High-throughput design method for gradient structure of thermal barrier coating resistant to ultrahigh-temperature thermal shock

Abstract

The invention belongs to the technical field of thermal stress of thermal barrier coatings, and particularly relates to a high-throughput design method for a gradient structure of an ultrahigh-temperature thermal shock resistant thermal barrier coating. Based on Abaqus finite element simulation software, high-throughput calculation is carried out on a coating gradient structure, a high-throughput calculation method used for analyzing thermal stress of different structures at high temperature and screening optimal coating structure design parameters of different coating materials, thicknesses and base materials is developed, the mapping relation between structure evolution and coating performance is obtained, and high-throughput calculation is carried out on the coating gradient structure. And subsequent screening of customized coating structure parameters is facilitated. The invention provides a full-process analysis algorithm from macroscopic parameter input to final result analysis, the coating structure optimization process is accelerated, and the optimization time is shortened; the high-throughput parameter generation speed and the model component efficiency are improved by several times, and output data can be directly used for a material genome plan database; the analysis of the change rule of part of parameters along with the coating structure has important significance for further prolonging the service life of the coating.

Description

Technical Field

[0001] The present invention belongs to the field of thermal stress technology of thermal barrier coatings (TBCs), and in particular relates to a high-throughput design method for gradient structures of ultra-high temperature thermal shock resistant TBCs. The method is used to design ultra-high temperature thermal shock resistant TBC structures and realize rapid evaluation and optimization of various TBC structural design schemes. Background Art

[0002] With the vigorous development of cutting-edge science and technology fields such as aerospace, energy and power, the performance requirements for materials in extreme ultra-high temperature environments are becoming increasingly stringent. Thermal barrier coatings (TBCs), as an advanced surface modification technology, play a vital role in key components such as gas turbines and aircraft engines. They can effectively isolate the substrate material from the erosion of high-temperature gases, significantly improve the substrate material's heat resistance limit and oxidation resistance, and thus significantly extend the overall service life of the components. However, the design and optimization process of traditional thermal barrier coatings is cumbersome, time-consuming, and costly. This situation seriously restricts the rapid iteration and innovation pace of modern industry.

[0003] In recent years, rapid advances in numerical simulation technology have opened up new avenues for the design and optimization of thermal barrier coatings. Leveraging high-performance computers and advanced simulation software, researchers can simulate the performance of thermal barrier coatings under various operating conditions with unprecedented accuracy and efficiency. However, traditional coating development methods rely on manual trial and error or simple iterative simulations, which present the following drawbacks:

[0004] (1) Low efficiency of parameter exploration: Parameters such as coating thickness and gradient distribution need to be manually adjusted, making it difficult to cover multidimensional parameter space;

[0005] (2) Model components are cumbersome: creating a single model is time-consuming, and batch simulation requires repeated operations, which can easily introduce errors;

[0006] (3) Waste of computing resources: The hardware operation efficiency in serial computing mode is less than 30%, and structural analysis relies on manual screening.

[0007] This limitation seriously hinders the innovation speed of thermal barrier coating design and the expansion of its application scope.

[0008] Patent publication number CN114580245A proposes a numerical simulation method for the deposition stress of functionally gradient thermal barrier coatings (FGBCs) on curved substrates. This method simulates the deposition stress evolution of FGBCs on curved substrates through layer-by-layer modeling and stress import, employing axisymmetric processing and meshing techniques. However, its shortcomings include: It only simulates the deposition stress of a single layer and fails to efficiently explore combinations of multiple gradient layer thicknesses; it lacks an integrated parameterized modeling script, requiring manual multi-layer model creation, resulting in low computational efficiency; and it fails to achieve high-throughput analysis and structural optimization under multiple operating conditions.

[0009] Patent publication number CN118627215A proposes a 3D modeling system and method for finite element simulation of thermal barrier coatings on complex turbine blades. Based on SOLIDWORKS and Abaqus, the system implements geometric modeling, cooling channel cutting, and finite element simulation through multiple modules. However, its shortcomings include: focusing on the 3D geometric modeling process and omitting high-throughput parameter generation and optimization algorithms; relying on manual operations for model adjustment and load setting, and failing to implement automated batch analysis; and failing to perform high-throughput screening for thermal stress distribution in gradient structures.

[0010] Patent publication number CN117077499A proposes a numerical modeling method for the pore structure of thermal barrier coatings for heavy-duty gas turbines. This method constructs a matrix model of the thermal barrier coating's pore structure using a random growth method, controlling porosity, location distribution, and growth probability. The method then combines Abaqus to analyze the effect of pore structure on thermal conductivity and Young's modulus. However, its shortcomings include: It only models the pore structure and does not address high-throughput design of gradient layer thickness distribution; it lacks parametric modeling and batch calculations, requiring manual setting of pore parameters and failing to cover multidimensional parameter spaces; and it fails to address the systematic optimization of thermal mismatch stress at the gradient coating interface.

