electrolytic copper additive ratio scheme is adjusted to generate a real-time additive scheme, and the electrolytic copper production is adjusted in real time through production equipment, specifically:

[0034] Through production defect prediction information, we can analyze the impact of different types of defects and evaluate yield calculations;

[0035] Determine whether the yield rate is lower than the expected value. If so, analyze the defect type based on the production defect prediction information, analyze the additive ratio based on different defect types, and generate a real-time additive solution;

[0036] Based on real-time additive solutions, electrolytic copper production can be adjusted in real time through production equipment.

[0037] A second aspect of the present invention further provides a management system for electrolytic copper additive production equipment, the system comprising: a memory and a processor, wherein the memory includes a management program for the electrolytic copper additive production equipment, and when the management program for the electrolytic copper additive production equipment is executed by the processor, the following steps are implemented:

[0038] In electrolytic copper-based production equipment, set up a production monitoring module;

[0039] During N electrolytic copper production cycles, the production monitoring module collects electrolyte monitoring data and electrolytic cell monitoring data, and the copper product inspection module inspects the product and obtains processing defect information;

[0040] Constructing a prediction model based on random forests, performing linear change analysis on the electrolytic cell monitoring data based on the two dimensions of current density parameters and voltage parameters, and obtaining first change characteristic data; performing linear change analysis on the electrolyte monitoring data based on the two dimensions of electrolyte temperature parameters and electrolyte concentration parameters, and obtaining second change characteristic data;

[0041] In the prediction model, the electrolytic cell monitoring data, the first change characteristic data, and the processing defect information are used as the first sub-sample set. A first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data, and the processing defect information are used as the second sub-sample set. A second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism.

[0042] In a real-time production cycle, real-time monitoring data of the electrolytic cell and electrolyte is obtained and pre-processed and extracted to obtain the data to be predicted. The data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time;

[0043] Through production defect prediction information, the defect category of electrolytic copper is analyzed and the corresponding electrolytic copper additive ratio scheme is adjusted, a real-time additive scheme is generated, and real-time adjustments to electrolytic copper production are made through production equipment.

[0044] The third aspect of the present invention also provides a computer-readable storage medium, which includes a management program for electrolytic copper additive production equipment. When the management program for electrolytic copper additive production equipment is executed by a processor, the steps of the management method for electrolytic copper additive production equipment as described in any one of the above items are implemented.

[0045] The present invention discloses a management method for electrolytic copper additive production equipment, which collects electrolytic cell and electrolyte data through a monitoring module, combines product detection defect information, and constructs a prediction model based on random forest. The model integrates the linear change characteristics of the electrolytic cell and electrolyte monitoring data, uses the CART algorithm to independently train decision trees, and predicts production defects through a voting mechanism. In real-time production, the preprocessed monitoring data is imported into the model to predict defects, and the defect categories are analyzed through the prediction results and the additive ratio is dynamically adjusted to achieve real-time optimization and adjustment of electrolytic copper production. Through the present invention, multi-dimensional real-time monitoring and analysis of electrolytic copper production and precise quality prediction can be achieved, dynamic analysis and regulation of additive schemes can be achieved, and production efficiency can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 A flow chart showing a method for managing electrolytic copper additive production equipment according to the present invention is provided;

[0047] Figure 2 A block diagram of a management system for electrolytic copper additive production equipment according to the present invention is shown. DETAILED DESCRIPTION

[0048] In order to more clearly understand the above-mentioned objects, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that, in the absence of conflict, the embodiments of the present application and the features therein can be combined with each other.

[0049] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present invention is not limited to the specific embodiments disclosed below.

[0050] Figure 1 A flow chart showing a method for managing electrolytic copper additive production equipment according to the present invention is shown.

[0051] like Figure 1 As shown, the first aspect of the present invention provides a management method for electrolytic copper additive production equipment, comprising:

[0052] S102, setting a production monitoring module in an electrolytic copper-based production device;

[0053] S104, during N electrolytic copper production cycles, collecting electrolyte monitoring data and electrolytic cell monitoring data through the production monitoring module, and inspecting the product and obtaining processing defect information through the copper product inspection module;

[0054] S106, constructing a prediction model based on a random forest, performing a linear change analysis on the electrolytic cell monitoring data based on the two dimensions of current density parameter and voltage parameter, and obtaining first change characteristic data, and performing a linear change analysis on the electrolyte monitoring data based on the two dimensions of electrolyte temperature parameter and electrolyte concentration parameter, and obtaining second change characteristic data;

[0055] S108, in the prediction model, using the electrolytic cell monitoring data, the first change characteristic data, and the processing defect information as a first subsample set, independently constructing and training a first decision tree based on the CART algorithm; using the electrolyte monitoring data, the second change characteristic data, and the processing defect information as a second subsample set, independently constructing and training a second decision tree based on the CART algorithm; combining the first decision tree and the second decision tree to form a prediction model; in the prediction model, selecting a prediction result through a voting mechanism;

[0056] S110, in a real-time production cycle, obtaining real-time monitoring data of the electrolytic cell and electrolyte, performing preprocessing and data extraction, obtaining data to be predicted, importing the data to be predicted into a prediction model to perform result prediction, and generating production defect prediction information in real time;

[0057] S112, using the production defect prediction information, conduct defect category analysis on the electrolytic copper and adjust the corresponding electrolytic copper additive ratio scheme, generate a real-time additive scheme, and make real-time adjustments to the electrolytic copper production through production equipment.

