Material qualification method applicable to airworthiness certification of civil aircraft engines
Through historical data or experimental verification, the statistical characteristics of civil aviation engine materials are determined, graded and approved and multi-level verification are carried out, which solves the stability and reliability of materials in airworthiness certification, ensures that the materials meet airworthiness requirements, and improves the approval Efficiency and accuracy.
Patent Information
- Application Number
- CN202411203359.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-30
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2044-08-30
AI Technical Summary
The lack of effective material qualification identification methods in the prior art has led to the inability of civil aviation engine materials to ensure continuous stability in the airworthiness certification process, low data reliability, and lack of sufficient evidence to show that the materials can continue to meet the standard requirements.
Through historical data or material qualification identification tests, the statistical characteristics of the material are determined, the material qualified product catalog is generated, the materials are divided into key parts, important parts and general parts, and the identification is carried out in grading. Multi-stage gradient interruption heat treatment process is used to eliminate residual stress, and The long-term performance verification of structural simulation parts ensures that the materials meet airworthiness requirements.
Improve the efficiency and accuracy of material qualification identification, ensure that the material standards are fully verified, provide the quality stability and consistency of the material, establish the material quality stability evaluation criteria that meets the airworthiness requirements, and provide the material quality stability evaluation criteria for civil aviation engine materials to be adapted to Airline approval provides technical support.
Smart Images

Figure CN118734717B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of civil aviation engine airworthiness certification, and in particular to a material qualification identification method applicable to civil aviation engine airworthiness certification. Background Art
[0002] my country's commercial aviation and general aviation are developing rapidly. In the future, civil aviation will inevitably give birth to a huge market. Civil aviation engines must comply with airworthiness regulations and obtain type certificates before they can be operated commercially. There is a special clause "33.15 Materials" in the airworthiness regulations for materials. During the airworthiness certification stage of civil aviation engines, engine materials must provide sufficient evidence to show that they meet all the requirements of the airworthiness clauses. Therefore, material qualification is the most important part of the material airworthiness compliance verification process, but there is currently no available material qualification method. At present, for material verification, it is generally sufficient to meet the standard requirements. The verification methods, processes and evaluation methods are not standardized, the reliability of the data obtained is not high, and there is insufficient evidence to show that the materials produced by suppliers can continuously and stably meet the material standard requirements. Therefore, there is an urgent need for a material qualification method suitable for civil aviation engine airworthiness certification. Summary of the invention
[0003] The present invention provides a material qualification method applicable to civil aviation engine airworthiness certification, which can obtain material standards with sufficient statistical characteristics through historical data or material qualification tests, and provide technical support for civil aviation engine materials to pass airworthiness certification.
[0004] According to one aspect of the present invention, there is provided a material qualification method applicable to civil aviation engine airworthiness certification, comprising the following contents:
[0005] Determine the materials for civil aviation engines;
[0006] Determine whether material qualification test is needed based on the application of the selected material and its historical data, and obtain the selected material's conformity statement report or material qualification test data;
[0007] Generate a catalog of qualified materials for civil aviation engines based on the selected material compliance report and material qualification test results;
[0008] The process of judging whether a material qualification test is required based on the application of the selected material and its historical data includes the following:
[0009] If the selected materials are mature materials that have been successfully applied in civil aviation products, and the historical data batches reach 10 batches of 10 batches, the number of sub-samples of each technical indicator of the material specification reaches 30, and the data source is traceable, the test process records are complete, and the formation process of the historical data meets the quality system requirements and technical requirements, then all conformity statements are prepared in the form of conformity statements, and combined with the subsequent whole machine test results to show that the requirements of the clause are met, then the material qualification test verification will no longer be carried out separately;
[0010] If the selected material is a mature material that has been successfully applied in civil aviation products, but the historical data only partially meets the requirements of 10 batches of 10 furnaces, and the number of sub-samples of the technical indicators of the material specification reaches 30, and the data source is traceable, the test process records are complete, and the formation process of the historical data meets the quality system requirements and technical requirements, then the part shall be prepared in the form of a conformity statement report, and combined with the subsequent whole machine test results to show that it meets the requirements of the clause, no separate test verification shall be carried out; the part that does not meet the requirements shall be subject to material qualification test verification;
[0011] If the selected material has not been successfully used in civil aviation products, or the historical data corresponding to the selected material does not meet the requirements, or the selected material is a new material, a separate material qualification test verification is required.
[0012] Furthermore, the material qualification test process includes the following:
[0013] According to the key characteristics of engine parts, the selected materials are divided into key parts materials, important parts materials and general parts materials;
[0014] Suppliers are divided into key-level suppliers, important-level suppliers and general-level suppliers according to the importance of the parts and technical capabilities they undertake. Key-level suppliers can undertake the qualification assessment of key parts materials, important parts materials and general parts materials, important-level suppliers can undertake the qualification assessment of important parts materials and general parts materials, and general-level suppliers can undertake the qualification assessment of general parts materials.
[0015] Prepare the first draft of material standards and material production process control documents;
[0016] Suppliers carry out the first batch production and inspection of key parts, important parts and general parts according to the material production process control documents and the first draft of material standards, and obtain the first piece identification data;
[0017] Carry out test verification on key parts, important parts and general parts together with the whole machine to obtain the whole machine test data;
[0018] Each technical indicator specified in the first draft of the material standard is verified and revised based on the first piece inspection data and the whole machine test data.
[0019] Furthermore, for key and important materials, after the first-article inspection is completed, the supplier must continuously produce 3 batches of materials in 3 furnaces in accordance with the material production process control documents, and conduct multiple batches of performance evaluation tests in accordance with the same requirements as the first-article inspection. After the multiple batches of performance evaluation tests are completed and qualified, a multiple batch performance evaluation report shall be prepared.
[0020] Furthermore, after passing multiple batches of performance evaluation tests, quality stability evaluation is also required.
[0021] Furthermore, for key component materials, it is also necessary to conduct long-term performance verification of the materials in combination with the actual operating conditions of the engine.
[0022] Furthermore, for key component materials, it is necessary to design structural simulation parts and conduct long-term performance verification of the structural simulation parts.
[0023] Furthermore, component / environmental testing verification is required for key and important parts materials.
[0024] Furthermore, the process of long-term performance verification of the structural simulation component includes the following contents:
[0025] Design structural simulation parts according to the structural characteristics of key parts and the temperature and load conditions during service. Through simulation and optimization iteration, make the structural characteristics and load conditions of the structural simulation parts consistent with the real parts.
[0026] Carry out structural simulation part processing, and use multi-level gradient interruption heat treatment process to eliminate the residual stress in the structural simulation part processing process, so as to avoid the residual stress in the processing process affecting the test results of the simulation part;
[0027] According to the engine test spectrum, a test spectrum of structural simulation parts is compiled, and the heat-treated structural simulation parts are subjected to long-term tests according to the test spectrum of structural simulation parts to obtain the evolution law between the organization and performance of the structural simulation parts under long-term use conditions;
[0028] A corrosion test spectrum is added to the test spectrum of the structural simulation parts. The heat-treated structural simulation parts are subjected to long-term tests according to the composite test spectrum to obtain the influence of corrosion on the evolution of the organization and performance of the structural simulation parts under long-term use conditions.
[0029] Furthermore, the process of eliminating residual stress during the processing of the structural simulation part by using a multi-level gradient interruption heat treatment process includes the following contents:
[0030] Place the structural simulation part in a high-pressure vacuum heat treatment furnace, and raise the temperature in the furnace to 1010-1080°C at a heating rate of 10-20°C / min;
[0031] Keep the structural simulation parts at 1010-1080℃ for 1-2h;
[0032] Cooling the structural simulation to room temperature;
[0033] Put the structural simulation part into a high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keep it warm for 0.5-1h, and then cool the structural simulation part to room temperature; put the structural simulation part back into the high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keep it warm for 0.5-1h, and then cool the structural simulation part to room temperature; put the structural simulation part back into the high-pressure vacuum heat treatment furnace again, increase the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keep it warm for 0.5-1h, and then cool the structural simulation part to room temperature;
[0034] Put the structural simulation part into a high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation part to room temperature; put the structural simulation part back into the high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation part to room temperature; put the structural simulation part back into the high-pressure vacuum heat treatment furnace again, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation part to room temperature;
[0035] Place the structural simulation part in a high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 620-650°C at a heating rate of 30-50°C / min, keep it warm for 2h-4h, and then cool the structural simulation part to room temperature; place the structural simulation part back in the high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 620-650°C at a heating rate of 30-50°C / min, keep it warm for 2h-4h, and then cool the structural simulation part to room temperature; place the structural simulation part back in the high-pressure vacuum heat treatment furnace again, increase the temperature in the furnace to 620-650°C at a heating rate of 30-50°C / min, keep it warm for 2h-4h, and then cool the structural simulation part to room temperature.