[0011] In view of this, it is particularly important to develop an efficient, accurate and high-throughput thermal barrier coating structure design method. Summary of the Invention

[0012] To address the above issues, the present invention aims to provide a high-throughput design method for gradient structures of thermal barrier coatings (TBCs) resistant to ultra-high temperature thermal shock. By developing high-throughput numerical analysis using Abaqus software and its Python script, the present invention can efficiently screen and optimize TBC materials and structural parameters, thereby improving the thermal shock resistance and service life of the coating.

[0013] In order to achieve the above object, the present invention provides the following technical solutions:

[0014] A high-throughput design method for gradient structures of thermal barrier coatings resistant to ultra-high temperature thermal shock is proposed. High-throughput calculations of the gradient structures of the coatings are performed using ABAQUS finite element simulation software. The finite element modeling process is sequentially implemented using Python code. The specific steps are as follows:

[0015] S1. Generate coating structure by high-throughput enumeration or grid method according to the number of coating gradient layers and total coating thickness input by the script;

[0016] S2. Using the Python API in Abaqus, a 2D geometric model of the thermal barrier coating and its substrate was created based on high-throughput structural design geometry parameters. The geometric model parameters included the coating substrate thickness, the bond layer thickness, the thermally grown oxide layer thickness, and the thickness distribution of each gradient ceramic layer.

[0017] S3. Read the material properties of the substrate, bonding layer, thermally grown oxide layer, and ceramic layer, and assign the material properties of the substrate, bonding layer, thermally grown oxide layer, and ceramic layer to the substrate, bonding layer, thermally grown oxide layer, and ceramic layer in the coating geometric model established in step S2 to obtain a geometric numerical model;

[0018] S4. Assemble the coating geometric numerical model to the instance, establish an analysis step, set boundary conditions for the geometric numerical model, and apply a temperature load to the geometric numerical model;

[0019] S5. Mesh the substrate and coating separately;

[0020] S6. Submit the geometric numerical model for analysis to obtain the stress and strain distribution of each layer of the structural coating, extract and save it in the database;

[0021] S7. Repeat steps S2 to S6, use the Python API of Abaqus to perform modeling and calculation analysis on the generated high-throughput coating structural parameters, and extract and save the data into the database.

[0022] In the high-throughput design method for the gradient structure of the ultra-high temperature thermal shock resistant thermal barrier coating, in step S1, the high-throughput generation process of the coating structure is as follows:

[0023] (1) According to the total thickness requirement of the coating to be analyzed and the number of gradient coatings, the coating thickness of the portion of the coating that requires high-throughput optimization is divided into multiple parts, and the fraction is required to be ≥ the number of gradient coating layers;

[0024] (2) All possible coating structures are screened out using an ordered combination method to form structural parameter data for high-throughput design.

[0025] In the high-throughput design method for the gradient structure of the thermal barrier coating resistant to ultra-high temperature thermal shock, in step S2, a coating structure design is sequentially taken out from the high-throughput data generated in step S1 to construct a two-dimensional geometric model. The width of the two-dimensional geometric model is set according to actual conditions. The parameters of the two-dimensional geometric model include the coating substrate thickness, the bonding layer thickness, the thermally grown oxide layer thickness and the gradient ceramic layer thickness distribution.

[0026] In the high-throughput design method for a gradient structure of a thermal barrier coating resistant to ultra-high temperature thermal shock, in step S3, the material properties of the substrate, bonding layer, thermally grown oxide, and ceramic layer are read from a material database. The material properties include thermal conductivity, density, Young's modulus, Poisson's ratio, thermal expansion coefficient, and density.

[0027] In the high-throughput design method for a gradient structure of a thermal barrier coating resistant to ultra-high temperature thermal shock, in step S4, the process of setting the boundary conditions of the geometric numerical model is as follows:

[0028] (1) The analysis step created is steady state, temperature and displacement coupling; the field output is created, and the output variables are stress, strain, displacement, action and reaction force, concentrated force and bending moment, contact stress, node temperature, heat flux vector and reaction flux; the process output request variables are total energy and thermal;

[0029] (2) Create thermal radiation parameters for the coating edge, create the left boundary as a symmetrical boundary, and the lower left corner as a fixed point;

[0030] (3) According to the experimental conditions, the temperature load applied to the coating specimens subjected to furnace thermal shock is the entire two-dimensional numerical model, and the temperature load applied to the coating specimens subjected to frontal flame thermal shock is the heat flux density on the front side of the coating.

[0031] In the high-throughput design method for a gradient structure of a thermal barrier coating resistant to ultra-high temperature thermal shock, in step S5, the implementation details of establishing mesh division are as follows: the size of the mesh division seed is designed according to the thickness of the thinnest coating, the unit shape adopts quadrilateral units as the main division method, the unit cell types are CPE4T and CPE3T, and the analysis type is plane strain; wherein, CPE4T is a four-node thermally coupled plane strain quadrilateral unit, with bilinear displacement and temperature; CPE3T is a three-node plane strain thermally coupled triangular unit, with linear displacement and temperature.