[0058] According to an embodiment of the present invention, in the electrolytic copper-based production equipment, a production monitoring module is set, specifically:

[0059] The production monitoring module includes a variety of sensors for monitoring the operation of the electrolytic cell and the status of the electrolyte during the electrolytic copper production process.

[0060] It should be noted that electrolyte status monitoring includes monitoring of parameters such as additive liquid level, electrolyte temperature, and concentration. The concentration is generally the concentration of liquids such as copper sulfate. Electrolytic cell operation monitoring includes monitoring of parameters such as current intensity and voltage.

[0061] According to an embodiment of the present invention, during N electrolytic copper production cycles, electrolyte monitoring data and electrolytic cell monitoring data are collected through the production monitoring module, and the copper product detection module is used to detect the product and obtain processing defect information, specifically:

[0062] During N electrolytic copper production cycles, different batches of electrolytic copper products are monitored in real time through the production monitoring module, and electrolyte monitoring data and electrolytic cell monitoring data are collected;

[0063] The copper product inspection module is used to inspect electrolytic copper products for defects, including surface smoothness, density, uniformity, and surface defects, and to obtain processing defect information.

[0064] It should be noted that the production equipment includes all relevant equipment for electrolytic copper production, including electrolytic cells, electrolyte dosing equipment, production monitoring modules, copper product detection modules, etc. The copper product detection module includes corresponding product detection equipment to perform defect analysis on copper finished products. The electrolyte monitoring data includes monitoring data such as additive liquid level, electrolyte temperature, concentration and other parameters, and the electrolytic cell monitoring data includes data such as current intensity and voltage. In N electrolytic copper production cycles, each cycle corresponds to a batch of production process, realizing production process monitoring and defect detection for multiple batches of products, and performing feature analysis on the collected monitoring data to explore the correlation between the electrolytic process data and product defects of copper products.

[0065] According to an embodiment of the present invention, when constructing the prediction model based on random forest, the linear change analysis of the electrolytic cell monitoring data is performed based on the two dimensions of current density parameter and voltage parameter, and the first change characteristic data is obtained. The linear change analysis of the electrolyte monitoring data is performed based on the two dimensions of electrolyte temperature parameter and electrolyte concentration parameter, and the second change characteristic data is obtained, specifically:

[0066] Perform data cleaning and standardization preprocessing on electrolyzer monitoring data;

[0067] Taking an electrolytic copper production cycle as the analysis unit, multiple monitoring nodes within an electrolytic copper production cycle are obtained;

[0068] Based on the two dimensions of current density parameters and voltage parameters, a linear change analysis of the electrolytic cell monitoring data is performed. The analysis sampling points of the data are based on multiple monitoring nodes, and the first change characteristic data is formed through the linear change analysis;

[0069] Based on the two dimensions of electrolyte temperature parameters and electrolyte concentration parameters, the electrolyte monitoring data is subjected to linear change analysis. The analysis sampling points of the data are based on multiple monitoring nodes, and the second change characteristic data is formed through linear change analysis.

[0070] It should be noted that the prediction model includes multiple decision trees. In an embodiment of the present invention, two decision trees are constructed, namely a first decision tree and a second decision tree. Through the random forest algorithm, based on repeated extraction of subsample sets, each decision tree can be constructed multiple times to achieve data prediction based on multi-dimensional feature analysis and improve prediction accuracy. The first change feature data includes the change characteristics of the current density parameter and the voltage parameter. The second change feature data includes the change characteristics of the electrolyte temperature parameter and the electrolyte concentration parameter.

[0071] It’s worth mentioning that random forest is an ensemble learning algorithm that improves classification and prediction accuracy by constructing multiple decision trees and combining their prediction results. In this paper, the multi-decision tree feature of random forest is utilized to construct decision trees for data of different dimensions for comprehensive prediction analysis.

[0072] The electrolyte monitoring data and electrolytic cell monitoring data include data collected under multiple cycles. When performing feature analysis and building a decision tree, data features can be extracted based on different cycles to form feature data change characteristics corresponding to different production cycles.

[0073] According to an embodiment of the present invention, in the prediction model, the electrolytic cell monitoring data, the first change characteristic data, and the processing defect information are used as a first subsample set, and a first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data, and the processing defect information are used as a second subsample set, and a second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism, specifically:

[0074] One electrolytic copper production cycle is used as the unit of analysis;

[0075] In the prediction model, the electrolytic cell monitoring data and the first change feature data are used as feature training data, and the processing defect information is used as the true prediction result to form a first sub-sample set;

[0076] Construct the first decision tree and initialize the maximum depth and minimum number of samples. Import the first subsample set into the first decision tree. Use the CART algorithm to perform feature selection and optimal split point calculation for each node of the decision tree. Use a cyclic recursive construction method until the decision tree reaches the maximum depth and obtains the complete first decision tree.

