Standardized optimization method for multi-process-parameter cooperative surface treatment of aeronautical parts

Abstract

The invention provides an aeronautical part multi-process parameter cooperative surface treatment standardization optimization method, and relates to the technical field of aeronautical manufacturing surface engineering, and the method comprises the following steps: S1, constructing a process parameter hierarchical coding system and a correlation mapping model; s2, establishing a parameter solving method based on multi-objective collaborative optimization; s3, constructing a modular process chain; s4, establishing real-time monitoring and closed-loop control of multi-source data fusion; s5, establishing a digital simulation body and verifying the digital simulation body; and S6, constructing a whole-process quality tracing database. According to the method, dynamic collaborative optimization of multiple process parameters is realized through parameter hierarchical coding, association mapping and a cross-process-chain intelligent decision model, the parameter conflict rate is reduced, and the process decision efficiency is improved. Meanwhile, a self-adaptive optimization engine based on reinforcement learning can correct process parameters in real time, the batch consistency of key performance indexes is improved, and the process stability and the product quality are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the technical field of aviation manufacturing surface engineering, and in particular to a standardized optimization method for collaborative surface treatment of multiple process parameters of aviation parts. Background Art

[0002] In the field of aviation manufacturing, aviation parts such as wing structures, engine blades, and landing gear components are subject to long-term exposure to extreme conditions such as high temperature, high pressure, high corrosion, and complex alternating loads. Their surface properties, such as wear resistance, corrosion resistance, and high-temperature oxidation resistance, directly determine the reliability, safety, and service life of the equipment. To meet this demand, surface treatment technology has become a key link in aviation parts manufacturing, requiring the integration of multiple processes such as laser cladding, micro-arc oxidation, electroless plating, thermal spraying, and vapor deposition to achieve surface functionalization such as strengthening, corrosion protection, and drag reduction. However, the existing surface treatment technology has significant technical bottlenecks: First, the isolated optimization of process parameters leads to the lack of multi-process synergy. For example, there is a complex thermal-electro-chemical coupling effect between the power and scanning speed of laser cladding and the voltage and frequency of micro-arc oxidation. If the parameters are not optimized in a coordinated manner, it is easy to cause interface thermal stress concentration, excessive coating porosity or microstructural defects such as cracks and unfusion. Especially in materials with large differences in thermophysical properties such as titanium alloys and high-temperature alloys, the probability of parameter conflict increases significantly; secondly, the process sequence relies on experience and lacks thermal-chemical coupling control. For example, if the cooling rate or interval time is not strictly controlled after micro-arc oxidation, it may cause The accumulation of residual stress causes deformation of the substrate. If the residual oxide layer is not completely removed before chemical plating, it may cause uncontrolled interfacial reaction and form a brittle phase. However, the existing technology lacks a quantitative model for the coupling of process sequence and parameters, resulting in poor process stability. Thirdly, the low degree of standardization and arbitrary parameter adjustment lead to large fluctuations in product quality. For example, the laser cladding power fluctuates by ±20% due to the equipment model and powder characteristics. The formula of the micro-arc oxidation electrolyte is not matched according to the thermal expansion coefficient of the material, resulting in coating peeling or insufficient corrosion resistance. Such problems are particularly prominent in the mass production of aviation parts. The performance difference between batches can reach more than 30%, making it difficult to meet zero-defect requirements.

[0003] In addition, full-process quality traceability and process reproduction are difficult. Existing technologies store process parameter records and quality inspection data independently, and lack correlation analysis. For example, parameter deviations such as laser cladding power fluctuations and micro-arc oxidation current density anomalies are not bound to coating hardness and bonding strength, resulting in low defect tracing efficiency. The dynamic adjustment of process parameters lacks historical data support, making it difficult to achieve full life cycle quality control. Finally, the integration of intelligent and digital technologies is insufficient. Existing surface treatment processes mostly remain at the single-machine automation stage, lacking real-time perception and collaborative control of process parameters, equipment status, and environmental variables. For example, anomalies such as laser cladding spot offset and micro-arc oxidation local breakdown are not identified in a timely manner, resulting in an increase in the coating defect rate. The virtual simulation and physical experimental data of multiple process chains have not been closed-loop verified, and the process synergy effect cannot be predicted, which limits the intelligent upgrade of aviation parts surface treatment towards high precision, high reliability, and high efficiency. Therefore, there is an urgent need for a standardized integrated method for collaborative surface treatment of multiple process parameters. Through parameter collaborative optimization, standardized design of the process chain, full-process closed-loop control and intelligent quality traceability, it can break through the limitations of traditional technologies such as parameter isolation, process disorder and uncontrollable quality, and provide technical support for high-performance manufacturing of aviation equipment. Summary of the Invention

[0004] In order to address the deficiencies of the above-mentioned prior art, the purpose of the present invention is to provide a standardized optimization method for collaborative surface treatment of multiple process parameters for aviation parts. This method aims to achieve collaborative optimization, standardized integration and full-process quality control of multiple process parameters through a series of technical measures. It can achieve dynamic collaborative optimization of multiple process parameters through parameter hierarchical coding and association mapping and cross-process chain intelligent decision-making models, reduce parameter conflict rate and improve process decision-making efficiency.