[0036] The present invention has the following beneficial effects:
[0037] The material qualification method applicable to the airworthiness certification of civil aviation engines of the present invention, after determining the material selection of civil aviation engines, firstly judges whether it is necessary to conduct a material qualification test according to the application of the selected materials and their historical data. If the amount of historical data of the selected materials meets the technical requirements, there is no need to conduct a separate material qualification test, and the compliance description report of the selected materials is directly obtained. If the amount of historical data of the selected materials does not meet the technical requirements, it is necessary to conduct a separate material qualification test. Finally, based on the compliance description report of the selected materials and the results of the material qualification test, a material qualified product catalog of civil aviation engines is generated. The entire material qualification process can obtain a material standard with sufficient statistical characteristics through historical data or material qualification test, and the material standard is fully verified through a large amount of historical data or test verification. At the same time, when the amount of historical data of the selected materials is sufficient, there is no need to conduct a material qualification test, which improves efficiency and provides technical support for civil aviation engine materials to pass airworthiness certification.
[0038] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0039] The accompanying drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the accompanying drawings:
[0040] Figure 1 The present invention is a flow chart of a material qualification method applicable to civil aviation engine airworthiness certification according to a preferred embodiment of the present invention.
[0041] Figure 2 It is a schematic diagram of the process of conducting a material qualification test in a preferred embodiment of the present invention.
[0042] Figure 3 It is a schematic diagram of a process for verifying the long-term performance of a structural simulation component in a preferred embodiment of the present invention.
[0043] Figure 4 It is a flow chart of a method for designing a titanium alloy composition with high comprehensive performance according to another embodiment of the present invention. DETAILED DESCRIPTION
[0044] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.
[0045] Reference Figure 1The preferred embodiment of the present invention provides a material qualification method applicable to civil aviation engine airworthiness certification, comprising the following contents:
[0046] Step S1: Determine the material selection for civil aviation engines;
[0047] Step S2: judging whether a material qualification test is required based on the application of the selected material and its historical data, and obtaining a conformity statement report or material qualification test data of the selected material;
[0048] Step S3: Generate a catalog of qualified materials for civil aviation engines based on the selected material's compliance statement report and material qualification test results.
[0049] It can be understood that the material qualification method applicable to the airworthiness certification of civil aviation engines in this embodiment, after determining the material selection of civil aviation engines, first judges whether it is necessary to conduct a material qualification test based on the application of the selected materials and their historical data. If the amount of historical data of the selected materials meets the technical requirements, there is no need to conduct a separate material qualification test, and the compliance statement report of the selected materials is directly obtained. If the amount of historical data of the selected materials does not meet the technical requirements, it is necessary to conduct a separate material qualification test. Finally, based on the compliance statement report of the selected materials and the results of the material qualification test, a material qualified product catalog for civil aviation engines is generated. The entire material qualification process can obtain material standards with sufficient statistical characteristics through historical data or material qualification tests, and the material standards are fully verified through a large amount of historical data or test verification. At the same time, when the amount of historical data of the selected materials is sufficient, there is no need to conduct a material qualification test, which improves efficiency and provides technical support for civil aviation engine materials to pass airworthiness certification.
[0050] It can be understood that in step S1, the material selection of the civil aviation engine is carried out based on the expected service conditions, environment and strength life requirements of the civil aviation engine, and taking into account the principles of applicability, processability, versatility, maintainability, maturity, economy, etc.
[0051] It can be understood that in step S2, if the selected material is a mature material that has been successfully used in civil aviation products, historical data of the delivered material is collected, and statistical analysis is performed to determine whether analytical description can be used instead of test verification to improve the efficiency of material qualification. Specifically, if the selected materials are mature materials that have been successfully applied in civil aviation products, and the historical data batches reach 10 furnaces and 10 batches, the number of sub-samples of each technical indicator of the material specification reaches 30, and the data source is traceable, the test process records are complete, and the formation process of the historical data meets the quality system requirements and technical requirements, then the conformity statement report shall be prepared in the form of a conformity statement, and combined with the subsequent whole machine test results, it is shown that the requirements of the clause are met, and no separate material qualification test verification shall be carried out; if the selected materials are mature materials that have been successfully applied in civil aviation products, but the historical data only partially meet the furnace batches of 10 furnaces and 10 batches, the number of sub-samples of some technical indicators of the material specification reaches 30, and the data source is traceable, the test process records are complete, and the formation process of the historical data meets the quality system requirements and technical requirements, then the conformity statement report shall be prepared in the form of a conformity statement for the selected materials, and combined with the subsequent whole machine test results, it is shown that the requirements of the clause are met, and no separate test verification shall be carried out; the selected materials that do not meet the requirements shall be subject to material qualification test verification. In addition, if the selected material has not been successfully applied in civil aviation products, but is a mature material used in other aviation products, or the historical data corresponding to the selected material does not meet the above requirements, or the selected material is a new material, a separate material qualification test is required. In addition, for new materials, the chemical composition, performance and organization design, as well as the molding process and heat treatment process must be completed first, and then the material qualification test can be carried out.
[0052] It can be understood that the present invention first determines whether the historical data of the selected material meets the preset requirements. If it meets the requirements of furnace batch, number of sub-samples, traceability, complete records, and compliance with specifications of the formation process, there is no need to conduct a separate material qualification test verification, and the compliance statement report of the selected material can be directly obtained, that is, it is determined that the selected material has passed the material qualification appraisal; if the historical data of the selected material does not meet the preset requirements, a separate material qualification test verification is required to evaluate whether the selected material is qualified by means of test verification, thereby ensuring the accuracy and reliability of the material qualification appraisal.
[0053] Among them, Figure 2 As mentioned above, the material qualification test process includes the following contents:
[0054] Step S21: Classifying the selected materials into key component materials, important component materials and general component materials according to the key characteristics of engine components;
[0055] Step S22: Divide suppliers into key-level suppliers, important-level suppliers and general-level suppliers according to the importance of the parts they undertake and their technical capabilities, wherein key-level suppliers can undertake the qualification assessment of key parts materials, important parts materials and general parts materials, important-level suppliers can undertake the qualification assessment of important parts materials and general parts materials, and general-level suppliers can undertake the qualification assessment of general parts materials;
[0056] Step S23: Prepare the first draft of material standards and material production process control documents;
[0057] Step S24: The supplier performs the first batch production and inspection of key parts, important parts and general parts according to the material production process control documents and the first draft of the material standards, and obtains the first piece identification data;
[0058] Step S25: test and verify the key parts, important parts and general parts together with the whole machine to obtain the whole machine test data;
[0059] Step S26: Verify and revise each technical indicator specified in the first draft of the material standard based on the first piece inspection data and the whole machine test data.
[0060] Specifically, the selected materials are first divided into key parts materials, important parts materials and general parts materials according to the key characteristics of engine parts, and the material qualification work is carried out in stages. Then, suppliers are divided into key-level suppliers, important-level suppliers and general-level suppliers according to the importance of the parts they undertake and their technical capabilities. Among them, key-level suppliers can undertake the qualification of key parts materials, important parts materials and general parts materials, important-level suppliers can undertake the qualification of important parts materials and general parts materials, and general-level suppliers can undertake the qualification of general parts materials. According to the different levels of suppliers, management requirements need to be put forward to suppliers in different stages.
[0061] Then, the historical data of existing international standards, national standards, navigation marks and other basic standards are collected, sorted and classified, and the distribution of the collected data matrix is determined based on the principle of statistical analysis. Then, the material process-organization-performance data is systematically analyzed based on digital simulation technology, and a three-dimensional model is established. The material process, organization, and performance characteristics are formulated according to the actual working conditions of the engine. Based on the collected historical data, the virtual augmented sample method is used to obtain the technical indicators with statistical characteristics of the material, thereby forming a preliminary draft of the material standard. In addition, the applicability and rationality of various requirements of the compiled preliminary draft of the material standard can also be analyzed. If some contents of the material standard have not been substantially changed, a compliance statement can be made directly. If there are substantial changes, re-test verification is required, which is conducive to improving the efficiency of identification.
[0062] Next, a material production process control document (PCD) is prepared for the material production process. The PCD should include but not be limited to identifying all process and management links, clarifying the process flow, controlled process parameters, key characteristics, process procedures and controlled process equipment, etc., to provide a reference for material trial production, inspection and testing, and on-site witnessing of key procedures and key processes during material qualification identification.
[0063] Then, the supplier conducts the first batch of production and inspection of key parts, important parts and general parts according to the material production process control documents and the draft of material standards, and obtains the first piece identification data.
