Method and system for performance prediction and process optimization of seven-system aluminum alloy PEO composite coating

Abstract

This application discloses a method and system for performance prediction and process optimization of a PEO composite coating of a seven-series aluminum alloy, relating to the field of electronic chemistry. The method includes obtaining a combination of process parameters to be tested for the composite coating, wherein the composite coating is a plasma electrolytic oxidation composite coating of a seven-series aluminum alloy synergistically reinforced with graphene and carbon nanotubes; inputting the combination of process parameters into a trained artificial neural network, an extreme gradient boosting model, and an integrated model of support vector regression and random forest regression to obtain corresponding performance parameter combinations; predicting the coating thickness and microhardness of the composite coating using the artificial neural network, predicting the friction coefficient and wear rate of the composite coating using the extreme gradient boosting model, and predicting the corrosion current density of the composite coating using the integrated model of support vector regression and random forest regression; the coating thickness, microhardness, friction coefficient, wear rate, and corrosion current density constitute the performance parameter combination of the composite coating, significantly improving prediction accuracy and efficiency.

Description

Technical Field

[0001] This application relates to the field of electronic chemistry, and in particular to a method and system for performance prediction and process optimization of PEO composite coatings for 7-series aluminum alloys. Background Technology

[0002] 7-series aluminum alloys (Al-Zn-Mg-Cu series, typical grades: 7075, 7050, 7055, 7085) are currently the strongest aluminum alloy series used in engineering applications. However, 7-series aluminum alloys have two significant engineering limitations that severely restrict their reliable application in highly corrosive environments: 1. High susceptibility to stress corrosion cracking (SCC); 2. Low surface hardness and insufficient wear resistance.

[0003] Plasma electrolytic oxidation (PEO) technology can grow ceramic coatings with α-Al₂O₃ and γ-Al₂O₃ as the main crystalline phases in situ on the surface of 700-series aluminum alloys, significantly improving surface hardness (600 HV–850 HV) and corrosion resistance while retaining the high strength of the substrate. Introducing graphene and carbon nanotube (CNT) hybrid carbon nanomaterials into the PEO electrolyte can prepare 700-series aluminum alloy composite PEO coatings with both self-lubricating and toughening functions. However, the above process involves more than ten strongly coupled process parameters, and the interaction of these parameters has a highly nonlinear effect on multiple key performance indicators. Traditional trial-and-error experimental methods are extremely inefficient, resulting in very high industrial-scale development costs for 700-series aluminum alloy composite PEO coatings. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for performance prediction and process optimization of PEO composite coatings for 7-series aluminum alloys, which significantly improves prediction accuracy and efficiency compared to traditional trial-and-error experimental methods.

[0005] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a method for predicting the performance of PEO composite coatings on 7-series aluminum alloys, including: The combination of process parameters to be tested for the composite coating is obtained. The composite coating is a plasma electrolytic oxidation composite coating of a seven-series aluminum alloy that is synergistically reinforced by graphene and carbon nanotubes.

[0006] The combination of process parameters to be tested is input into the trained artificial neural network, extreme gradient boosting model, and integrated support vector regression and random forest regression model to obtain the corresponding performance parameter combination.

[0007] Specifically, the coating thickness and microhardness of the composite coating are predicted by the artificial neural network, the friction coefficient and wear rate of the composite coating are predicted by the extreme gradient boosting model, and the corrosion current density of the composite coating is predicted by the integrated model of support vector regression and random forest regression. The coating thickness, microhardness, friction coefficient, wear rate and corrosion current density constitute the performance parameter combination of the composite coating.

[0008] Secondly, this application provides a process optimization method for PEO composite coating of 7-series aluminum alloys, including: Obtain the target performance parameter combination for the composite coating.

[0009] The NSGA-II multi-objective genetic algorithm is used to call the performance prediction method of the seven-series aluminum alloy PEO composite coating described in the first aspect to obtain the relationship between different combinations of process parameters and different combinations of performance parameters, and then to determine the target combination of process parameters that satisfies the target performance parameter combination.