[0005] Specifically, the present invention provides a standardized optimization method for collaborative surface treatment of multiple process parameters of aviation parts, which includes the following steps: S1. Constructing a hierarchical coding system and an associative mapping model for process parameters: Dividing surface treatment process parameters into basic parameters and collaborative parameters and establishing a three-level coding rule; forming a standardized parameter library and dynamically modifying the coupling threshold in the standardized parameter library; S2. Establish a parameter solution method based on multi-objective collaborative optimization: Taking the key performance indicators of the surface treatment process as the optimization target, combined with physical constraints, a multi-process parameter collaborative optimization model is constructed to solve the optimal parameter combination that meets multiple constraints: ; ; ; ; Where: objective function Take m key performance indicators as optimization targets; There are p physical constraints; are q equality constraints; is the value range of the i-th process parameter; S3. Build a modular process chain: Decompose the surface treatment process into three modules: pre-treatment, main treatment, and post-treatment. Each module is divided into multiple process units. Establish rules for transferring parameters and data between modules. Use a hierarchical coding system to achieve cross-module matching of process parameters and automatically match the optimal process unit combination. S4. Establish real-time monitoring and closed-loop control based on multi-source data fusion: compare key status parameters with the threshold range in the standardized parameter library to ensure that the key status parameters are within the threshold range; S5. Establish a digital simulation and perform verification; S6. Build a full-process quality traceability database.

[0006] Preferably, in step S1, the process type is identified by the first-level coding, the parameter category is identified by the second-level coding, and the range and accuracy of the parameter value are identified by the third-level coding, so as to realize the standardized identification and traceability of the parameters; the correlation between the basic parameters and the collaborative parameters is quantified by experimental design multi-field coupling simulation, a parameter coupling model is established and the collaborative threshold range is defined, so as to form a standardized parameter library including parameter classification, coding rules and collaborative constraints; finally, the coupling threshold in the parameter library is dynamically corrected through the feedback mechanism of process execution data.

[0007] Preferably, in step S1, the coupling threshold in the standardized parameter library is dynamically corrected by the process execution data, and the coupling threshold correction formula is as follows: ; Among them, the correction amount The calculation formula is as follows: ; Where: is the coupling threshold between the corrected process parameters i and j; is the coupling threshold between the original process parameters i and j in the standardized parameter library; Is the correction coefficient, used to control the correction range, the range is 0< ≤1; is the correlation coefficient between process parameters i and j, reflecting the correlation strength between basic parameters and synergistic parameters, and the range is 0≤ ≤1; is the occasional error between process parameters i and j in actual execution; The maximum allowable occasional error is determined based on process requirements and historical data; and are the upper and lower limits of the coupling threshold between process parameters i and j, respectively.

[0008] Preferably, in step S3, each module is refined into an independent process unit in combination with the process content specified in the surface treatment process standard, and standardized input and output interfaces of the three major modules are defined.

[0009] Preferably, the specific steps of using the multi-process parameter collaborative optimization model to solve the optimal parameter combination that meets multiple constraints in step S2 are: using a combination of genetic algorithm and response surface method to generate an initial parameter population through experimental design, and using finite element simulation and physical experimental data to iteratively correct the objective function and quantify the nonlinear coupling effect between parameters; introducing an adaptive weight adjustment strategy to dynamically balance the competitive relationship between multiple objectives and generate an optimal parameter combination that meets multiple constraints.

[0010] Preferably, in step S3, the optimal process unit combination is automatically matched based on a decision tree algorithm according to the material type, performance requirements and geometric characteristics of the aviation parts.

[0011] Preferably, in step S4, when the key state parameters deviate from the preset threshold, the closed-loop control mechanism is automatically triggered, and the process parameters are dynamically adjusted using PID control or fuzzy control algorithm, and the process effect after the parameter adjustment is predicted through the digital simulation body in step S5; the parameter data, state monitoring values ​​and quality inspection results before and after the adjustment are associated and stored in the full-process quality traceability database in step S6, forming a complete closed loop of monitoring-decision-making-execution-feedback.