[0064] Among them, for key parts, all tests and inspection items specified in the material specification (i.e. the first draft of the material standard) must be carried out. The number and location of the cut specimens should include the locations specified in the material specification, and compared with the material specification, the sampling locations and number of samples need to be increased in dangerous parts under service conditions. Sampling for all test items needs to be cut from the actual part blanks. All tests must be carried out in accordance with ASTM standards. During the test, people, machines, materials, methods, environment and measurements must be recorded in detail and photos or videos must be retained. The test results must meet the requirements of the material specification. In addition, for key parts, we also characterize the surface state, chemical element composition distribution, element segregation, microstructure, mechanical properties and other multi-parameter states of the raw materials and blanks in the whole cross-section based on different characterization technologies, so as to obtain the basic data of the surface state, metallurgy, organization and mechanical properties of the raw materials and blanks in the production process, and conduct quantitative statistical distribution analysis on the surface state, chemical composition, element segregation, grain structure, alloy precipitation phase, inclusions, mechanical properties, etc. of the raw materials and blanks, and reconstruct the raw materials and blanks in three dimensions by computer to obtain the three-dimensional models of the surface state, chemical composition, microstructure and mechanical properties of the raw materials and blanks; then, we simulate and calculate the production process of the raw materials and blanks through simulation software to obtain the similar The same result as the above three-dimensional reconstruction model is obtained, and the production process is further optimized and iterated until various parameters such as surface state, chemical composition, microstructure and mechanical properties are uniform on each section; then, the obtained production process parameters are used to produce raw materials and blanks, and high-throughput test characterization of multiple parameters such as surface state, chemical element composition distribution, element segregation, microstructure, mechanical properties, etc. of each section is carried out, and the test results are used to correct the simulation results, so as to achieve precise control of the surface state, chemical composition, microstructure and mechanical properties of raw materials and blanks through simulation methods, and achieve uniformity of various parameters such as surface state, chemical composition, microstructure and mechanical properties of each section, thereby eliminating the hidden dangers of the structure caused by material heterogeneity.
[0065] For important parts, all tests and inspection items specified in the material specifications must be carried out. The number and locations of sample cuts should include the locations specified in the material specifications, and the sampling locations and number of samples need to be increased in dangerous parts under service conditions compared to the material specifications. Sampling for all test items needs to be cut from actual part blanks. All tests must be carried out in accordance with ASTM standards. During the test, people, machines, materials, methods, environment and measurements must be recorded in detail and photos or videos must be retained. The test results must meet the requirements of the material specifications. In addition, by characterizing the surface state, chemical element composition distribution, element segregation, microstructure, mechanical properties and other multi-parameter states of the raw materials and blanks in the entire cross-section based on different characterization techniques, the basic data of metallurgy, organization and mechanical properties in the production process of the raw materials and blanks are obtained, and the surface state, chemical composition, element segregation, grain structure, alloy precipitation phase, inclusions, mechanical properties and the like of the raw materials and blanks are quantitatively statistically analyzed. The raw materials and blanks are reconstructed in three dimensions by computer to obtain three-dimensional models of the surface state, chemical composition, microstructure and mechanical properties of the raw materials and blanks. Then, the production process of the raw materials and blanks is simulated and calculated by simulation software to obtain the three-dimensional models consistent with the above three-dimensional reconstruction. The same result as the model is obtained, and the production process is optimized and iterated until the parameters such as surface state, chemical composition, microstructure and mechanical properties are uniform on each section; then, the obtained production process parameters are used to produce raw materials and blanks, and high-throughput test characterization of multiple parameters such as surface state, chemical element composition distribution, element segregation, microstructure, mechanical properties, etc. of each section is carried out, and the simulation results are corrected, so as to achieve precise control of the surface state, chemical composition, microstructure and mechanical properties of raw materials and blanks through simulation methods, and achieve uniformity of various parameters such as surface state, chemical composition, microstructure and mechanical properties of each section, eliminating the hidden dangers of the structure caused by the non-uniformity of the material. In addition, for casings with containment requirements, high strain rate compression or tensile tests are also required. The compression or tensile stress-strain curve under high strain rate can be obtained through the following process: first, obtain quasi-static and compression or tensile test data at different temperature states and different high strain rates through experiments, and draw stress-strain curves; then use numerical simulation methods to simulate and calculate the stress-strain curves of materials at different temperatures; then compare the stress-strain curves calculated by the numerical simulation method with the stress-strain curves of the experimental test, and save the best simulation parameters; then adjust the temperature and strain rate, and use numerical methods to simulate and calculate the stress-strain curves of materials at high temperature and high speed under the best simulation parameter conditions; finally, based on the stress-strain curve, use the strain deviation method or the JC constitutive model to obtain the typical stress-strain curves and constitutive parameters of the material under different test conditions.
[0066] For general parts, all tests and inspection items specified in the material specification must be carried out. The number and location of the cut specimens should include the locations specified in the material specification. The sampling locations and number of samples can be increased as appropriate. Sampling for all test items must be cut from the actual part blanks. All tests must be carried out in accordance with ASTM standards. During the test process, people, machines, materials, methods, environment and measurements must be recorded in detail and photos or videos must be retained. The test results must meet the requirements of the material specification. In addition, by characterizing the surface state, chemical element composition distribution, element segregation, microstructure, mechanical properties and other multi-parameter states of the raw materials and blanks in the entire cross-section based on different characterization techniques, the basic data of metallurgy, organization and mechanical properties in the production process of the raw materials and blanks are obtained, and the surface state, chemical composition, element segregation, grain structure, alloy precipitation phase, inclusions, mechanical properties and the like of the raw materials and blanks are quantitatively statistically analyzed. The raw materials and blanks are reconstructed in three dimensions by computer to obtain three-dimensional models of the surface state, chemical composition, microstructure and mechanical properties of the raw materials and blanks. Then, the production process of the raw materials and blanks is simulated and calculated by simulation software to obtain the three-dimensional models consistent with the above three-dimensional reconstruction. The same result as the model is obtained, and the production process is optimized and iterated until the parameters such as surface state, chemical composition, microstructure and mechanical properties are uniform on each section; then, the obtained production process parameters are used to produce raw materials and blanks, and high-throughput test characterization of multiple parameters such as surface state, chemical element composition distribution, element segregation, microstructure, mechanical properties, etc. of each section is carried out, and the simulation results are corrected, so as to achieve precise control of the surface state, chemical composition, microstructure and mechanical properties of raw materials and blanks through simulation methods, and achieve uniformity of various parameters such as surface state, chemical composition, microstructure and mechanical properties of each section, eliminating the hidden dangers of the structure caused by the non-uniformity of the material.
[0067] In addition, material suppliers shall produce in accordance with PCD documents. When key procedures or key processes are involved in the production process, material suppliers shall invite the review representative of the Civil Aviation Airworthiness Certification Center of China and the inventor of the Type Certificate (TC) to conduct on-site witnessing. For non-conformities and suggestions found during the on-site witnessing, the supplier shall complete the rectification within half a month and submit supporting materials to the civil aviation review representative and TC inventor.
[0068] Optionally, for key and important materials, after the first article inspection is completed, the supplier must continue to produce 3 batches of materials in 3 furnaces in accordance with the material production process control documents, and conduct multiple batches of performance evaluation tests in accordance with the same requirements as the first article inspection. After the multiple batches of performance evaluation tests are completed and qualified, a multiple batch performance evaluation report is compiled. If the 3 furnaces and 3 batches are unqualified, the material supplier needs to analyze the reasons for the unqualified, provide solutions, and verify whether the solutions are effective. If effective, the PCD file is corrected and re-produced, and the multiple batches of performance evaluation work is completed again; if the revised PCD file involves process changes, equipment changes, or other important parameter changes, the first article inspection work must also be repeated.
[0069] Optionally, after key and important materials have passed multiple batches of performance evaluation tests, they also need to undergo quality stability evaluation. Specifically, first collect historical data of the same material in different periods, batches, and models, as well as data obtained from first-piece identification and multiple batches of performance evaluation tests, and classify and manage them. Then, use the principle of data statistical analysis to determine the distribution of the data matrix. Then, based on big data and digital simulation technology, train the data to obtain the organization and performance characteristics that meet the process characteristics of the material, and establish a three-dimensional model of process-organization-performance, thereby forming criteria for the evaluation of the quality stability of multiple batches of materials. Using the criteria for the evaluation of the quality stability of multiple batches of materials, the stability of key and important materials that have passed the first-piece identification and multiple batches of performance evaluation is evaluated. If they meet the design requirements, they enter the next verification stage; if they do not meet the design requirements, analyze the reasons, improve them, and re-verify them until the stability meets the design requirements and enters the next verification stage.