[0077] The electrolyte monitoring data and the second change feature data are used as feature training data, and the processing defect information is used as the true prediction result to form a second sub-sample set;

[0078] Construct a second decision tree and initialize the maximum depth and minimum number of samples. Import the second subsample set into the second decision tree. Use the CART algorithm to perform feature selection and optimal split point calculation for each node of the decision tree. Use a cyclic recursive construction method until the decision tree reaches the maximum depth and obtains a complete second decision tree.

[0079] Based on N electrolytic copper production cycles and corresponding electrolytic cell monitoring data and electrolyte monitoring data, a plurality of first sub-sample sets and a second sub-sample set are generated, and a plurality of first decision trees and a plurality of second decision trees are constructed;

[0080] The plurality of first decision trees are combined with the plurality of second decision trees to form a prediction model based on a random forest, in which prediction results are selected through a voting mechanism.

[0081] It should be noted that the first sub-sample set includes the electrolytic cell monitoring data, the first change characteristic data and the processing defect information. The minimum number of samples is the minimum number of samples required for a leaf node.

[0082] It is worth mentioning that in the traditional production applications of electrolytic copper, there is a lack of data mining analysis of the electrolytic copper monitoring process and dynamic prediction analysis of the electrolytic copper production quality. At the same time, there is a lack of multi-dimensional analysis of monitoring data, and analysis is often based on unilateral data, such as only focusing on reaction rate or reaction temperature. In production regulation and finished product testing, production regulation is generally based on manual experience, and production adjustment and additive analysis are carried out based on manual experience analysis of production test results. This has poor practicality and makes it difficult to achieve effective analysis of production laws, resulting in low production efficiency and benefits.

[0083] In the present invention, a prediction model based on random forests is constructed. The prediction model specifically includes two types of decision trees. The first type is constructed based on the data characteristics of electrolytic cell monitoring data (current density, voltage, etc.). The purpose is to analyze the relationship between the changing patterns of electrolytic cell monitoring data and the defects of the final copper product. The second type is constructed based on the data characteristics of electrolyte monitoring data (electrolyte temperature, concentration, etc.). The purpose is to analyze the relationship between the changing patterns of electrolyte monitoring data and the defects of the final copper product. The model learns the changing patterns of the monitoring data and simultaneously learns and predicts the correlation between the finished product defects and the monitoring data. Multiple decision trees are combined based on random forests to form a prediction model. Based on a method with replacement, multiple subsample sets can be obtained from the monitoring data to construct decision trees, achieving diversity in the two types of decision trees. In the prediction model, two decision trees can be trained, and the prediction results can be compared with the real verification data to optimize the parameters of the decision trees, and finally a product prediction model that can combine the two dimensions of electrolyte monitoring data and electrolytic cell monitoring data is obtained. This model can be comprehensively applied to the dynamic prediction and evaluation of the quality of electrolytic copper products, and realize multi-dimensional real-time monitoring and analysis of electrolytic copper and precise product prediction.

[0084] In the prediction model, the prediction results of multiple decision trees are used to select the final prediction information through a voting mechanism.

[0085] According to an embodiment of the present invention, in a real-time production cycle, real-time monitoring data of the electrolytic cell and the electrolyte is obtained and preprocessed and data extracted to obtain data to be predicted. The data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time. Specifically,

[0086] In a real-time production cycle, obtain real-time monitoring data of the electrolytic cell and the electrolyte, and clean and standardize the data;

[0087] Set a time window and divide it into K time nodes based on preset intervals. In the real-time production cycle, based on the most recent time window, extract data from the electrolytic cell real-time monitoring data and the electrolyte real-time monitoring data and perform linear change analysis. The data analysis sampling points are based on K time nodes to obtain real-time detection change data.

[0088] The real-time monitoring data of the electrolytic cell, the real-time monitoring data of the electrolyte and the real-time detection change data are imported into the prediction model as the data to be predicted to predict processing defects and obtain production defect prediction information.

[0089] It should be noted that the time window is a set time length. The size of K is related to the preset interval. The electrolytic cell and electrolyte real-time monitoring data includes electrolytic cell real-time monitoring data and electrolyte real-time monitoring data.

[0090] During the real-time production cycle, the real-time monitoring data of the electrolytic cell and the electrolyte are stored in real time. As the production process continues, the amount of monitoring data continues to increase. In the present invention, when performing actual predictive analysis, a corresponding time window is set to intercept data within the most recent preset time period for change analysis. Based on the change characteristics, a predictive model is imported to predict the processing results, thereby achieving real-time analysis and prediction of data at relatively recent time points.

[0091] According to an embodiment of the present invention, the defect category analysis of electrolytic copper is performed based on production defect prediction information, and the corresponding electrolytic copper additive ratio scheme is adjusted to generate a real-time additive scheme, and the electrolytic copper production is adjusted in real time through production equipment, specifically:

[0092] Through production defect prediction information, we can analyze the impact of different types of defects and evaluate yield calculations;

[0093] Determine whether the yield rate is lower than the expected value. If so, analyze the defect type based on the production defect prediction information, analyze the additive ratio based on different defect types, and generate a real-time additive solution;

[0094] Based on real-time additive solutions, electrolytic copper production can be adjusted in real time through production equipment.