[0012] Preferably, in step S5, a digital simulation body covering the entire process of pre-processing, main processing and post-processing is constructed, and the physical field evolution and surface treatment microstructure formation process under the synergistic effect of multiple process parameters are simulated in real time through the finite element model and the standardized parameter library; the real-time monitoring data in step S4 are bidirectionally mapped with the virtual simulation results, and the accuracy of the digital simulation body is dynamically iterated through a data-driven model correction mechanism.

[0013] Preferably, the specific steps of constructing the full-process quality traceability database in step S6 are: S61. Using the standardized parameter coding system of step S1 and the modular process chain of step S3, the aviation part's base material, process path, process parameters, and real-time monitoring data are structured and stored, and the quality inspection results are bound to the part's unique identity. S62. Build a multi-dimensional association model based on graph database technology to conduct penetrating tracing by part number, process type, time interval, or quality defect type, and locate abnormal parameter nodes. Integrate machine learning algorithms to conduct in-depth mining of historical quality data, automatically extract association rules between parameter deviations and defect patterns, and generate visual defect tracing reports and process improvement suggestions. S63. Achieve cross-departmental quality data collaboration through a cloud-based data sharing platform, promote continuous optimization of the standardized parameter library through closed-loop feedback of historical defect data, and form a virtuous iterative mechanism of quality monitoring-defect tracing-process improvement.

[0014] Preferably, the method further includes step S7, constructing a process intelligent decision-making and adaptive optimization system, and the specific steps are as follows: S71. Based on the historical process data in the full-process quality traceability database in step S6, the real-time monitoring information of key status parameters in step S4, and the virtual verification results in step S5, a nonlinear relationship between process parameters, environmental variables, and surface treatment performance is obtained through a deep learning algorithm, and an intelligent decision-making model across the process chain is established; S72, real-time evaluation of process execution results based on reinforcement learning algorithms, and dynamic correction of parameter decision logic through closed-loop feedback mechanisms; S73. Conduct virtual-reality interactive verification between the digital simulation and the physical production line.

[0015] Compared with the prior art, the present invention has the following beneficial effects: (1) Through parameter hierarchical coding and association mapping and a cross-process chain intelligent decision-making model, this invention can achieve dynamic collaborative optimization of multiple process parameters, reduce parameter conflict rates, and improve process decision-making efficiency. At the same time, the adaptive optimization engine based on reinforcement learning can correct process parameters in real time, improve the batch consistency of key performance indicators, and significantly enhance process stability and product quality.

[0016] (2) This invention achieves standardized reuse and rapid matching of process modules through modular process chain construction and digital simulation virtual verification, thereby improving process chain design efficiency and reducing development costs. In addition, virtual verification technology can predict defects in advance, shorten the process development cycle, improve product qualification rate, and reduce trial and error costs and resource waste.

[0017] (3) This invention builds a full-process quality traceability database. Through the process-parameter-quality ternary correlation model, it achieves penetrating traceability from part number to process node, improving the efficiency of defect tracing and stabilizing the success rate of process reproduction. At the same time, intelligent early warning of defect patterns based on machine learning compresses the range of quality fluctuations, providing a highly reliable and consistent surface treatment solution for aviation manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 This is a flow chart of the standardized optimization method for collaborative surface treatment of multiple process parameters for aviation parts of the present invention; Figure 2 Schematic diagram of parameter hierarchical coding and collaborative optimization of the present invention; Figure 3 It is a modular process chain dynamic reorganization flow chart of the present invention; Figure 4 This is a diagram showing the interaction between the digital simulation virtual verification and the physical production line of the present invention; Figure 5 This is the architecture diagram of the full-process quality traceability database of the present invention; Figure 6 Schematic diagram of intelligent early warning of defect modes and parameter optimization of the present invention. DETAILED DESCRIPTION

[0019] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings.

[0020] Specifically, a standardized optimization method for collaborative surface treatment of multiple process parameters of aviation parts includes the following steps: S1. Construct a hierarchical coding system and an association mapping model for process parameters: divide the surface treatment process parameters into basic parameters and collaborative parameters and establish a three-level coding rule; quantify the correlation between basic parameters and collaborative parameters, establish a parameter coupling model and define the collaborative threshold range, form a standardized parameter library containing parameter classification, coding rules and collaborative constraints, and dynamically correct the coupling threshold in the standardized parameter library through process execution data; in step S1, the process type is identified by the first-level coding, the parameter category is identified by the second-level coding, and the parameter value range and accuracy are identified by the third-level coding, so as to realize the standardized identification and traceability of parameters; quantify the correlation between basic parameters and collaborative parameters through experimental design and multi-field coupling simulation, establish a parameter coupling model and define the collaborative threshold range, and form a standardized parameter library containing parameter classification, coding rules and collaborative constraints; finally, the coupling threshold in the parameter library is dynamically corrected through the feedback mechanism of process execution data.