[0070] Optionally, for key component materials, it is also necessary to conduct long-term performance verification of the materials in combination with the actual engine operating conditions. Specifically, a long-term performance verification test spectrum of the materials is compiled based on the actual engine test spectrum, and standard specimens are used to conduct tests on the long-term organizational stability, mechanical properties, etc. of the materials. The test time is the total life of the engine design, and the evolution law between the organization and performance of the materials under long-term use is obtained; the standard specimens are subjected to the same surface treatment as the parts, and the long-term organizational stability, mechanical properties, etc. of the materials are tested according to the above-mentioned long-term performance verification test spectrum of the materials. The test time is the total life of the engine design, and the evolution law between the organization and performance of the materials under long-term use is obtained due to the surface treatment; on the basis of the above-mentioned long-term performance verification test spectrum of the materials, a corrosion test spectrum is added to obtain a composite test spectrum, and standard specimens are used to conduct tests on the long-term organizational stability, mechanical properties, etc. of the materials according to the composite test spectrum. The test time is the total life of the engine design, and the influence of corrosion on the evolution between the organization and performance of the materials under long-term use is obtained. If the evolution law between the material's organization and performance under long-term use, the evolution law between the material's organization and performance under long-term use due to surface treatment, and the influence of corrosion on the evolution between the material's organization and performance under long-term use all meet the requirements, the material is judged to have passed the long-term performance verification.
[0071] Optionally, for key component materials, it is also necessary to design structural simulation components and conduct long-term performance verification of the structural simulation components. Figure 3 As shown, the process of long-term performance verification of the structural simulation component includes the following contents:
[0072] Step S101: designing a structural simulation part according to the structural characteristics of the key parts and the temperature and load conditions during service, and optimizing and iterating through simulation to make the structural characteristics and load conditions of the structural simulation part consistent with the real parts;
[0073] Step S102: Processing the structural simulation parts, and using a multi-level gradient interruption heat treatment process to eliminate the residual stress in the processing of the structural simulation parts, so as to avoid the residual stress in the processing affecting the test results of the simulation parts; wherein, for the blanks of larger sizes such as disks and shafts, directly cut the specimens from the blanks of disks and shafts and machine them into the size of the simulation parts, and for the simulation parts of thin-walled parts such as blades and casings, it is necessary to redesign the mold with the thin-walled parts such as blades and casings, prepare the simulation parts blanks according to the same preparation process as the blades and casings, and finally process them into the size requirements of the simulation parts;
[0074] Step S103: compile a test spectrum of the structural simulation part according to the engine test spectrum, and perform a long-term test on the heat-treated structural simulation part according to the test spectrum of the structural simulation part to obtain the evolution law between the organization and performance of the structural simulation part under the condition of long-term use;
[0075] Step S104: Add a corrosion test spectrum based on the test spectrum of the structural simulation part, and perform a long-term test on the heat-treated structural simulation part according to the composite test spectrum to obtain the influence of corrosion on the evolution of the organization and performance of the structural simulation part under long-term use conditions.
[0076] The process of eliminating residual stress during the processing of the structural simulation part by using a multi-level gradient interruption heat treatment process includes the following:
[0077] Place the structural simulation part in a high-pressure vacuum heat treatment furnace, and raise the temperature in the furnace to 1010-1080°C at a heating rate of 10-20°C / min;
[0078] Insulation: keep the structural simulation parts at 1010-1080℃ for 1-2h;
[0079] Cooling: cooling the structural simulation part to room temperature. The cooling method may be water cooling, air cooling or oil cooling;
[0080] Pretreatment: placing the structural simulation part in a high-pressure vacuum heat treatment furnace, raising the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keeping it warm for 0.5-1h, and then cooling the structural simulation part to room temperature; placing the structural simulation part back in the high-pressure vacuum heat treatment furnace, raising the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keeping it warm for 0.5-1h, and then cooling the structural simulation part to room temperature; placing the structural simulation part back in the high-pressure vacuum heat treatment furnace again, raising the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keeping it warm for 0.5-1h, and then cooling the structural simulation part to room temperature. The cooling method may be any one of water cooling, air cooling or oil cooling;
[0081] Interrupt the high-temperature treatment, put the structural simulation into a high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation to room temperature; put the structural simulation back into the high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation to room temperature; put the structural simulation back into the high-pressure vacuum heat treatment furnace again, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation to room temperature. The cooling method can be any one of water cooling, air cooling or oil cooling;
[0082] For the final aging heat treatment, the structural simulation is placed in a high-pressure vacuum heat treatment furnace, and the temperature in the furnace is increased to 620-650°C at a heating rate of 30-50°C / min. After keeping it warm for 2h-4h, the structural simulation is cooled to room temperature; the structural simulation is placed back in the high-pressure vacuum heat treatment furnace, and the temperature in the furnace is increased to 620-650°C at a heating rate of 30-50°C / min. After keeping it warm for 2h-4h, the structural simulation is cooled to room temperature; the structural simulation is placed back in the high-pressure vacuum heat treatment furnace again, and the temperature in the furnace is increased to 620-650°C at a heating rate of 30-50°C / min. After keeping it warm for 2h-4h, the structural simulation is cooled to room temperature. The cooling method can be any one of water cooling, air cooling or oil cooling.
[0083] If the evolution law between the organization and performance of the structural simulation parts under long-term use conditions and the influence of corrosion on the evolution of the organization and performance of the structural simulation parts under long-term use conditions meet the requirements, the material is judged to have passed the long-term use performance verification of the structural simulation parts.
[0084] Optionally, for key and important parts materials, component / environmental test verification is also required. Specifically, low cycle, over-rotation cracking, over-temperature, vibration, containment test and corrosion sensitivity test are carried out on key and important parts. The test time should exceed the total life of the engine design, so as to verify whether the parts processed by key and important parts materials meet the design requirements.
[0085] Optionally, for stainless steel materials that need to be used in harsh marine environments, it is also necessary to predict the corrosion rate evolution law. Specifically, the corrosion rate evolution law prediction process includes the following contents:
[0086] 1) Determine the relevant influencing factors in the natural corrosion process of stainless steel;
[0087] The relevant influencing factors that may exist in the natural corrosion process of stainless steel are analyzed and determined from the perspectives of internal factors, external factors, and process factors. In the natural corrosion process of stainless steel, internal factors include the content of stainless steel components, such as Fe content, Cr content, Ni content, Ti content, etc.; external factors include the chloride ion content, temperature, humidity, pH, etc. in the natural environment where the stainless steel is located; process factors include the length of time the stainless steel is placed in the natural environment, etc. Based on the determined relevant influencing factors, a data set of influencing factors of natural corrosion of stainless steel is formed.
[0088] (2) The PCA method was used to reduce the dimensionality of the data set of factors affecting natural corrosion of stainless steel;
[0089] For the data set X = {x1, x2, ..., x n}, where n is the total amount of data, and each piece of data x i It contains m components (i.e., the dimension is m), representing the data corresponding to each factor. The specific operations of using the principal component analysis (PCA) method to reduce the dimension of the data set are as follows.
[0090] a. Preprocess data;
[0091] A centralized preprocessing method is used for all data in the corrosion influencing factor dataset. The average value of each column attribute is subtracted from all data, so that the average value of each column attribute data after preprocessing is 0. The calculation formula is as follows: (The preprocessed data set is recorded as
[0092]
[0093] b. Solve the eigenvalues and eigenvectors;
[0094] Calculate the covariance matrix of the preprocessed dataset And perform eigenvalue decomposition on the covariance matrix to obtain the eigenvalues λ1≥λ2≥...≥λ sorted from large to small m and their corresponding eigenvectors ω1, ω2, ..., ω m .
[0095] c. Set the reconstruction threshold t or the dimension m′ after dimensionality reduction;
[0096] The dimension after dimensionality reduction is determined in two ways: one is determined by the reconstruction threshold t, and the other is directly set to m′. The method determined by the reconstruction threshold t is as follows: the cumulative eigenvalues are calculated in descending order, and the proportion of the total eigenvalues is called the cumulative contribution rate t i ,Right now If the cumulative contribution rate corresponding to the current eigenvalue is t m′ If it just exceeds the reconstruction threshold t, the dimension after dimensionality reduction is set to m′.
[0097] d. Construct the projection matrix W;
[0098] According to the dimension m′ after dimensionality reduction, take the first m′ eigenvalues λ1, λ2, ..., λ m′ The corresponding eigenvectors ω1, ω2, ..., ω m′ Construct the projection matrix W = (ω1, ω2, ..., ω m′ ).
[0099] e. Forming a reduced-dimensional dataset
[0100] Through the projection matrix W, we can obtain the reduced dimension dataset Dataset The dimension of is m′<m, that is, each piece of data has fewer components.