[0010] Thirdly, this application provides a performance prediction and process optimization system for PEO composite coatings on 7-series aluminum alloys, including: The forward prediction module is used to execute the performance prediction method for the PEO composite coating of the seven-series aluminum alloy described in the first aspect to predict the combination of performance parameters of the composite coating under different combinations of process parameters to be tested.

[0011] The reverse optimization module is used to execute the process optimization method for the PEO composite coating of the seventh-series aluminum alloy described in the second aspect to determine the target process parameter combination when the composite coating meets the target performance parameter combination.

[0012] According to the specific embodiments provided in this application, the following technical effects are disclosed: This application provides a method and system for performance prediction and process optimization of PEO composite coatings on 7-series aluminum alloys. Different types of machine learning models are selected for different formation indices of the composite coating. Given a combination of process parameters, different types of machine learning models provide targeted predictions for the corresponding performance parameters. Among them, the coating thickness prediction model R... 2 ≥0.95; Predictive model for microhardness R 2 ≥0.94; Prediction model for friction coefficient R 2 ≥0.93; Wear rate prediction model R 2 ≥0.93; Integrated prediction model for corrosion current density R 2The relative error of experimental verification for all models is within ±6%, thus enabling high-precision performance prediction of PEO composite coatings on 7-series aluminum alloys. Compared to traditional trial-and-error experimental methods, this significantly improves prediction accuracy and efficiency. It can reduce the amount of verification experiments in the development of 7-series aluminum alloy PEO composite coating processes by no less than 75%, and shorten the process development cycle to 15%-20% of traditional trial-and-error experimental methods. Attached Figure Description

[0013] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0014] Figure 1 This is a flowchart illustrating a method for predicting the performance of a PEO composite coating on a 7-series aluminum alloy according to an embodiment of this application.

[0015] Figure 2 This is a flowchart illustrating a process optimization method for a PEO composite coating of a 7-series aluminum alloy according to an embodiment of this application. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] The performance prediction method for the PEO composite coating of 7-series aluminum alloys provided in this application embodiment is executed by a computer device, which can be a terminal device or a server. The terminal device includes host computers, laptops, and smartphones, etc., and the server can be a server cluster or a single server.

[0019] like Figure 1 As shown, the performance prediction method for the PEO composite coating of the 7-series aluminum alloy provided in this application embodiment includes steps S110 and S120.

[0020] Step S110: Obtain the test process parameter combination of the composite coating. The composite coating is a plasma electrolytic oxidation composite coating of a seven-series aluminum alloy with synergistic enhancement of graphene and carbon nanotubes.

[0021] Specifically, the combination of process parameters includes electrolyte formulation and forward current density J. + Negative current density J - Pulse frequency, forward duty cycle, processing time, total carbon addition, graphene to carbon nanotube mass ratio (G:CNT mass ratio), alloy grade, and matrix heat treatment state.

[0022] Preferably, the process parameter combination also includes charge density. Q=J + ·D·t·60 (A·s / cm) 2 ), positive and negative current ratio r=J + / J — Effective specific surface area of ​​carbon materials A eff And alloy Zn equivalent; alloy Zn equivalent is used to characterize the contribution of key elements affecting the performance of composite coatings in different alloy grades of VII series aluminum alloys. The specific expression of alloy Zn equivalent is: Zn eq =w Zn +5w Cu ,in, w Zn and w Cu These represent the mass fractions of Zn and Cu elements in VII series aluminum alloys, respectively. t For processing time; D The area is the electrode area.

[0023] For example, the alloy grades can be selected from 7075, 7055, 7085, and 7050, and the base heat treatment states can be selected from T6, T73, T76, and T7351. The electrolyte formulation is a quaternary system of NaAlO2, KOH, Na2SiO3, and EDTA-2Na, with the mass concentration of NaAlO2 ranging from 6 g / L to 12 g / L, the mass concentration of KOH ranging from 0.5 g / L to 2.0 g / L, the mass concentration of Na2SiO3 ranging from 1 g / L to 8 g / L, and the mass concentration of EDTA-2Na ranging from 1 g / L to 4 g / L. The forward current density ranges from 0.15 A / cm². 2 ~0.60A / cm 2The processing time ranges from 5 min to 30 min, the total carbon addition ranges from 0 g / L to 2.0 g / L, and the mass ratio of graphene to carbon nanotubes ranges from 0:1 to 4:0.