[0095] It should be noted that the expected value is user-set and varies across different electrolytic copper products. In this invention, product defect information is dynamically predicted based on real-time data within a time window. The defect prediction results are used to analyze different defect types and adjust the additive ratios. For example, if fine copper crystals and an uneven surface are predicted, the appropriate glue (including gelatin, bone glue, etc.) can be added. If the copper surface is predicted to be rough, non-dense, or too low, the product can be improved by adding thiourea. Production defect prediction information includes smoothness, density, uniformity, surface defects, and other predictions, as well as information on the degree of defects and the range of defect quantities for different defect types.

[0096] According to an embodiment of the present invention, the further embodiment includes:

[0097] In a real-time production cycle, based on the real-time additive plan, the electrolytic copper production additives are replenished in real time using the periodic replenishment method;

[0098] Dynamically obtain corresponding data from the real-time monitoring data of the electrolytic cell and electrolyte through a recent time window and perform defect prediction evaluation based on the prediction model to obtain the current defect prediction information;

[0099] Compare the current defect prediction information with the production defect prediction information of the previous time window;

[0100] Information comparison includes defect category comparison, defect degree comparison, and defect prediction quantity comparison. The degree of improvement of current production defects is judged through comparative analysis, and real-time production early warning analysis of production equipment is conducted based on the degree of improvement.

[0101] It should be noted that the degree of improvement of the predicted defects can reflect whether the monitored changes of electrolytic copper are tending to expectations, and based on this, accurate dynamic prediction and dynamic early warning of the production process can be achieved, reducing the manual analysis process and improving the informatization level of electrolytic copper production early warning.

[0102] Figure 2 A block diagram of a management system for electrolytic copper additive production equipment according to the present invention is shown.

[0103] A second aspect of the present invention further provides a management system 2 for electrolytic copper additive production equipment, the system comprising: a memory 21 and a processor 22, wherein the memory 21 includes a management program for the electrolytic copper additive production equipment, and when the management program for the electrolytic copper additive production equipment is executed by the processor 22, the following steps are implemented:

[0104] In electrolytic copper-based production equipment, set up a production monitoring module;

[0105] During N electrolytic copper production cycles, the production monitoring module collects electrolyte monitoring data and electrolytic cell monitoring data, and the copper product inspection module inspects the product and obtains processing defect information;

[0106] Constructing a prediction model based on random forests, performing linear change analysis on the electrolytic cell monitoring data based on the two dimensions of current density parameters and voltage parameters, and obtaining first change characteristic data; performing linear change analysis on the electrolyte monitoring data based on the two dimensions of electrolyte temperature parameters and electrolyte concentration parameters, and obtaining second change characteristic data;

[0107] In the prediction model, the electrolytic cell monitoring data, the first change characteristic data, and the processing defect information are used as the first sub-sample set. A first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data, and the processing defect information are used as the second sub-sample set. A second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism.

[0108] In a real-time production cycle, real-time monitoring data of the electrolytic cell and electrolyte is obtained and pre-processed and extracted to obtain the data to be predicted. The data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time;

[0109] Through production defect prediction information, the defect category of electrolytic copper is analyzed and the corresponding electrolytic copper additive ratio scheme is adjusted, a real-time additive scheme is generated, and real-time adjustments to electrolytic copper production are made through production equipment.

[0110] According to an embodiment of the present invention, in the electrolytic copper-based production equipment, a production monitoring module is set, specifically:

[0111] The production monitoring module includes a variety of sensors for monitoring the operation of the electrolytic cell and the status of the electrolyte during the electrolytic copper production process.

[0112] It should be noted that electrolyte status monitoring includes monitoring of parameters such as additive liquid level, electrolyte temperature, and concentration. The concentration is generally the concentration of liquids such as copper sulfate. Electrolytic cell operation monitoring includes monitoring of parameters such as current intensity and voltage.

[0113] According to an embodiment of the present invention, during N electrolytic copper production cycles, electrolyte monitoring data and electrolytic cell monitoring data are collected through the production monitoring module, and the copper product detection module is used to detect the product and obtain processing defect information, specifically:

[0114] During N electrolytic copper production cycles, different batches of electrolytic copper products are monitored in real time through the production monitoring module, and electrolyte monitoring data and electrolytic cell monitoring data are collected;

[0115] The copper product inspection module is used to inspect electrolytic copper products for defects, including surface smoothness, density, uniformity, and surface defects, and to obtain processing defect information.