[0021] The coupling threshold correction formula is as follows: ; Among them, the correction amount The calculation formula is as follows: ; Where: is the coupling threshold between the corrected process parameters i and j; is the coupling threshold between the original process parameters i and j in the standardized parameter library; Is the correction coefficient, used to control the correction range, the range is 0< ≤1; is the correlation coefficient between process parameters i and j, reflecting the correlation strength between basic parameters and synergistic parameters, and the range is 0≤ ≤1; is the occasional error between process parameters i and j in actual execution; The maximum allowable occasional error is determined based on process requirements and historical data; and are the upper and lower limits of the coupling threshold between process parameters i and j, respectively.

[0022] S2. Establish a parameter solution method based on multi-objective collaborative optimization: First, take the key performance indicators of the surface treatment process as the optimization target, combine with physical constraints, build a multi-process parameter collaborative optimization model, and use the multi-process parameter collaborative optimization model to solve the optimal parameter combination that meets the multi-constraint conditions; the specific steps of using the multi-process parameter collaborative optimization model to solve the optimal parameter combination that meets the multi-constraint conditions in step S2 are: using a combination of genetic algorithm and response surface method, generating the initial parameter population through experimental design, and using finite element simulation and physical experimental data to iteratively correct the objective function and quantify the nonlinear coupling effect between parameters; introducing an adaptive weight adjustment strategy to dynamically balance the competitive relationship between multiple objectives and generate the optimal parameter combination that meets the multi-constraint conditions. The multi-process parameter collaborative optimization model is specifically: ; ; ; ; Where: objective function Take m key performance indicators as optimization objectives; the constraint conditions are inequality constraints There are p physical constraints; the constraint equality constraint are q equality constraints; is the value range of the i-th process parameter. and are the minimum and maximum values ​​of the i-th process parameter respectively.

[0023] This innovative algorithm, based on the improvement and integration of classic algorithms such as genetic algorithms, the corresponding surface method, and the adaptive weight adjustment strategy, can more efficiently handle high-dimensional, nonlinear process parameter optimization problems. It also performs iterative optimization to improve the reliability of the optimization results.

[0024] S3. Build a modular process chain: First, decompose the surface treatment process into three modules: pretreatment, main treatment, and post-treatment. Each module is further refined into independent process units, and the standardized input and output interfaces of the three modules are defined. Then, establish the rules for transferring parameters and data between modules, and realize cross-module matching of process parameters through the hierarchical coding system of step S1. Automatically match the optimal process module combination according to the material type, performance requirements, and geometric characteristics of aviation parts.

[0025] First, clarify the transfer content such as process parameters, part feature data, quality inspection data, etc., determine the transfer method such as direct data transfer, intermediate database transfer or message queue transfer, and formulate data format and conversion rules; secondly, according to the coding system established in S1, cross-module matching is performed according to the process type and corresponding process parameters; finally, a part feature database is established, and a decision tree algorithm is used to output the optimal process module combination based on the process characteristics.

[0026] S4. Establish real-time monitoring and closed-loop control of multi-source data fusion: collect multi-dimensional data, pre-process the data, and compare the key state parameters with the threshold range in the standardized parameter library; in step S4, when the monitoring data deviates from the preset threshold, the system automatically triggers the closed-loop control mechanism, uses PID control or fuzzy control algorithm to dynamically adjust the process parameters, and predicts the process effect after parameter adjustment through the digital simulation model to ensure that the adjusted parameter combination meets the performance target; in addition, the parameter data, state monitoring values ​​and quality inspection results before and after adjustment are associated and stored in the full-process quality traceability database, forming a complete closed loop of monitoring-decision-making-execution-feedback. Afterwards, the real-time monitoring data in step S4 is bidirectionally mapped with the virtual simulation results, and the accuracy of the digital simulation body is dynamically iterated through the data-driven model correction mechanism.

[0027] S5. Establish a digital simulation body and verify it: First, build a digital simulation body covering the entire process of pre-processing, main processing and post-processing. By integrating the finite element model with the process parameter standardization library, the physical field evolution and surface treatment microstructure formation process under the synergistic effect of multiple process parameters are simulated in real time.

[0028] S6, building a full-process quality traceability database. Specifically, the specific steps of building a full-process quality traceability database in step S6 are: S61. Through the standardized parameter coding system of step S1 and the process chain modular identification of step S3, the base material, process path, process parameters and real-time monitoring data of the aviation part are structured and stored, and the quality inspection results are bound to the unique identity of the part.