[0101] For the parameter setting of the PCA method, it is recommended that the reconstruction threshold t be 0.9, or the dataset m′≤0.6m after dimensionality reduction.
[0102] (3) Use the RF model to simulate and predict the evolution of the natural corrosion rate of stainless steel;
[0103] The dataset after dimension reduction based on the influencing factors of natural corrosion of stainless steel And the corresponding stainless steel corrosion rate set Y = {y1, y2, ..., y n}, which constitutes the data set for RF model training Each piece of data x i Containing m' components (i.e., the dimension is m'), the RF model is trained to simulate and predict the evolution law of the natural corrosion rate of stainless steel. The specific operations are as follows:
[0104] a. Set RF model parameters;
[0105] Determine the number of base learners (decision trees) n tree And the number of random attributes to be extracted k.
[0106] b. Construct a base learner sub-training set;
[0107] For each base learner, a bootstrap sampling method is used to extract n data with replacement from a data set Z containing n data to form a sub-training set Z t , the unsampled data constitutes the sub-validation set Z v .
[0108] c. Training of base learners;
[0109] Using each sampled sub-training set Z t Conduct sub-decision tree training. During the attribute selection and division training of each node of the sub-decision tree, the attribute set of each node randomly generates a subset containing k attributes, and then determines the optimal attribute from the subset for node division. Through different sub-training sets and different attribute subsets, n tree Sub-decision trees (base learners).
[0110] d. Base learner result integration method;
[0111] The RF model takes the average of the prediction results of each sub-decision tree to obtain the final regression result of the model.
[0112] e. Model evaluation;
[0113] Using sub-validation set Z v, verify and compare the corrosion rate predicted by the RF model with the actual corrosion rate, and evaluate and judge whether the model performance meets the requirements. Use various regression performance indicators, such as root mean square error (RMSE) and goodness of fit (R-squared), to evaluate the accuracy and generalization ability of the model. If the indicators do not meet the requirements, return to step a to reset the parameters and perform subsequent steps to update the RF model; if the indicators meet the requirements, the RF model that can finally describe the evolution law of stainless steel corrosion rate is obtained.
[0114] For the parameter setting of the RF model, the recommended number of base learners (decision trees) is n tree =100, the number of random attributes to be extracted is k=log2 m or k=m / 3, and the base learner result integration method adopts the average method.
[0115] For the evaluation of RF model, the indicators can be root mean square error RMSE, goodness of fit R 2 The RMSE calculation formula is as follows: 2 The calculation formula is as follows:
[0116]
[0117]
[0118] in, is the corrosion rate predicted by the RF model, is the true mean corrosion rate. It is recommended that when the RMSE index satisfies or R 2 The index meets R 2 When ≥0.65, the model training ends and the final RF model is obtained.
[0119] Then, the key parts, important parts and general parts are tested and verified together with the whole machine. The whole machine test is carried out in accordance with the airworthiness certification requirements. The test time should exceed the total design life of the engine, so as to verify whether the materials and components verified by the previous steps meet the design requirements of the whole machine.
[0120] Finally, statistical analysis is performed on the first-article identification data, multi-batch performance evaluation data, long-term use performance data, simulated part long-term use performance data, component test data, and whole machine test data, and each technical indicator specified in the preliminary draft of the material standard is verified. The technical indicators in the preliminary draft of the material standard are further revised to form the final draft of the material standard to reach an approvable status.
[0121] It can be understood that the material qualification test of the present invention performs statistical analysis and verification of material standards based on material qualification data, and strictly manages the material production process. The material is fully verified and the quality stability and consistency are evaluated. The obtained material data is highly reliable, forming sufficient evidence to show that the materials produced by the material supplier can continuously and stably meet the material specification requirements, which can ensure that the civil aviation engine materials pass the airworthiness certification smoothly. In addition, the material quality stability is evaluated by using big data and digital simulation technology, and a three-dimensional model of process-organization-performance is established. The material quality stability evaluation criteria and methods that meet the airworthiness requirements are established, providing support for the quality consistency and stability evaluation of materials in the process of civil aviation engine material qualification. In addition, the key parts materials and important parts materials have been verified by multi-level tests of standard specimens, structural simulation parts, parts and complete machine tests, and the materials have been fully verified. In addition, by characterizing and analyzing the distribution laws of chemical composition, microstructure and mechanical properties through high-throughput testing, the production process flow of materials and the processing process flow of parts can be optimized, and data support can be provided for the design optimization of civil aviation engines.
[0122] It can be understood that in step S3, if the amount of historical data of the selected material meets the requirements, its compliance statement report is directly obtained, that is, the selected material is judged to have passed the material qualification appraisal; and for the selected material that needs to undergo material qualification appraisal test, the corresponding final draft of the material standard is obtained, and the PCD file of the material supplier is formally approved, indicating that the supplier can stably batch produce, and a catalog of qualified material products is established. After the verification of the final draft of the material standard is completed and approved, the selected material is judged to have passed the qualification appraisal.
[0123] In addition, titanium alloy, as a new material commonly used in aircraft engines, has been widely used in aircraft engines, such as titanium alloy blades, Figure 4 As shown, another embodiment of the present invention also provides a method for designing a titanium alloy composition with high comprehensive performance, which is applicable to the composition design of titanium alloy materials, and includes the following contents:
[0124] S100: Establishing an initial data set, based on historical data of titanium alloys, establishing an initial data set including titanium alloy composition, material descriptors reflecting intrinsic properties of the material, and titanium alloy properties, wherein the titanium alloy properties include at least two of 600°C high temperature strength, elongation, creep strength, and fatigue life;
[0125] S200: Key material descriptor screening, which gradually screens data through Pearson correlation screening, recursive elimination and feature weight sorting to determine the key material descriptors that affect the performance of each titanium alloy;
[0126] S300: Establishing a training data set, establishing a training data set including titanium alloy composition, key material descriptors, and titanium alloy performance data;
[0127] S400: Establishing a prediction model, taking titanium alloy composition and key material descriptors as inputs, taking at least two titanium alloy properties as outputs, establishing at least two composition-property prediction models, and training the at least two composition-property prediction models using a training data set;
[0128] S500: Multi-objective optimization, using a multi-objective optimization solution algorithm to iteratively and collaboratively optimize and solve at least two component-performance prediction models to obtain a titanium alloy composition with the best comprehensive performance.
[0129] It can be understood that the present invention proposes a machine learning composition rational design strategy that combines material descriptors, multi-step key feature screening, and multi-objective optimization solution algorithms, which can solve the problems of the traditional "trial and error method" with long R&D cycle, high cost, and difficulty in finding multi-objective optimal solutions, and realize the rapid and low-cost design of new titanium alloys with multi-objective performance synergistic enhancement under the background of large composition space. In addition, by screening key features through Pearson correlation screening, recursive elimination, and feature importance sorting, the key material descriptors closely related to the performance of each titanium alloy can be quickly and accurately determined and used as input features, which improves the prediction accuracy, generalization ability, and interpretability of the model, and provides a reference for the rational design of high-performance titanium alloy components.
[0130] Optionally, in step S100, the titanium alloy composition includes at least one of Ti, Al, Sn, Zr, Mo, Nb, Ta, Si, W, Hf, Nd, Y, and C elements, and the material descriptors reflecting the intrinsic properties of the material include at least one of the alloy theoretical melting point, valence electron concentration, electron concentration, electronegativity, Poisson's ratio, shear modulus, Young's modulus, compression modulus, a lattice constant, b lattice constant, c lattice constant, mixing enthalpy, and lattice distortion energy, and the titanium alloy properties include but are not limited to 600°C high temperature strength (UTS), elongation (EL), creep strength (σ c ) and fatigue life (N), etc.
[0131] It can be understood that in step S200, the step of screening key material descriptors specifically includes the following contents:
[0132] Pearson correlation screening: calculate the Pearson coefficient between any two material descriptors. If the correlation coefficient is greater than 0.95, one feature is removed from the two and the other is retained, and finally n key material descriptors are obtained;
[0133] Recursive elimination: n key material descriptors selected by titanium alloy composition and Pearson correlation are used as input, and at least two titanium alloy properties are used as output. For each titanium alloy property, n RF machine learning models are established, and the extracted features (i.e., key material descriptors) corresponding to the minimum model error are removed, and n-1 features remain. Then, the above feature elimination is repeated until the minimum error changes from decreasing to being stable.
[0134] The feature weights are sorted, and the titanium alloy composition and the key material descriptors remaining after recursive elimination are taken as input. The performance of each target titanium alloy is output. The RF machine learning model is established to calculate the importance of each feature, and the top 4 to 8 key material descriptors are selected as the final key material descriptors, thereby determining the key material descriptors that are closely related to the performance of each titanium alloy.