[0116] It should be noted that the production equipment includes all relevant equipment for electrolytic copper production, including electrolytic cells, electrolyte dosing equipment, production monitoring modules, copper product detection modules, etc. The copper product detection module includes corresponding product detection equipment to perform defect analysis on copper finished products. The electrolyte monitoring data includes monitoring data such as additive liquid level, electrolyte temperature, concentration and other parameters, and the electrolytic cell monitoring data includes data such as current intensity and voltage. In N electrolytic copper production cycles, each cycle corresponds to a batch of production process, realizing production process monitoring and defect detection for multiple batches of products, and performing feature analysis on the collected monitoring data to explore the correlation between the electrolytic process data and product defects of copper products.

[0117] According to an embodiment of the present invention, when constructing the prediction model based on random forest, the linear change analysis of the electrolytic cell monitoring data is performed based on the two dimensions of current density parameter and voltage parameter, and the first change characteristic data is obtained. The linear change analysis of the electrolyte monitoring data is performed based on the two dimensions of electrolyte temperature parameter and electrolyte concentration parameter, and the second change characteristic data is obtained, specifically:

[0118] Perform data cleaning and standardization preprocessing on electrolyzer monitoring data;

[0119] Taking an electrolytic copper production cycle as the analysis unit, multiple monitoring nodes within an electrolytic copper production cycle are obtained;

[0120] Based on the two dimensions of current density parameters and voltage parameters, a linear change analysis of the electrolytic cell monitoring data is performed. The analysis sampling points of the data are based on multiple monitoring nodes, and the first change characteristic data is formed through the linear change analysis;

[0121] Based on the two dimensions of electrolyte temperature parameters and electrolyte concentration parameters, the electrolyte monitoring data is subjected to linear change analysis. The analysis sampling points of the data are based on multiple monitoring nodes, and the second change characteristic data is formed through linear change analysis.

[0122] It should be noted that the prediction model includes multiple decision trees. In an embodiment of the present invention, two decision trees are constructed, namely a first decision tree and a second decision tree. Through the random forest algorithm, based on repeated extraction of subsample sets, each decision tree can be constructed multiple times to achieve data prediction based on multi-dimensional feature analysis and improve prediction accuracy. The first change feature data includes the change characteristics of the current density parameter and the voltage parameter. The second change feature data includes the change characteristics of the electrolyte temperature parameter and the electrolyte concentration parameter.

[0123] It’s worth mentioning that random forest is an ensemble learning algorithm that improves classification and prediction accuracy by constructing multiple decision trees and combining their prediction results. In this paper, the multi-decision tree feature of random forest is utilized to construct decision trees for data of different dimensions for comprehensive prediction analysis.

[0124] The electrolyte monitoring data and electrolytic cell monitoring data include data collected under multiple cycles. When performing feature analysis and building a decision tree, data features can be extracted based on different cycles to form feature data change characteristics corresponding to different production cycles.

[0125] According to an embodiment of the present invention, in the prediction model, the electrolytic cell monitoring data, the first change characteristic data, and the processing defect information are used as a first subsample set, and a first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data, and the processing defect information are used as a second subsample set, and a second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism, specifically:

[0126] One electrolytic copper production cycle is used as the unit of analysis;

[0127] In the prediction model, the electrolytic cell monitoring data and the first change feature data are used as feature training data, and the processing defect information is used as the true prediction result to form a first sub-sample set;

[0128] Construct the first decision tree and initialize the maximum depth and minimum number of samples. Import the first subsample set into the first decision tree. Use the CART algorithm to perform feature selection and optimal split point calculation for each node of the decision tree. Use a cyclic recursive construction method until the decision tree reaches the maximum depth and obtains the complete first decision tree.

[0129] The electrolyte monitoring data and the second change feature data are used as feature training data, and the processing defect information is used as the true prediction result to form a second sub-sample set;

[0130] Construct a second decision tree and initialize the maximum depth and minimum number of samples. Import the second subsample set into the second decision tree. Use the CART algorithm to perform feature selection and optimal split point calculation for each node of the decision tree. Use a cyclic recursive construction method until the decision tree reaches the maximum depth and obtains a complete second decision tree.

[0131] Based on N electrolytic copper production cycles and corresponding electrolytic cell monitoring data and electrolyte monitoring data, a plurality of first sub-sample sets and a second sub-sample set are generated, and a plurality of first decision trees and a plurality of second decision trees are constructed;

[0132] The plurality of first decision trees are combined with the plurality of second decision trees to form a prediction model based on a random forest, in which prediction results are selected through a voting mechanism.

[0133] It should be noted that the first sub-sample set includes the electrolytic cell monitoring data, the first change characteristic data and the processing defect information. The minimum number of samples is the minimum number of samples required for a leaf node.

[0134] It is worth mentioning that in the traditional production applications of electrolytic copper, there is a lack of data mining analysis of the electrolytic copper monitoring process and dynamic prediction analysis of the electrolytic copper production quality. At the same time, there is a lack of multi-dimensional analysis of monitoring data, and analysis is often based on unilateral data, such as only focusing on reaction rate or reaction temperature. In production regulation and finished product testing, production regulation is generally based on manual experience, and production adjustment and additive analysis are carried out based on manual experience analysis of production test results. This has poor practicality and makes it difficult to achieve effective analysis of production laws, resulting in low production efficiency and benefits.