[0029] S62. Establish a multi-dimensional association model based on graph database technology, support penetrating tracing by part number, process type, time interval or quality defect type, and quickly locate abnormal parameter nodes; integrate machine learning algorithms to conduct in-depth mining of historical quality data, automatically extract association rules between parameter deviations and defect patterns, and generate visual defect tracing reports and process improvement suggestions.

[0030] S63. Achieve cross-departmental quality data collaboration through a cloud-based data sharing platform. At the same time, through closed-loop feedback of historical defect data, promote continuous optimization of the standardized parameter library, and form a virtuous iterative mechanism of "quality monitoring-defect tracing-process improvement."

[0031] It also includes building a process intelligent decision-making and adaptive optimization system. The specific steps for building a process intelligent decision-making and adaptive optimization system are: S71. Based on the historical process data in the full-process quality traceability database in step S6, the real-time monitoring information in step S4, and the virtual verification results in step S5, the nonlinear relationship between process parameters, environmental variables and surface treatment performance is obtained through a deep learning algorithm, and an intelligent decision-making model across the process chain is established to support automatic recommendation of the optimal process parameter combination based on part type, performance requirements and real-time working conditions.

[0032] S72. Deploy an adaptive optimization engine to evaluate process execution results in real time based on reinforcement learning algorithms, and dynamically correct parameter decision logic through a closed-loop feedback mechanism to enable the process system to have autonomous learning and continuous improvement capabilities.

[0033] S73. Through virtual-reality interactive verification between digital simulation and physical production lines, the stability and reliability of the optimized parameter combination under complex working conditions are ensured. Specific embodiments The embodiment of the present invention provides a standardized optimization method for collaborative surface treatment of multiple process parameters of aviation parts, such as Figure 1 As shown, it includes the following steps: S1. Constructing a hierarchical coding system and an associated mapping model for process parameters: Figure 2As shown in the figure, the surface treatment process parameters are first classified into basic parameters and collaborative parameters. Basic parameters include single process independent variables such as laser power, spot size, and micro-arc oxidation termination voltage, and collaborative parameters include cross-process coupling variables such as laser scanning rate and micro-arc oxidation pulse frequency. A three-level coding rule is established, that is, the process type is identified by the first-level coding, such as LC-laser cladding, MAO-micro-arc oxidation, SP-spraying, CN-nitriding or CE-chemical plating; the parameter category is identified by the second-level coding, such as P-power, T-temperature, V-voltage, F-frequency, D-thickness or C-concentration; the parameter value range and accuracy are identified by the third-level coding. For example, P-LC-1500±50W represents the laser cladding power range and accuracy, realizing standardized identification and traceability of parameters. Subsequently, through experimental design, multi-field coupling simulation is used to quantify the correlation between basic parameters and collaborative parameters, a parameter coupling model is established, and the collaborative threshold range is defined. For example, when the laser power P-LC ≥ 1450W, the micro-arc oxidation voltage V-MAO needs to be ≤ 460V to avoid excessive interface thermal stress, forming a standardized parameter library containing parameter classification, coding rules and collaborative constraints. Finally, the coupling threshold in the parameter library is dynamically corrected through the feedback mechanism of process execution data, and parameter reuse and collaborative optimization across the process chain are supported.

[0035] S2. Establish a parameter solution method based on multi-objective collaborative optimization: In this embodiment, the key performance indicators of surface treatment such as coating hardness, bonding strength, porosity and residual stress are first used as optimization targets, and physical constraints such as material thermal expansion coefficient and substrate deformation are combined to construct a multi-process parameter collaborative optimization model. Subsequently, a genetic algorithm is combined with a response surface method to generate an initial population of parameters through experimental design, and the objective function is iteratively corrected using finite element simulation and physical experimental data to quantify the nonlinear coupling effect between parameters, such as the effect of the synergistic effect of laser power and micro-arc oxidation voltage on the crack sensitivity of the coating. On this basis, an adaptive weight adjustment strategy is introduced to dynamically balance the competitive relationship between multiple objectives, such as the contradiction between hardness improvement and porosity control, and generate the optimal parameter combination that meets multiple constraints, such as the collaborative matching of laser power 1520W, scanning speed 5.5mm / s and micro-arc oxidation voltage 452V.