[0024] For example, the bipolar pulse uses a 40% positive duty cycle; the negative current density can be determined by combining the positive current density and the positive-negative current ratio; the pulse frequency is an adjustable process parameter, which is 1000Hz in a preferred embodiment.

[0025] Step S120 involves inputting the combination of process parameters to be tested into the trained artificial neural network, extreme gradient boosting model, and support vector regression and random forest regression ensemble model to obtain the corresponding performance parameter combination. Specifically, this includes: predicting the coating thickness and microhardness of the composite coating using the artificial neural network; predicting the friction coefficient and wear rate of the composite coating using the extreme gradient boosting model; and predicting the corrosion current density of the composite coating using the support vector regression and random forest regression ensemble model. The coating thickness, microhardness, friction coefficient, wear rate, and corrosion current density constitute the performance parameter combination of the composite coating.

[0026] In this step, a single model is not used to predict all performance parameters. Instead, different models are used to predict different performance parameters. A multi-algorithm ensemble prediction model is constructed based on the DMME framework. Artificial Neural Network (ANN) is used for coating thickness and microhardness, Extreme Gradient Boosting (XGBoost) is used for friction coefficient and wear rate, and an ensemble model of Support Vector Regression (SVR) and Random Forest Regression (RFR) is used for corrosion current density.

[0027] During model selection, hyperparameter tuning was performed using Bayesian optimization, and R-squared was performed using 5-fold cross-validation. 2 The mean is the objective function. The final result is: for coating thickness and hardness prediction, artificial neural networks (ANNs) perform best (R0). 2 ≥0.95); for predicting friction coefficient and wear rate, extreme gradient boosting (XGBoost) performs best (R 2 ≥0.93); For corrosion current density prediction, the ensemble model of Support Vector Regression (SVR) and Random Forest Regression (RFR) performs best (R 2 (≥0.94). The relative error of experimental validation for all models was within ±6%.

[0028] Specifically, the hyperparameters for ANN used to predict coating thickness are: hidden layer structure (64, 32), activation function ReLU, regularization coefficient α = 0.051, initial learning rate η0 = 0.0068, and early stopping mechanism; the hyperparameters for ANN used to predict microhardness are: hidden layer structure (64, 32, 16), regularization coefficient α = 0.058, and initial learning rate η0 = 0.0062; the hyperparameters for XGBoost used to predict friction coefficient are: learning rate η = 0.05, and maximum tree depth d. max =6, sample (row) subsampling rate subsample=0.80, feature (column) subsampling rate colsample_bytree=0.80, L2 regularization weight λ=1.3, number of trees n_estimators=320; the hyperparameters of SVR for corrosion current density prediction are: penalty coefficient / regularization strength C=13.21, insensitive loss threshold ε=0.548, kernel function γ='scale'; the hyperparameters of RFR are: number of decision trees n_estimators=120, maximum tree depth max_depth=15, minimum number of samples required for internal node splitting min_samples_split=3; the weighted ensemble weights of SVR and RFR are 0.55 and 0.45, respectively.

[0029] It should be noted that in this embodiment, the wear rate and corrosion current density need to be logarithmically transformed to base 10 so that they are on the same order of magnitude as other performance parameters.

[0030] In summary, this embodiment provides a machine learning-based method for predicting the performance of PEO composite coatings on 7-series aluminum alloys. Different types of machine learning models are selected for different formation indices of the composite coating. Given a combination of process parameters, different types of machine learning models provide targeted predictions for the corresponding performance parameters. Among them, the coating thickness prediction model R... 2 ≥0.95; Predictive model for microhardness R 2 ≥0.94; Prediction model for friction coefficient R 2 ≥0.93; Wear rate prediction model R 2 ≥0.93; Integrated prediction model for corrosion current density R 2 The relative error of experimental verification for all models is within ±6%, thus enabling high-precision performance prediction of PEO composite coatings on 7-series aluminum alloys. Compared to traditional trial-and-error experimental methods, this significantly improves prediction accuracy and efficiency. It can reduce the amount of verification experiments in the development of 7-series aluminum alloy PEO composite coating processes by no less than 75%, and shorten the process development cycle to 15%-20% of traditional trial-and-error experimental methods.