[0135] It can be understood that the present invention performs key feature screening through Pearson correlation screening, recursive elimination and feature importance sorting, quickly and accurately determines the key material descriptors closely related to the performance of each titanium alloy and uses them as input features, thereby improving the prediction accuracy, generalization ability and interpretability of the model, and providing a reference for the rational design of high-performance titanium alloy compositions.
[0136] It can be understood that in step S300, based on the key material descriptors of each titanium alloy performance screened out in step S200, a training data set including titanium alloy composition, key material descriptors, and titanium alloy performance data is established, wherein a training data set is constructed for each prediction model.
[0137] It can be understood that in step S400, the initial model of each component-performance prediction model is an XGBoost model, and its general function is formula (3):
[0138]
[0139] In the formula, f i p represents the prediction result of sample i after the pth iteration, f i p-1 represents the prediction results of the first p-1 decision trees, f i (x i ) represents the model of the i-th decision tree.
[0140] Each training data set is randomly divided into a training set and a test set according to a ratio of 8:2. The component-prediction model is trained using the corresponding training set. During the training process, the hyperparameters of the XGBoost initial model are doubly optimized using the Grid Search and K-Fold CrossValidation algorithms to obtain the XGBoost improved model, where K=5 and 10. The hyperparameters include the number of trees (n_estimator), the learning rate (learning_rate), and the maximum tree depth (max_depth). The optimization ranges of the hyperparameters are 1-1000, 0.01-0.5, and 1-10, respectively.
[0141] Then, the corresponding test set is used to evaluate the prediction accuracy and optimize the design of the XGBoost improved model to obtain the best XGBoost model. The evaluation indicators used are mean absolute error (MAE), root mean square error (RMSE), and determination coefficient (R). 2 , as shown in equations (4)-(6):
[0142]
[0143]
[0144]
[0145] In the formula, is the model’s predicted value for the sample, y i is the true value of the sample, is the average of the true values of the samples, and n is the total number of samples.
[0146] It can be understood that in step S500, the multi-objective optimization solution algorithm adopts a non-dominated sorting genetic evolutionary algorithm (NSGA II), and the optimization target is as shown in formula (7). Weight coefficients a, b, ..., m are introduced into the non-dominated sorting genetic evolutionary algorithm to respectively represent the importance of different titanium alloy properties. According to the actual needs of different application scenarios, the values of a, b, ..., m are adjusted to optimize the properties of various titanium alloys to different degrees, so as to obtain a titanium alloy composition with the best comprehensive performance, and perform experimental verification and iterative optimization;
[0147]
[0148] Among them, f1, f2, ..., f mis the function expression of m optimization objectives; UTS, EL, ..., M are the fitness functions for optimizing the performance of different titanium alloys, i.e., multiple performance prediction models established in step S400; max and min represent the maximum and minimum values of each performance, respectively, and are set according to actual needs; a, b, ..., m are the weight coefficients of the m optimization objectives, respectively, representing the importance of the m target alloy properties, where m ≥ 2. According to the actual needs of different application scenarios, the weights of the optimization objectives are flexibly adjusted to adjust the optimization focus, giving priority to meeting the most important objectives while taking into account other target performances, and finally providing an alloy design solution that meets specific application needs.
[0149] Specifically, the NSGA-II algorithm selects 3 to 10 solutions with the best performance from the obtained Pareto front solutions (Pareto) for experimental verification. If the error is less than 5%, the titanium alloy composition with the smallest relative error is taken as the optimal composition. If the relative error is greater than 5%, the experimental results are put into the training set as training data to iteratively optimize the model until the relative error between the predicted value and the experimental value is less than 5%. At this time, a new titanium alloy composition with excellent comprehensive performance can be obtained.
[0150] In addition, another embodiment of the present invention further provides a method for preparing a novel titanium alloy blade, which adopts the titanium alloy composition designed in the above embodiment, and comprises the following steps:
[0151] Step 1: Ingot smelting, smelting the titanium alloy composition obtained according to the titanium alloy composition design method of the above embodiment to obtain an alloy ingot;
[0152] Step 2: Forging, forging the alloy ingot in step 1 to obtain a forging;
[0153] Step three: heat treatment, subjecting the forgings described in step two to solid solution aging heat treatment to obtain titanium alloy blades.
[0154] Optionally, the impurity content of the titanium alloy component in the step one is O<0.1%, N<0.01%, and Fe<0.1%; the smelting temperature is 1850-2350°C, and the time from heating and smelting to cooling and taking out of the furnace is 3-8h; the smelting is carried out in a vacuum consumable arc furnace, the vacuum degree of the vacuum consumable arc furnace is ≤1Pa, the smelting current is controlled at 3-6kA according to the ingot size, and the smelting voltage is 23-30V.
[0155] Optionally, the forging process in step 2 includes:
[0156] Open the billet, heat the alloy ingot to 1000-1150℃, keep it warm for 6-12h, then take it out of the furnace for forging to form a billet with a forging ratio of ≥4;
[0157] Free forging, subjecting the blank to 2-4 fires of upsetting and drawing deformation at 20-60°C above the β transformation point to form a forging blank, wherein the upsetting and drawing ratio of each fire is not less than 3, and the total forging ratio is ≥10;
[0158] Finish forging, grinding the obtained forging blank, heating it to 10-50°C above the β transformation point, and performing 2-4 rounds of finish forging to form a bar;
[0159] Die forging: placing the rod into a matching die, heating it to 10-50° C. above the β phase transformation point, and performing a fire forming to obtain a new type of titanium alloy blade.
[0160] Optionally, the solution temperature of the solution aging heat treatment in step three is 1000°C to 1060°C, the holding time is 0.5 to 1.5 hours, and the cooling method is water cooling; the aging temperature is 650°C to 800°C, the holding time is 5 to 10 hours, and the cooling method is air cooling.
[0161] In an embodiment of the present invention, a heat treatment process combining solution treatment and aging treatment is adopted to regulate the blade microstructure, and a special Widmanstätten structure with equiaxed ɑ phase intercalated in lamellar ɑ+β phase is designed, thereby obtaining a new titanium alloy blade with high thermal strength, creep resistance and fatigue performance.
[0162] Optionally, the microstructure in the new titanium alloy blade is a special Widmanstatten structure in which equiaxed ɑ phase is interspersed in lamellar ɑ+β phase, wherein the lamellar ɑ+β phase has a length of about 10 to 25 μm and an aspect ratio of 5 to 15; the equiaxed ɑ phase is distributed therein with a length of about 30 to 45 μm and an aspect ratio of 3.5 to 6.4, and some equiaxed ɑ phases are connected together, and the content of the equiaxed ɑ phase is between 20 and 30%.
[0163] The following examples more specifically describe the titanium alloy composition design and titanium alloy blade preparation disclosed in the present invention, and these examples are only for illustrative purposes.
[0164] Example 1
[0165] 1) Establish an initial data set: establish alloy composition, alloy properties and material descriptors that reflect the intrinsic characteristics of the material; the alloy composition includes Ti, Al, Sn, Zr, Mo, Nb, Ta, Si, W, Mo, Y, Nd, C, the alloy properties include 600℃ high temperature strength (UTS) and elongation (EL), and the material descriptors include alloy theoretical melting point, valence electron concentration, electron concentration, electronegativity, Poisson's ratio, shear modulus, Young's modulus, compression modulus, a lattice constant, b lattice constant, c lattice constant, mixing enthalpy, and lattice distortion energy.
[0166] 2) Determine the key material descriptor combination that affects the 600℃ high temperature strength (UTS) and elongation (EL) properties of titanium alloy; among them, shear modulus, c lattice constant, valence electron concentration, and Poisson's ratio are finally selected as key material descriptors through step-by-step Pearson correlation screening, recursive elimination, and feature importance sorting.
[0167] 3) Establishing a data set: Establishing a data set based on the above alloy composition, the selected key material descriptors, and the alloy's 600°C high temperature strength (UTS) and elongation (EL).