[0036] S3, building a modular process chain: Figure 3As shown in the figure, the whole process of surface treatment is first decomposed into three modules: pretreatment, main treatment and post-treatment. Pretreatment includes cleaning or sandblasting, main treatment includes laser cladding, micro-arc oxidation or chemical plating, and post-treatment includes vacuum sealing or heat treatment for stress relief. Each module is further refined into independent process units. For example, the cleaning module includes alkaline cleaning, acid cleaning, deionized water rinsing and other sub-units, and the standardized input and output interfaces of the modules are defined. For example, the input of the pretreatment module is the substrate material and the initial surface roughness, and the output is the cleanliness and activation energy state; the input of the main treatment module is the substrate surface after pretreatment, and the output is the coating thickness and microstructure characteristics; then the transmission rules of parameters and data between modules are established, and a hierarchical coding system is used as shown in the steps S1 realizes cross-module matching of process parameters. For example, the surface roughness parameters after sandblasting need to be dynamically associated with the power-scanning speed combination of laser cladding, and designs a dynamic reorganization algorithm for the process chain. According to the material type of aviation parts such as titanium alloy and high-temperature alloy, performance requirements such as wear resistance and corrosion resistance, and geometric features such as thin-walled parts and complex cavities, the optimal process module combination is automatically matched. For example, for TC4 titanium alloy engine blades, the recommended process chain is "ultrasonic cleaning → sandblasting (Ra3.2μm) → laser cladding (NiCrBSi alloy) → micro-arc oxidation (ceramic coating) → vacuum sealing", while for high-temperature alloy turbine discs, it is adjusted to "chemical degreasing → laser shock strengthening → chemical nickel plating → heat treatment".

[0037] S4. Establish real-time monitoring and closed-loop control of multi-source data fusion: First, deploy high-precision sensor arrays at key process nodes such as laser cladding molten pool, micro-arc oxidation electrolytic cell, and chemical plating reaction chamber to collect multi-dimensional data such as temperature field, stress distribution, electrochemical signals, and spectral characteristics in real time, and pre-process the data through edge computing nodes, such as filtering and noise reduction, feature extraction, and compare key state parameters such as molten pool temperature fluctuation rate and electrolyte conductivity change rate with the threshold range in the standardized parameter library; when the key state parameter monitoring data deviates from the preset threshold, the system automatically triggers the closed-loop control mechanism, using fuzzy PID The control algorithm dynamically adjusts the process parameters. If the molten pool temperature exceeds the upper threshold of 5% during the laser cladding process, the laser power is simultaneously reduced by 2% and the scanning speed is increased by 3%. At the same time, the substrate cooling airflow compensation is started. The digital simulation model in step S5 is used to predict the process effects after parameter adjustment, such as changes in the coating crystal orientation and residual stress evolution trends, to ensure that the adjusted parameter combination still meets the performance goals. In addition, the parameter data before and after adjustment, status monitoring values ​​and quality inspection results such as microhardness and bonding strength are associated and stored in the full-process quality traceability database, forming a complete closed loop of "monitoring-decision-making-execution-feedback".

[0038] S5. Create a digital simulation and verify it: Figure 4As shown in the figure, first, a digital simulation body covering the entire process of pre-treatment, main treatment and post-treatment is constructed. Through the deep integration of high-precision finite element models such as thermal-electro-chemical multi-field coupling models and process parameter standardization libraries, the physical field evolution under the synergistic action of multiple process parameters, such as laser cladding molten pool flow, micro-arc oxidation plasma discharge, chemical plating growth dynamics and coating microstructure formation process such as grain orientation and phase composition distribution, is simulated in real time. Subsequently, the real-time monitoring data in step S4 is bidirectionally mapped with the virtual simulation results. Through data-driven model correction mechanisms such as Bayesian optimization-based parameter calibration, the accuracy of the dynamic iterative digital simulation body is improved, so that the prediction error of key properties such as coating hardness, bonding strength and porosity is controlled within 3%. On this basis, the virtual simulation body is used to carry out process limit tests such as coating failure simulation and fault tolerance analysis under extreme parameter combinations, such as process robustness evaluation when the sensor fails, to identify potential defects such as crack initiation location and interface debonding risk in advance, and output optimization suggestions such as adjusting the laser power gradient and optimizing the micro-arc oxidation pulse waveform. Finally, the parameter combination that passes the virtual verification is pushed to the actual production line for execution.