[0135] In the present invention, a prediction model based on random forests is constructed. The prediction model specifically includes two types of decision trees. The first type is constructed based on the data characteristics of electrolytic cell monitoring data (current density, voltage, etc.). The purpose is to analyze the relationship between the changing patterns of electrolytic cell monitoring data and the defects of the final copper product. The second type is constructed based on the data characteristics of electrolyte monitoring data (electrolyte temperature, concentration, etc.). The purpose is to analyze the relationship between the changing patterns of electrolyte monitoring data and the defects of the final copper product. The model learns the changing patterns of the monitoring data and simultaneously learns and predicts the correlation between the finished product defects and the monitoring data. Multiple decision trees are combined based on random forests to form a prediction model. Based on a method with replacement, multiple subsample sets can be obtained from the monitoring data to construct decision trees, achieving diversity in the two types of decision trees. In the prediction model, two decision trees can be trained, and the prediction results can be compared with the real verification data to optimize the parameters of the decision trees, and finally a product prediction model that can combine the two dimensions of electrolyte monitoring data and electrolytic cell monitoring data is obtained. This model can be comprehensively applied to the dynamic prediction and evaluation of the quality of electrolytic copper products, and realize multi-dimensional real-time monitoring and analysis of electrolytic copper and precise product prediction.

[0136] In the prediction model, the prediction results of multiple decision trees are used to select the final prediction information through a voting mechanism.

[0137] According to an embodiment of the present invention, in a real-time production cycle, real-time monitoring data of the electrolytic cell and the electrolyte is obtained and preprocessed and data extracted to obtain data to be predicted. The data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time. Specifically,

[0138] In a real-time production cycle, obtain real-time monitoring data of the electrolytic cell and the electrolyte, and clean and standardize the data;

[0139] Set a time window and divide it into K time nodes based on preset intervals. In the real-time production cycle, based on the most recent time window, extract data from the electrolytic cell real-time monitoring data and the electrolyte real-time monitoring data and perform linear change analysis. The data analysis sampling points are based on K time nodes to obtain real-time detection change data.

[0140] The real-time monitoring data of the electrolytic cell, the real-time monitoring data of the electrolyte and the real-time detection change data are imported into the prediction model as the data to be predicted to predict processing defects and obtain production defect prediction information.

[0141] It should be noted that the time window is a set time length. The size of K is related to the preset interval. The electrolytic cell and electrolyte real-time monitoring data includes electrolytic cell real-time monitoring data and electrolyte real-time monitoring data.

[0142] During the real-time production cycle, the real-time monitoring data of the electrolytic cell and the electrolyte are stored in real time. As the production process continues, the amount of monitoring data continues to increase. In the present invention, when performing actual predictive analysis, a corresponding time window is set to intercept data within the most recent preset time period for change analysis. Based on the change characteristics, a predictive model is imported to predict the processing results, thereby achieving real-time analysis and prediction of data at relatively recent time points.

[0143] According to an embodiment of the present invention, the defect category analysis of electrolytic copper is performed based on production defect prediction information, and the corresponding electrolytic copper additive ratio scheme is adjusted to generate a real-time additive scheme, and the electrolytic copper production is adjusted in real time through production equipment, specifically:

[0144] Through production defect prediction information, we can analyze the impact of different types of defects and evaluate yield calculations;

[0145] Determine whether the yield rate is lower than the expected value. If so, analyze the defect type based on the production defect prediction information, analyze the additive ratio based on different defect types, and generate a real-time additive solution;

[0146] Based on real-time additive solutions, electrolytic copper production can be adjusted in real time through production equipment.

[0147] It should be noted that the expected value is user-set and varies across different electrolytic copper products. In this invention, product defect information is dynamically predicted based on real-time data within a time window. The defect prediction results are used to analyze different defect types and adjust the additive ratios. For example, if fine copper crystals and an uneven surface are predicted, the appropriate glue (including gelatin, bone glue, etc.) can be added. If the copper surface is predicted to be rough, non-dense, or too low, the product can be improved by adding thiourea. Production defect prediction information includes smoothness, density, uniformity, surface defects, and other predictions, as well as information on the degree of defects and the range of defect quantities for different defect types.

[0148] The third aspect of the present invention also provides a computer-readable storage medium, which includes a management program for electrolytic copper additive production equipment. When the management program for electrolytic copper additive production equipment is executed by a processor, the steps of the management method for electrolytic copper additive production equipment as described in any one of the above items are implemented.

[0149] The present invention discloses a management method for electrolytic copper additive production equipment, which collects electrolytic cell and electrolyte data through a monitoring module, combines product detection defect information, and constructs a prediction model based on random forest. The model integrates the linear change characteristics of the electrolytic cell and electrolyte monitoring data, uses the CART algorithm to independently train decision trees, and predicts production defects through a voting mechanism. In real-time production, the preprocessed monitoring data is imported into the model to predict defects, and the defect categories are analyzed through the prediction results and the additive ratio is dynamically adjusted to achieve real-time optimization and adjustment of electrolytic copper production. Through the present invention, multi-dimensional real-time monitoring and analysis of electrolytic copper production and precise quality prediction can be achieved, dynamic analysis and regulation of additive schemes can be achieved, and production efficiency can be improved.