[0168] 4) Establishing alloy composition, key material descriptor-high temperature strength prediction model and alloy composition, key material descriptor-elongation prediction model respectively;
[0169] The training process of the high temperature strength prediction model is as follows: alloy composition and selected key material descriptors are used as input data, 600℃ high temperature strength of alloy is used as output data, 80% of the total data volume is used as training set, and the remaining 20% is used as test set. The XGBoost model is used to train the initial high temperature strength prediction model, and then the Grid Search and K-Fold Cross Validation algorithms are used to double optimize the hyperparameters of the XGBoost initial high temperature strength prediction model to obtain the XGBoost optimal high temperature strength prediction model, where K = 5, XGBoost hyperparameters n_estimator = 680, learning_rate = 0.3, max_depth = 6, and the determination coefficient R of the optimized model is 2.34. 2 Reached 0.961;
[0170] The training process of the elongation prediction model is as follows: alloy composition and the selected key material descriptors are used as input data, the alloy elongation at 600℃ is used as output data, 80% of the total data volume is used as the training set, and the remaining 20% is used as the test set. The XGBoost model is used to train the initial elongation prediction model, and then the Grid Search and K-Fold CrossValidation algorithms are used to double optimize the hyperparameters of the XGBoost initial elongation prediction model to obtain the XGBoost optimal elongation model, where K = 10, XGBoost hyperparameters n_estimator = 850, learning_rate = 0.06, max_depth = 8, and the determination coefficient R of the optimized model is 2 Reached 0.957.
[0171] 5) Obtain the titanium alloy composition with the best comprehensive performance: Based on the XGBoost optimal high temperature strength prediction model and the XGBoost optimal elongation prediction model, the NSGA-Ⅱ algorithm is used to perform multi-objective optimization with the maximum high temperature strength and maximum elongation as the optimization targets. The weight coefficients of the two are 0.6 and 0.4, respectively. The first five solutions of the Pareto front solution are used as experimental verification objects, and the relative errors between the measured 600℃ tensile strength and elongation and the predicted values are calculated. If the error is less than 5%, the alloy composition with the smallest relative error is the optimal composition. If the relative error is greater than 5%, the experimental results are put into the training set as training data to continue iterative optimization of the model until the relative error between the predicted value and the experimental value is less than 5%. The final optimized alloy composition and properties are as follows: Ti-5.7Al-4Sn-3.5Zr-0.3Mo-0.4Si-0.3Nb-2.0Ta-0.8W-0.05C, high temperature tensile strength 710MPa, elongation 20.5%.
[0172] 6) Ingot smelting: The raw materials are smelted according to the above-mentioned component ratio to obtain an alloy ingot; wherein the O content in the raw materials is 0.062%, the N content is 0.003%, and the Fe content is 0.045%. The smelting is carried out in a vacuum consumable arc furnace, the smelting temperature is 2200°C, the heating and smelting time to cooling and furnace discharge is 6h, the vacuum degree of the vacuum consumable arc furnace is 1Pa, the smelting current is 5kA, and the smelting voltage is 25V.
[0173] 7) Forging: The alloy ingot is forged, which includes the following four steps, wherein the blanking process parameters are: heating temperature 1120°C, holding time 8h, forging ratio 9, to obtain a blank; the free forging process parameters are: the blank is subjected to 4 times of upsetting and drawing deformation at 1100°C, each time the upsetting and drawing ratio is 4, to obtain a forged blank; the fine forging process parameters are: the forging blank is subjected to 4 times of fine forging at 1080°C to obtain a rod; the die forging process parameters are: the rod is loaded into a matching die, and the rod is formed by one fire at 1080°C to obtain a titanium alloy blade.
[0174] 8) Heat treatment: The titanium alloy blades obtained as above are subjected to solution aging heat treatment to finally obtain new titanium alloy blades with high thermal strength and high elongation; the solution temperature is 1040°C, the holding time is 1h, and the cooling method is water cooling; the aging temperature is 720°C, the holding time is 8h, and the cooling method is air cooling.
[0175] The titanium alloy blades made using the above-mentioned optimized alloy composition have a special Widmanstatten structure with equiaxed ɑ phase interspersed in the lamellar ɑ+β phase, wherein the lamellar ɑ+β phase is about 18μm long and has an aspect ratio of about 12; the equiaxed ɑ phase is distributed therein with a length of about 34μm and an aspect ratio of 5.2, and some equiaxed ɑ phases are connected together, with the content of equiaxed ɑ phase being about 24%; the high-temperature strength and elongation are 740MPa and 24.5%, respectively.
[0176] Example 2
[0177] 1) Establish the initial data set: establish alloy composition, alloy properties and material descriptors reflecting the intrinsic properties of the material; the alloy composition includes Ti, Al, Sn, Zr, Mo, Nb, Ta, Si, W, Mo, Y, Nd, C, and the alloy properties include 600℃ high temperature strength (UTS), elongation (EL), creep strength (σ c ), material descriptors include alloy theoretical melting point, valence electron concentration, electron concentration, electronegativity, Poisson's ratio, shear modulus, Young's modulus, compression modulus, a lattice constant, b lattice constant, c lattice constant, mixing enthalpy, and lattice distortion energy.
[0178] 2) Determine the factors affecting the 600℃ high temperature strength (UTS), elongation (EL), creep strength (σ c ) of the key material descriptors; among them, compression modulus, Young's modulus, shear modulus, c lattice constant, valence electron concentration and Poisson's ratio were finally selected as the key material descriptors through step-by-step Pearson correlation screening, recursive elimination and feature importance sorting.
[0179] 3) Establish a data set: Based on the above alloy composition, the selected key material descriptors and the alloy 600℃ high temperature strength (UTS), elongation (EL), creep strength (σ c ) creates a data set for the content.
[0180] 4) Establish titanium alloy composition, key material descriptor-high temperature strength prediction model, titanium alloy composition, key material descriptor-elongation prediction model, and titanium alloy composition, key material descriptor-creep strength prediction model respectively;
[0181] The high temperature strength prediction model establishment process is as follows: alloy composition and selected key material descriptors are used as input data, 600℃ high temperature strength of alloy is used as output data, 80% of the total data volume is used as training set, and the remaining 20% is used as test set. The initial high temperature strength prediction model is trained with XGBoost model, and then Grid Search and K-FoldCross Validation algorithms are used to double optimize the hyperparameters of XGBoost initial high temperature strength prediction model to obtain the XGBoost optimal high temperature strength prediction model, where K=5, XGBoost hyperparameters n_estimator=1000, learning_rate=0.1, max_depth=5, and the determination coefficient R of the optimized model is 2 Reached 0.969;
[0182] The process of establishing the elongation prediction model is as follows: alloy composition and the selected key material descriptors are used as input data, alloy elongation is used as output data, 80% of the total data volume is used as the training set, and the remaining 20% is used as the test set. The XGBoost model is used to train the initial elongation prediction model, and then the Grid Search and K-Fold CrossValidation algorithms are used to double optimize the hyperparameters of the XGBoost initial elongation prediction model to obtain the XGBoost optimal elongation model, where K = 10, XGBoost hyperparameters n_estimator = 500, learning_rate = 0.2, max_depth = 4, and the determination coefficient R of the optimized model is 0. 2 Reached 0.971;
[0183] The creep strength prediction model establishment process is as follows: alloy composition and the selected key material descriptors are used as input data, alloy creep strength is used as output data, 80% of the total data volume is used as the training set, and the remaining 20% is used as the test set. The XGBoost model is used to train the initial creep strength prediction model, and then the Grid Search and K-Fold CrossValidation algorithms are used to double optimize the hyperparameters of the XGBoost initial creep strength prediction model to obtain the XGBoost optimal creep strength model, where K = 10, XGBoost hyperparameters n_estimator = 900, learning_rate = 0.05, max_depth = 6, and the determination coefficient R of the optimized model is 0. 2 Reached 0.962.
[0184] 5) Obtain the titanium alloy composition with the best comprehensive performance: Based on the XGBoost optimal high temperature strength prediction model, the XGBoost optimal elongation prediction model and the XGBoost optimal creep strength prediction model, the NSGA-Ⅱ algorithm is used to perform multi-objective optimization with the maximum high temperature strength, maximum elongation and maximum creep strength as the optimization targets. The weight coefficients of the three are 0.3, 0.3 and 0.4 respectively. The first five solutions of the Pareto front solution are used as experimental verification objects to calculate the measured 600℃ tensile strength, elongation and creep strength and the predicted The relative error of the value is calculated. If the error is less than 5%, the alloy composition with the smallest relative error is taken as the optimal composition. If the relative error is greater than 5%, the experimental results are put into the training set as training data to continue iterative optimization of the model until the relative error between the predicted value and the experimental value is less than 5%. The final optimized alloy composition and properties are as follows: Ti-5.7Al-4Sn-3.5Zr-0.5Mo-0.4Si-0.3Nb-2.0Ta-1W-0.1Y, high temperature tensile strength 700MPa, elongation 25%, creep strength 360MPa.
[0185] 6) Ingot smelting: The raw materials are smelted according to the above-mentioned component ratio to obtain an alloy ingot; wherein the O content in the raw materials is 0.062%, the N content is 0.003%, and the Fe content is 0.042%. The smelting is carried out in a vacuum consumable arc furnace, the smelting temperature is 2300°C, the heating and smelting time to cooling and taking out of the furnace is 5h, the vacuum degree of the vacuum consumable arc furnace is 1Pa, the smelting current is 5kA, and the smelting voltage is 28V.