[0031] Based on the high-precision performance prediction of PEO composite coatings for 7-series aluminum alloys, this embodiment also provides a process optimization method for 7-series aluminum alloy PEO composite coatings, such as... Figure 2 As shown, it specifically includes steps S210 and S220.

[0032] Step S210: Obtain the target performance parameter combination of the composite coating.

[0033] Step S220: Using the NSGA-II multi-objective genetic algorithm, the performance prediction method for the PEO composite coating of the seven-series aluminum alloy provided in this embodiment is called to obtain the relationship between different combinations of process parameters and different combinations of performance parameters, and then the target combination of process parameters that meets the target performance parameter combination is determined.

[0034] Among them, the NSGA-II multi-objective genetic algorithm is a multi-objective optimization evolutionary algorithm used to find a set of "optimal compromise" solutions when multiple objectives conflict with each other. Since there are contradictions among the different performance parameters of the composite coating, if all different performance parameters are taken as objectives, the NSGA-II multi-objective genetic algorithm can find a set of "optimal compromise" performance parameter combinations by searching the Pareto optimal front.

[0035] For example, the target process parameter combination is as follows: alloy grade 7075, heat treatment state T73, electrolyte type I standard formula (NaAlO2 8g / L, KOH 1g / L, Na2SiO3 3g / L, EDTA-2Na 2g / L), and forward current density 0.35A / cm². 2 J - / J + =0.50, pulse frequency 1000Hz, forward duty cycle 40%, processing time 18min, total carbon addition 1.0g / L, and G:CNT mass ratio 1:1. The target performance parameter combination is: coating thickness 35.5μm, microhardness 718HV, coefficient of friction 0.248, and wear rate 0.62×10⁻⁶. -5 mm 3 / (N·m), corrosion current density 3.2×10 -7 A / cm 2 The experimental verification showed that the error was within ±5%. In this example, the various performance parameters were relatively balanced.

[0036] In this embodiment, the NSGA-II multi-objective genetic algorithm can also find a combination of target process parameters that meets preset performance requirements (target performance parameter combination).

[0037] The NSGA-II multi-objective genetic algorithm for finding target process parameter combinations is based on the performance prediction method for PEO composite coatings of 7-series aluminum alloys provided in this embodiment. During its operation, it needs to forward calculate the performance parameter combinations corresponding to different process parameter combinations, thereby traversing different possibilities to find the target process parameter combination corresponding to the target performance parameter combination.

[0038] In this embodiment, the process optimization method for the PEO composite coating of 7-series aluminum alloys may further include: Step S230: Quantitative analysis of the relationship between different combinations of process parameters and different combinations of performance parameters is conducted through SHAP (SHapley Additive exPlanations) value analysis, KDE (Kernel Density Estimation) joint distribution map and process response heatmap, to determine the influence of each process parameter on each performance parameter.

[0039] Through the above steps, the relationship between process parameters and performance parameters can be analyzed quantitatively and interpretably.

[0040] For example, the substrate heat treatment state S HT The contribution of the SHAP to the prediction of corrosion current density (logarithmic value) is not less than 20%, and the SHAP value corresponding to the T73 heat treatment state is the most negative among all heat treatment states; alloy Zn equivalent Zn eq The Pearson correlation coefficient for microhardness prediction is no higher than -0.55, and the Pearson correlation coefficient for wear rate (logarithmic value) is no lower than +0.45; the contribution of the G:CNT mass ratio to the prediction of friction coefficient SHAP is no less than 33%, and the SHAP value is most negative when G:CNT=1:1.

[0041] This application also provides a performance prediction and process optimization system for PEO composite coatings of 7-series aluminum alloys, specifically including a forward prediction module and a reverse optimization module.