[0150] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of the devices or units can be electrical, mechanical or other forms.

[0151] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed across multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.

[0152] In addition, all functional units in the embodiments of the present invention may be integrated into one processing unit, or each unit may be separately used as a unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0153] Those skilled in the art will appreciate that all or part of the steps of the above-mentioned method embodiments may be implemented by hardware associated with program instructions, and the aforementioned program may be stored in a computer-readable storage medium. When the program is executed, the program executes the steps of the above-mentioned method embodiments. The aforementioned storage medium includes various media that can store program codes, such as mobile storage devices, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical disks.

[0154] Alternatively, if the integrated units described above are implemented as software modules and sold or used as standalone products, they can also be stored on a computer-readable storage medium. Based on this understanding, the technical solutions of the embodiments of the present invention, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product, stored on a storage medium, includes instructions for enabling a computer device (such as a personal computer, server, or network device) to execute all or part of the methods described in various embodiments of the present invention. The aforementioned storage media include various media capable of storing program code, such as removable storage devices, ROM, RAM, magnetic disks, or optical disks.

[0155] The above description is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any technician familiar with this technical field can easily think of changes or replacements within the technical scope disclosed by the present invention, which should be covered by the scope of protection of the present invention.

Claims

1. A management method for electrolytic copper additive production equipment, characterized in that: include: In electrolytic copper-based production equipment, set up a production monitoring module; During N electrolytic copper production cycles, the production monitoring module collects electrolyte monitoring data and electrolytic cell monitoring data, and the copper product detection module detects the product and obtains processing defect information; Construct a prediction model based on random forest, perform linear change analysis on the electrolytic cell monitoring data based on the two dimensions of current density parameter and voltage parameter, and obtain first change characteristic data, perform linear change analysis on the electrolyte monitoring data based on the two dimensions of electrolyte temperature parameter and electrolyte concentration parameter, and obtain second change characteristic data; In the prediction model, the electrolytic cell monitoring data, the first change characteristic data and the processing defect information are used as the first sub-sample set, and the first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data and the processing defect information are used as the second sub-sample set, and the second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism; In a real-time production cycle, real-time monitoring data of the electrolytic cell and electrolyte are obtained and preprocessed and extracted to obtain the data to be predicted. The data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time; Through the production defect prediction information, the defect category of electrolytic copper is analyzed and the corresponding electrolytic copper additive ratio scheme is adjusted, a real-time additive scheme is generated, and the electrolytic copper production is adjusted in real time through production equipment.

2. The management method of electrolytic copper additive production equipment according to claim 1, characterized in that: In the production equipment based on electrolytic copper, a production monitoring module is set, specifically: The production monitoring module includes a variety of sensors for monitoring the operation of the electrolytic cell and the state of the electrolyte during the electrolytic copper production process.

3. The management method of electrolytic copper additive production equipment according to claim 2, characterized in that: In the N electrolytic copper production cycles, the production monitoring module is used to collect electrolyte monitoring data and electrolytic cell monitoring data, and the copper product detection module is used to detect the product and obtain processing defect information, specifically: During N electrolytic copper production cycles, different batches of electrolytic copper products are monitored in real time through the production monitoring module, and electrolyte monitoring data and electrolytic cell monitoring data are collected; Through the copper product detection module, the electrolytic copper products are inspected for defects, including surface smoothness, density, uniformity, surface defects, and processing defect information is obtained.

4. The management method of electrolytic copper additive production equipment according to claim 3, characterized in that: When constructing the prediction model based on random forest, the linear change analysis of the electrolytic cell monitoring data is performed based on the two dimensions of current density parameter and voltage parameter, and the first change characteristic data is obtained. The linear change analysis of the electrolyte monitoring data is performed based on the two dimensions of electrolyte temperature parameter and electrolyte concentration parameter, and the second change characteristic data is obtained, which is specifically: Clean the electrolyzer monitoring data and perform standardization preprocessing; Taking an electrolytic copper production cycle as the analysis unit, multiple monitoring nodes within an electrolytic copper production cycle are obtained; Based on the two dimensions of current density parameter and voltage parameter, a linear change analysis is performed on the electrolytic cell monitoring data. The data analysis sampling points are based on multiple monitoring nodes, and the first change characteristic data is formed through the linear change analysis. Based on the two dimensions of electrolyte temperature parameters and electrolyte concentration parameters, the electrolyte monitoring data is analyzed for linear changes. The analysis sampling points of the data are based on multiple monitoring nodes, and the second change characteristic data is formed through linear change analysis.