[0186] 7) Forging: The alloy ingot is forged, which includes the following four steps, wherein the blanking process parameters are: heating temperature 1050°C, holding time 9h, forging ratio 5, to obtain a blank; the free forging process parameters are: the blank is subjected to 4 upsetting and drawing deformation at 1100°C, each upsetting and drawing ratio is 4, to obtain a forged blank; the fine forging process parameters are: the forging blank is subjected to 4 fine forging at 1080°C to obtain a rod; the die forging process parameters are: the rod is loaded into a matching die, and the rod is formed by one fire at 1080°C to obtain a titanium alloy blade.
[0187] 8) Heat treatment: The titanium alloy blades obtained above are subjected to solution aging heat treatment to finally obtain new titanium alloy blades with high thermal strength, elongation and creep strength; the solution temperature is 1040°C, the holding time is 0.5h, and the cooling method is water cooling; the aging temperature is 750°C, the holding time is 6h, and the cooling method is air cooling.
[0188] The titanium alloy blade made of the above-mentioned optimized alloy composition has a special Widmanstatten structure with equiaxed ɑ phase intercalated in the lamellar ɑ+β phase, wherein the lamellar ɑ+β phase is about 15μm long and has an aspect ratio of about 13; the equiaxed ɑ phase is distributed therein with a length of about 30μm and an aspect ratio of 5, and some equiaxed ɑ phases are connected together, with the content of equiaxed ɑ phase being about 26%; the high temperature tensile strength is 755MPa, the elongation is 24%, and the creep strength is 368MPa.
[0189] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A material qualification method applicable to civil aviation engine airworthiness certification, characterized in that: Includes the following: Determine the materials for civil aviation engines, select materials for civil aviation engines based on the expected service conditions, environment and strength life requirements of civil aviation engines, and consider the principles of applicability, processability, versatility, maintainability, maturity and economy; Determine whether material qualification test is needed based on the application of the selected material and its historical data, and obtain the selected material's conformity statement report or material qualification test data; Generate a catalog of qualified materials for civil aviation engines based on the selected material compliance report and material qualification test results; The process of judging whether a material qualification test is required based on the application of the selected material and its historical data includes the following: If the selected materials are mature materials that have been successfully applied in civil aviation products, and the historical data batches reach 10 batches of 10 batches, the number of sub-samples of each technical indicator of the material specification reaches 30, and the data source is traceable, the test process records are complete, and the formation process of the historical data meets the quality system requirements and technical requirements, then all conformity statements are prepared in the form of conformity statements, and combined with the subsequent whole machine test results to show that the requirements of the clause are met, then the material qualification test verification will no longer be carried out separately; If the selected material is a mature material that has been successfully applied in civil aviation products, but the historical data only partially meets the requirements of 10 batches of 10 batches of 10 furnaces, and the number of sub-samples of the technical indicators of the material specification reaches 30, and the data source is traceable, the test process records are complete, and the formation process of the historical data meets the quality system requirements and technical requirements, then this part shall be prepared in the form of a conformity statement report, and combined with the subsequent whole machine test results to show that it meets the requirements of the clause, no separate test verification shall be carried out; The parts that do not meet the requirements need to be verified by material qualification tests; If the selected material has not been successfully applied in civil aviation products, or the historical data corresponding to the selected material does not meet the requirements, or the selected material is a new material, a separate material qualification test verification is required; The material qualification test process includes the following: According to the key characteristics of engine parts, the selected materials are divided into key parts materials, important parts materials and general parts materials; Suppliers are divided into key-level suppliers, important-level suppliers and general-level suppliers according to the importance of the parts and technical capabilities they undertake. Key-level suppliers can undertake the qualification assessment of key parts materials, important parts materials and general parts materials, important-level suppliers can undertake the qualification assessment of important parts materials and general parts materials, and general-level suppliers can undertake the qualification assessment of general parts materials. Prepare the first draft of material standards and material production process control documents; Suppliers carry out the first batch production and inspection of key parts, important parts and general parts according to the material production process control documents and the first draft of material standards, and obtain the first piece identification data; Carry out test verification on key parts, important parts and general parts together with the whole machine to obtain the whole machine test data; Each technical indicator specified in the first draft of the material standard is verified and revised based on the first piece inspection data and the whole machine test data.
2. The material qualification method for civil aviation engine airworthiness certification according to claim 1, characterized in that: For key and important materials, after the first piece inspection is completed, the supplier must continue to produce 3 batches of materials in 3 furnaces in accordance with the material production process control documents, and conduct multiple batches of performance evaluation tests in accordance with the same requirements as the first piece inspection. After the multiple batches of performance evaluation tests are completed and qualified, prepare multiple batches of performance evaluation reports.
3. The material qualification method for civil aviation engine airworthiness certification according to claim 2, characterized in that: After passing multiple batches of performance evaluation tests, quality stability evaluation is also required.
4. The material qualification method for civil aviation engine airworthiness certification according to claim 3, characterized in that: For key component materials, it is also necessary to conduct long-term performance verification of the materials in combination with the actual operating conditions of the engine.
5. The material qualification evaluation method for civil aviation engine airworthiness certification according to claim 4, characterized in that: For key component materials, it is also necessary to design structural simulation parts and conduct long-term performance verification of the structural simulation parts.
6. The material qualification method for civil aviation engine airworthiness certification according to claim 5, characterized in that: For key and important parts materials, component / environmental testing verification is also required.
7. The material qualification method for civil aviation engine airworthiness certification according to claim 5, characterized in that: The process of long-term performance verification of the structural simulation parts includes the following: Design structural simulation parts according to the structural characteristics of key parts and the temperature and load conditions during service. Through simulation and optimization iteration, make the structural characteristics and load conditions of the structural simulation parts consistent with the real parts. Carry out structural simulation part processing, and use multi-level gradient interruption heat treatment process to eliminate the residual stress in the structural simulation part processing process, so as to avoid the residual stress in the processing process affecting the test results of the simulation part; According to the engine test spectrum, a test spectrum of structural simulation parts is compiled, and the heat-treated structural simulation parts are subjected to long-term tests according to the test spectrum of structural simulation parts to obtain the evolution law between the organization and performance of the structural simulation parts under long-term use conditions; A corrosion test spectrum is added to the test spectrum of the structural simulation parts. The heat-treated structural simulation parts are subjected to long-term tests according to the composite test spectrum to obtain the influence of corrosion on the evolution of the organization and performance of the structural simulation parts under long-term use conditions.
8. The material qualification method for civil aviation engine airworthiness certification according to claim 7, characterized in that: The process of eliminating residual stress during the processing of the structural simulation part by using a multi-level gradient interruption heat treatment process includes the following contents: Place the structural simulation part in a high-pressure vacuum heat treatment furnace, and raise the temperature in the furnace to 1010-1080°C at a heating rate of 10-20°C / min; Keep the structural simulation parts at 1010-1080℃ for 1-2h; Cooling the structural simulation to room temperature; Put the structural simulation part into a high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keep it warm for 0.5-1h, and then cool the structural simulation part to room temperature; put the structural simulation part back into the high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keep it warm for 0.5-1h, and then cool the structural simulation part to room temperature; put the structural simulation part back into the high-pressure vacuum heat treatment furnace again, increase the temperature in the furnace to 580-620°C at a heating rate of 30-50°C / min, keep it warm for 0.5-1h, and then cool the structural simulation part to room temperature; Put the structural simulation part into a high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation part to room temperature; put the structural simulation part back into the high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation part to room temperature; put the structural simulation part back into the high-pressure vacuum heat treatment furnace again, increase the temperature in the furnace to 720-760°C at a heating rate of 30-50°C / min, keep it warm for 20-50 minutes, and then cool the structural simulation part to room temperature; Place the structural simulation part in a high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 620-650°C at a heating rate of 30-50°C / min, keep it warm for 2h-4h, and then cool the structural simulation part to room temperature; place the structural simulation part back in the high-pressure vacuum heat treatment furnace, increase the temperature in the furnace to 620-650°C at a heating rate of 30-50°C / min, keep it warm for 2h-4h, and then cool the structural simulation part to room temperature; place the structural simulation part back in the high-pressure vacuum heat treatment furnace again, increase the temperature in the furnace to 620-650°C at a heating rate of 30-50°C / min, keep it warm for 2h-4h, and then cool the structural simulation part to room temperature.
Citation Information
Patent Citations
Sealant material verification method suitable for airworthiness certification
CN109592065A
Airworthiness conformity judgment method for aviation turbine engine lubricating oil
CN115343450A