[0039] S6. Build a full-process quality traceability database: Figure 5 As shown in the figure, first, through the standardized parameter coding system and process chain modular identification, the base material of each batch of aviation parts, process path such as "laser cladding + micro-arc oxidation", each process parameter such as laser power, electrolyte concentration and real-time monitoring data such as temperature curve and stress distribution are structured and stored, and the quality inspection results such as coating hardness, bonding strength, corrosion resistance test value and the unique identity of the part such as QR code / RFID are bound; then, a multi-dimensional association model is established based on graph database technology to support penetrating tracing by part number, process type, time interval or quality defect type such as excessive porosity and crack initiation, and to quickly locate abnormal parameter nodes. For example, if the query finds that the coating bonding strength of a batch of parts is insufficient, it can be traced back to micro-arc oxidation. Abnormal records of voltage fluctuations exceeding the threshold of 5% in the chemical process; in addition, the integrated machine learning algorithm conducts in-depth mining of historical quality data, automatically extracts association rules between parameter deviations and defect patterns, such as "when the laser scanning speed is lower than 8mm / s and the substrate temperature exceeds 200℃, the risk of coating cracks increases by 40%", and generates visual defect tracing reports and process improvement suggestions; finally, cross-departmental and cross-enterprise quality data collaboration is achieved through the cloud-based data sharing platform, which improves the defect tracing efficiency by more than 90%, and stabilizes the process reproduction success rate at more than 98%. At the same time, through the closed-loop feedback of historical defect data, it promotes the continuous optimization of the standardized parameter library of step S1, forming a virtuous iterative mechanism of "quality monitoring-defect tracing-process improvement".

[0040] S7. Build process intelligent decision-making and adaptive optimization system: Figure 6As shown in the figure, firstly, the historical process data, real-time monitoring information and virtual verification results in the whole process quality traceability database are integrated, and the complex nonlinear relationship between process parameters, environmental variables and coating performance is mined through deep learning algorithms such as convolutional networks in the figure, and an intelligent decision-making model across the process chain is established. It supports automatic recommendation of the optimal process parameter combination based on part types such as engine blades, fuselage frames, performance requirements such as high-temperature wear resistance and corrosion resistance, and real-time working conditions such as equipment status fluctuations and changes in ambient temperature and humidity. For example, for a certain high-temperature alloy part, the model can integrate historical data and real-time equipment status to dynamically adjust the coordinated matching relationship between laser cladding power and micro-arc oxidation voltage; then, an adaptive optimization engine is deployed, based on reinforcement learning algorithms such as the PPO policy gradient algorithm in real time Evaluate the process execution effect and dynamically correct the parameter decision logic through a closed-loop feedback mechanism. For example, when the coating hardness is detected to deviate from the target value, the pH value of the chemical plating solution and the deposition time compensation are automatically adjusted to enable the process system to have autonomous learning and continuous improvement capabilities; finally, through the virtual-reality interactive verification of the digital simulation body and the physical production line, the stability and reliability of the optimized parameter combination under complex working conditions are ensured, the process decision efficiency is improved by more than 70%, and the batch consistency of key performance indicators such as coating life and bonding strength is increased to 95%. At the same time, through the self-evolution mechanism of the intelligent decision-making system, the surface treatment process is promoted from an experience-driven to a data-driven paradigm transformation, which significantly enhances the intelligence level and core competitiveness of the surface treatment of complex parts in the aviation manufacturing field.

[0041] The embodiments described above are merely descriptions of preferred implementations of the present invention and are not intended to limit the scope of the present invention. Without departing from the spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should fall within the scope of protection determined by the claims of the present invention.

Claims

1. A standardized optimization method for collaborative surface treatment of multiple process parameters for aviation parts, characterized by: It includes: S1. Constructing a hierarchical coding system and an associative mapping model for process parameters: Dividing surface treatment process parameters into basic parameters and collaborative parameters and establishing a three-level coding rule; forming a standardized parameter library and dynamically correcting the coupling threshold in the standardized parameter library; S2. Establish a parameter solution method based on multi-objective collaborative optimization: Taking the key performance indicators of the surface treatment process as the optimization target, combined with physical constraints, a multi-process parameter collaborative optimization model is constructed to solve the optimal parameter combination that meets multiple constraints: ； ； ； ； Where: objective function Take m key performance indicators as optimization targets; There are p physical constraints; are q equality constraints; is the value range of the i-th process parameter; S3. Build a modular process chain: Decompose the surface treatment process into three modules: pre-treatment, main treatment, and post-treatment. Each module is divided into multiple process units. Establish rules for transferring parameters and data between modules. Use a hierarchical coding system to achieve cross-module matching of process parameters and automatically match the optimal process unit combination. S4. Establish real-time monitoring and closed-loop control based on multi-source data fusion: compare key status parameters with the threshold range in the standardized parameter library to ensure that the key status parameters are within the threshold range; S5. Establish a digital simulation and perform verification; S6. Build a full-process quality traceability database.

2. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 1 is characterized by: In step S1, the process type is identified by the first-level coding, the parameter category is identified by the second-level coding, and the range and accuracy of the parameter value are identified by the third-level coding, so as to realize the standardized identification and traceability of the parameters; the correlation between the basic parameters and the collaborative parameters is quantified by experimental design multi-field coupling simulation, and a parameter coupling model is established and the collaborative threshold range is defined to form a standardized parameter library containing parameter classification, coding rules and collaborative constraints; finally, the coupling threshold in the parameter library is dynamically corrected through the feedback mechanism of process execution data.

3. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 2 is characterized by: In step S1, the coupling threshold in the standardized parameter library is dynamically corrected using the process execution data. The coupling threshold correction formula is as follows: ； Among them, the correction amount The calculation formula is as follows: ； Where: is the coupling threshold between the corrected process parameters i and j; is the coupling threshold between the original process parameters i and j in the standardized parameter library; Is the correction coefficient, used to control the correction range, the range is 0< ≤1; is the correlation coefficient between process parameters i and j, reflecting the correlation strength between basic parameters and synergistic parameters, and the range is 0≤ ≤1; is the occasional error between process parameters i and j in actual execution; The maximum allowable occasional error is determined based on process requirements and historical data; and are the upper and lower limits of the coupling threshold between process parameters i and j, respectively.

4. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 1 is characterized by: In step S3, each module is refined into an independent process unit in combination with the process content specified in the surface treatment process standard, and the standardized input and output interfaces of the three major modules are defined.

5. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 1 is characterized by: The specific steps of using the multi-process parameter collaborative optimization model in step S2 to solve the optimal parameter combination that meets multiple constraints are as follows: using a combination of genetic algorithm and response surface method to generate an initial parameter population through experimental design, and using finite element simulation and physical experimental data to iteratively correct the objective function and quantify the nonlinear coupling effect between parameters; introducing an adaptive weight adjustment strategy to dynamically balance the competitive relationship between multiple objectives and generate the optimal parameter combination that meets multiple constraints.

6. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 1 is characterized by: In step S3, the optimal process unit combination is automatically matched based on the decision tree algorithm according to the material type, performance requirements and geometric characteristics of the aviation parts.

7. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 1 is characterized by: In step S4, when the key state parameters deviate from the preset threshold, the closed-loop control mechanism is automatically triggered, and the process parameters are dynamically adjusted using PID control or fuzzy control algorithms. The process effects after parameter adjustment are predicted through the digital simulation body in step S5; the parameter data, state monitoring values ​​and quality inspection results before and after adjustment are associated and stored in the full-process quality traceability database in step S6, forming a complete closed loop of monitoring-decision-making-execution-feedback.

8. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 7 is characterized by: In step S5, a digital simulation body covering the entire process of pre-processing, main processing and post-processing is constructed, and the physical field evolution and surface treatment microstructure formation process under the synergistic effect of multiple process parameters are simulated in real time through the finite element model and standardized parameter library; The real-time monitoring data in step S4 is bidirectionally mapped to the virtual simulation results, and the accuracy of the digital simulation body is dynamically iterated through a data-driven model correction mechanism.

9. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 8 is characterized by: The specific steps for constructing the full-process quality traceability database in step S6 are: S61. Using the standardized parameter coding system of step S1 and the modular process chain of step S3, the aviation part's base material, process path, process parameters, and real-time monitoring data are structured and stored, and the quality inspection results are bound to the part's unique identity. S62. Build a multi-dimensional association model based on graph database technology to conduct penetrating tracing by part number, process type, time interval, or quality defect type, and locate abnormal parameter nodes. Integrate machine learning algorithms to conduct in-depth mining of historical quality data, automatically extract association rules between parameter deviations and defect patterns, and generate visual defect tracing reports and process improvement suggestions. S63. Achieve cross-departmental quality data collaboration through a cloud-based data sharing platform, promote continuous optimization of the standardized parameter library through closed-loop feedback of historical defect data, and form a virtuous iterative mechanism of quality monitoring-defect tracing-process improvement.

10. The standardized optimization method for collaborative surface treatment of aviation parts with multiple process parameters according to claim 1 is characterized by: The process also includes step S7, constructing a process intelligent decision-making and adaptive optimization system, and the specific steps are as follows: S71. Based on the historical process data in the full-process quality traceability database in step S6, the real-time monitoring information of key status parameters in step S4, and the virtual verification results in step S5, a nonlinear relationship between process parameters, environmental variables, and surface treatment performance is obtained through a deep learning algorithm, and an intelligent decision-making model across the process chain is established; S72, real-time evaluation of process execution results based on reinforcement learning algorithms, and dynamic correction of parameter decision logic through closed-loop feedback mechanisms; S73. Conduct virtual-reality interactive verification between the digital simulation and the physical production line.

Images