5. The management method of electrolytic copper additive production equipment according to claim 4, characterized in that: In the prediction model, the electrolytic cell monitoring data, the first change characteristic data and the processing defect information are used as the first sub-sample set, and the first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data and the processing defect information are used as the second sub-sample set, and the second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism, specifically: One electrolytic copper production cycle is taken as the unit of analysis; In the prediction model, the electrolytic cell monitoring data and the first change characteristic data are used as characteristic training data, and the processing defect information is used as the true prediction result to form a first sub-sample set; Construct the first decision tree and initialize the maximum depth and minimum number of samples, import the first sub-sample set into the first decision tree, perform feature selection and optimal segmentation point calculation on each node of the decision tree through the CART algorithm, and adopt a cyclic recursive construction method until the decision tree reaches the maximum depth and obtains a complete first decision tree; The electrolyte monitoring data and the second change characteristic data are used as characteristic training data, and the processing defect information is used as a true prediction result to form a second sub-sample set; Construct a second decision tree and initialize the maximum depth and minimum number of samples, import the second sub-sample set into the second decision tree, perform feature selection and optimal segmentation point calculation on each node of the decision tree through the CART algorithm, and adopt a cyclic recursive construction method until the decision tree reaches the maximum depth and obtains a complete second decision tree; Based on N electrolytic copper production cycles and corresponding electrolytic cell monitoring data and electrolyte monitoring data, a plurality of first sub-sample sets and a second sub-sample set are generated, and a plurality of first decision trees and a plurality of second decision trees are constructed; The plurality of first decision trees are combined with the plurality of second decision trees to form a prediction model based on a random forest, in which a prediction result is selected through a voting mechanism.

6. The management method of electrolytic copper additive production equipment according to claim 5, characterized in that: In a real-time production cycle, real-time monitoring data of the electrolytic cell and the electrolyte are obtained and preprocessed and extracted to obtain the data to be predicted, the data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time, specifically: In a real-time production cycle, obtain the real-time monitoring data of the electrolytic cell and the electrolyte, and clean and standardize the data; A time window is set, and the time window is divided into K time nodes based on a preset interval. In the real-time production cycle, based on the most recent time window, data is extracted from the real-time monitoring data of the electrolytic cell and the real-time monitoring data of the electrolyte, and linear change analysis is performed. The data analysis sampling points are based on K time nodes to obtain real-time detection change data; The real-time monitoring data of the electrolytic cell, the real-time monitoring data of the electrolyte and the real-time detection change data are imported into the prediction model as the data to be predicted to predict processing defects and obtain production defect prediction information.

7. The management method of electrolytic copper additive production equipment according to claim 6, characterized in that: The defect category analysis of electrolytic copper is carried out through production defect prediction information, and the corresponding electrolytic copper additive ratio scheme is adjusted to generate a real-time additive scheme, and the electrolytic copper production is adjusted in real time through production equipment, specifically: Through production defect prediction information, impact analysis and yield calculation evaluation of different types of defects are performed; Determine whether the yield is lower than the expected value. If so, analyze the defect type based on the production defect prediction information, analyze the additive ratio based on different defect types, and generate a real-time additive solution; Based on real-time additive solutions, electrolytic copper production is adjusted in real time through production equipment.

8. A management system for electrolytic copper additive production equipment, characterized in that: The system includes: a memory and a processor, wherein the memory includes a management program for electrolytic copper additive production equipment, and when the management program for electrolytic copper additive production equipment is executed by the processor, the following steps are implemented: In electrolytic copper-based production equipment, set up a production monitoring module; During N electrolytic copper production cycles, the production monitoring module collects electrolyte monitoring data and electrolytic cell monitoring data, and the copper product detection module detects the product and obtains processing defect information; Construct a prediction model based on random forest, perform linear change analysis on the electrolytic cell monitoring data based on the two dimensions of current density parameter and voltage parameter, and obtain first change characteristic data, perform linear change analysis on the electrolyte monitoring data based on the two dimensions of electrolyte temperature parameter and electrolyte concentration parameter, and obtain second change characteristic data; In the prediction model, the electrolytic cell monitoring data, the first change characteristic data and the processing defect information are used as the first sub-sample set, and the first decision tree is independently constructed and trained based on the CART algorithm. The electrolyte monitoring data, the second change characteristic data and the processing defect information are used as the second sub-sample set, and the second decision tree is independently constructed and trained based on the CART algorithm. The first decision tree and the second decision tree are combined to form a prediction model. In the prediction model, the prediction result is selected through a voting mechanism; In a real-time production cycle, real-time monitoring data of the electrolytic cell and electrolyte are obtained and preprocessed and extracted to obtain the data to be predicted. The data to be predicted is imported into the prediction model for result prediction, and production defect prediction information is generated in real time; Through the production defect prediction information, the defect category of electrolytic copper is analyzed and the corresponding electrolytic copper additive ratio scheme is adjusted, a real-time additive scheme is generated, and the electrolytic copper production is adjusted in real time through production equipment.

9. The management system for electrolytic copper additive production equipment according to claim 8, characterized in that: In the production equipment based on electrolytic copper, a production monitoring module is set, specifically: The production monitoring module includes a variety of sensors for monitoring the operation of the electrolytic cell and the state of the electrolyte in the electrolytic copper production process.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a management program for electrolytic copper additive production equipment. When the management program for electrolytic copper additive production equipment is executed by a processor, the steps of the management method for electrolytic copper additive production equipment according to any one of claims 1 to 7 are implemented.

Images