[0042] The forward prediction module is used to execute the performance prediction method for the PEO composite coating of the 7-series aluminum alloy provided in this embodiment to predict the combination of performance parameters of the composite coating under different combinations of process parameters to be tested.

[0043] For example, the forward prediction module is configured to perform the following steps: obtaining the test process parameter combination of the composite coating, wherein the composite coating is a synergistic enhancement of graphene and carbon nanotubes in a 7-series aluminum alloy plasma electrolytic oxidation composite coating; the process parameter combination includes electrolyte formulation, positive current density, negative current density, pulse frequency, positive duty cycle, processing time, total carbon addition, mass ratio of graphene to carbon nanotubes, alloy grade, substrate heat treatment state, charge density, positive and negative current ratio, effective specific surface area of ​​carbon material, and alloy Zn equivalent; inputting the test process parameter combination into the trained artificial neural network, extreme gradient boosting model, and support vector regression and random forest regression ensemble model respectively to obtain the corresponding performance parameter combination; wherein, the coating thickness and microhardness of the composite coating are predicted by the artificial neural network, the friction coefficient and wear rate of the composite coating are predicted by the extreme gradient boosting model, and the corrosion current density of the composite coating is predicted by the support vector regression and random forest regression ensemble model; the coating thickness, microhardness, friction coefficient, wear rate, and corrosion current density constitute the performance parameter combination of the composite coating.

[0044] The reverse optimization module is used to execute the process optimization method for the PEO composite coating of the 7-series aluminum alloy provided in this embodiment to determine the target process parameter combination when the composite coating meets the target performance parameter combination.

[0045] For example, the reverse optimization module is configured to perform the following steps: obtain the target performance parameter combination of the composite coating; use the NSGA-II multi-objective genetic algorithm to call the performance prediction method of the PEO composite coating of the 7-series aluminum alloy to obtain the relationship between different process parameter combinations and different performance parameter combinations, and then determine the target process parameter combination that satisfies the target performance parameter combination; use SHAP value analysis, KDE joint distribution map and process response heatmap to quantitatively analyze the relationship between different process parameter combinations and different performance parameter combinations, and determine the influence law of each process parameter on each performance parameter.

[0046] Meanwhile, based on the performance prediction method of the PEO composite coating of the seven-series aluminum alloy in this embodiment, the system can simultaneously predict the performance comparison of composite coatings of 2-4 alloy grades under the same process parameters, and can display them side by side using radar charts.

[0047] Furthermore, based on the process optimization method for the PEO composite coating of the 7-series aluminum alloy in this embodiment, the system can output recommended process parameters for different performance requirements. For example, the recommended maximum coating thickness mode is 7075-T73, type II electrolyte (NaAlO2 10g / L, KOH 1g / L, Na2SiO3 5g / L, EDTA-2Na 3g / L), and the recommended forward current density is J. + =0.45A / cm 2The processing time is t=25min; other process parameters have little effect on the coating thickness. The recommended maximum microhardness mode is 7075-T73, Type I, with a positive current density of J. + =0.35A / cm 2 The processing time is t=18min; other process parameters have little effect on microhardness. The recommended minimum friction coefficient mode is 7075-T73, Type I, with a forward current density of J. + =0.35A / cm 2 Total carbon addition C total =1.3g / L, graphene to carbon nanotube mass ratio G:CNT=1:1. Other process parameters have little effect on the friction coefficient; the optimal corrosion protection mode recommended is 7075-T73, Type I, forward current density J. + =0.35A / cm 2 The processing time t=22min, the mass ratio of graphene to carbon nanotubes G:CNT=1:1, and other process parameters have little effect on the corrosion current density.

[0048] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0049] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0050] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0051] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0052] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0053] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0054] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0055] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method for predicting the performance of a PEO composite coating on a 7-series aluminum alloy, characterized in that, include: The combination of process parameters to be tested for the composite coating is obtained. The composite coating is a plasma electrolytic oxidation composite coating of a seven-series aluminum alloy that is synergistically reinforced by graphene and carbon nanotubes. The combination of process parameters to be tested is input into the trained artificial neural network, extreme gradient boosting model and support vector regression and random forest regression ensemble model respectively to obtain the corresponding performance parameter combination; Specifically, the coating thickness and microhardness of the composite coating are predicted by the artificial neural network, the friction coefficient and wear rate of the composite coating are predicted by the extreme gradient boosting model, and the corrosion current density of the composite coating is predicted by the integrated model of support vector regression and random forest regression. The coating thickness, microhardness, friction coefficient, wear rate and corrosion current density constitute the performance parameter combination of the composite coating.

2. The performance prediction method for PEO composite coatings on 7-series aluminum alloys according to claim 1, characterized in that, The process parameter combination includes electrolyte formulation, positive current density, negative current density, pulse frequency, positive duty cycle, processing time, total carbon addition, mass ratio of graphene to carbon nanotubes, alloy grade, and substrate heat treatment state.

3. The performance prediction method for PEO composite coatings on 7-series aluminum alloys according to claim 2, characterized in that, The process parameter combination also includes charge density, positive and negative current ratio, effective specific surface area of ​​carbon material and alloy Zn equivalent; the alloy Zn equivalent is used to characterize the contribution of key elements affecting the performance of the composite coating in different alloy grades of VII series aluminum alloys.

4. The performance prediction method for the PEO composite coating of 7-series aluminum alloys according to claim 2, characterized in that, The range of alloy grades includes 7075, 7055, 7085 and 7050, and the range of base heat treatment states includes T6, T73, T76 and T7351.

5. The performance prediction method for PEO composite coatings on 7-series aluminum alloys according to claim 2, characterized in that, The electrolyte formulation is a quaternary system of NaAlO2, KOH, Na2SiO3 and EDTA-2Na, wherein the mass concentration of NaAlO2 ranges from 6 g / L to 12 g / L, the mass concentration of KOH ranges from 0.5 g / L to 2.0 g / L, the mass concentration of Na2SiO3 ranges from 1 g / L to 8 g / L, and the mass concentration of EDTA-2Na ranges from 1 g / L to 4 g / L.

6. The performance prediction method for PEO composite coatings on 7-series aluminum alloys according to claim 2, characterized in that, The forward current density is in the range of 0.15 A / cm. 2 ~0.60A / cm 2 The processing time ranges from 5 min to 30 min, the total carbon addition ranges from 0 g / L to 2.0 g / L, the mass ratio of graphene to carbon nanotubes ranges from 0:1 to 4:0, and the forward duty cycle is selected as 40%.

7. The performance prediction method for PEO composite coatings on 7-series aluminum alloys according to claim 3, characterized in that, The Zn equivalent of the alloy Zn eq =w Zn +5w Cu ,in, w Zn and w Cu These represent the mass fractions of Zn and Cu elements in VII series aluminum alloys, respectively.

8. A process optimization method for a PEO composite coating of 7-series aluminum alloys, characterized in that, include: Obtain the target performance parameter combination of the composite coating; The NSGA-II multi-objective genetic algorithm is used to call the performance prediction method of the seven-series aluminum alloy PEO composite coating as described in any one of claims 1-7 to obtain the relationship between different combinations of process parameters and different combinations of performance parameters, and then to determine the target combination of process parameters that satisfies the target performance parameter combination.

9. The process optimization method for the PEO composite coating of 7-series aluminum alloys according to claim 8, characterized in that, Also includes: The relationship between different combinations of process parameters and different combinations of performance parameters was quantitatively analyzed by using SHAP value analysis, KDE joint distribution diagram and process response heatmap, to determine the influence of each process parameter on each performance parameter.

10. A performance prediction and process optimization system for PEO composite coatings on 7-series aluminum alloys, characterized in that, include: A forward prediction module is used to execute the performance prediction method of any one of claims 1-7 for the 7-series aluminum alloy PEO composite coating to predict the performance parameter combination of the composite coating under different combinations of test process parameters. The reverse optimization module is used to execute the process optimization method for the PEO composite coating of the seven-series aluminum alloy as described in claim 8 or 9 to determine the target process parameter combination when the composite coating meets the target performance parameter combination.

Images