[0032] In the high-throughput design method for gradient structures of thermal barrier coatings resistant to ultra-high temperature thermal shock, in step S6, the node information, unit information, stress information and interface information of each layer of the coating structure are obtained by reading the current structure data file odb and saving them in the database.

[0033] The high-throughput design method for the gradient structure of the ultra-high temperature thermal shock resistant thermal barrier coating, in order to achieve high-throughput model analysis, sequentially retrieves data from the high-throughput structure design parameter database, and sequentially executes steps S2 to S6 to obtain numerical calculation results of the high-throughput structure design.

[0034] The design concept of the present invention is:

[0035] First, the present invention provides a full-process analysis algorithm from macro-parameter input to final result analysis, which accelerates the coating structure optimization process and shortens the optimization time; second, the high-throughput parameter generation speed and model component efficiency are improved several times, and the output data can be directly used in the Materials Genome Initiative (MGI) database; finally, analyzing the change pattern of some parameters with the coating structure is of great significance for further improving the life of the coating.

[0036] Unlike traditional finite element modeling, this method deeply integrates Python scripting with the Abaqus API, enabling high-throughput automated modeling and parametric design of thermal barrier coating structures. This approach automatically generates coating structure parameters in batches and sequentially completes a series of operations, from geometric model creation, material property assignment, analysis step setting, meshing, and computational analysis. This eliminates the need for manual adjustment of model parameters, significantly improving design efficiency and automation.

[0037] This method aims to achieve rapid and comprehensive analysis and optimization of thermal barrier coating structures by integrating advanced numerical simulation techniques with efficient parameter optimization algorithms. This approach effectively addresses the cumbersome, time-consuming, and costly design and optimization processes of traditional thermal barrier coatings, as well as the shortcomings of traditional numerical simulation methods, such as inefficient parameter exploration, cumbersome model construction, and wasted computational resources. By enabling high-throughput parameter generation and improving the efficiency of model components, it can rapidly cover multidimensional parameter spaces and identify optimal design solutions, significantly reducing design costs and shortening R&D cycles. This provides strong technical support for continued advancements in aerospace, energy, and power generation.

[0038] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0039] This paper provides a high-throughput numerical analysis method for coating structure screening. By integrating Python scripts with the Abaqus API, it enables high-throughput automated modeling and parametric design of thermal barrier coating structures, eliminating the need for manual adjustment of model parameters. Compared to traditional finite element modeling methods, this method improves design efficiency by several times. It is particularly suitable for the rapid screening of complex parameter combinations such as gradient coatings and multilayer composite structures, significantly improving design efficiency and automation.

[0040] 2. The present invention adopts a gradient ceramic layer thickness distribution design and screens the optimal gradient parameter combination through high-throughput calculation, which effectively alleviates the thermal mismatch stress caused by the difference in thermal expansion coefficient at the coating / substrate interface. Combined with the dynamic call of the material property database, it can optimize the key parameters such as thermal conductivity and thermal expansion coefficient of each layer of material in a targeted manner, significantly reducing the risk of interface crack initiation and propagation during thermal cycling.

[0041] 3. This invention supports adaptive design for multiple operating conditions. By flexibly setting temperature load boundary conditions (such as furnace thermal shock and flame front thermal shock), it can simulate the complex thermal shock environment experienced during actual aircraft engine operation. High-throughput calculation results provide data support for coating structural design under different service scenarios, improving the coating's comprehensive resistance to failure mechanisms such as CMAS corrosion and high-temperature oxidation.

[0042] 4. This invention, based on database-driven parametric modeling and batch calculations, can quickly eliminate low-performance structural solutions and prioritize high-potential design candidates. Compared with traditional trial-and-error methods, it significantly reduces the number of physical experiments, shortens the R&D cycle, and reduces material preparation and testing costs.

[0043] 5. This invention achieves a synergistic match between material performance and structural design through the combined optimization of material properties and geometric parameters. For example, to address the sintering densification problem of zirconia-based ceramics, gradient porosity design can compensate for the increasing trend in thermal conductivity and extend the service life of the coating.

[0044] 6. This invention supports the structural design of thermal barrier coatings in ultra-high temperature environments (>1600°C). Combined with the performance prediction of advanced ceramic materials, it can promote the development of next-generation aircraft engines, gas turbines and other equipment towards higher thrust-to-weight ratios.

[0045] 7. The data-driven intelligent decision support of the present invention can construct a thermal barrier coating performance-parameter relationship model through the accumulation of a database of high-throughput calculation results, provide machine learning training data for subsequent design, and gradually realize intelligent material screening and structural optimization. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 Start running the flow chart for high-throughput algorithms.

[0047] Figure 2 An example diagram of modeling a test piece coating.

[0048] Figure 3 An example diagram of meshing for a numerical model.

[0049] Figure 4 The following are typical stress analysis diagrams for a structure, where (a) is the tangential stress distribution cloud diagram and (b) is the normal stress distribution cloud diagram.

[0050] Figure 5-Figure 7 For high-throughput screening analysis, the maximum tensile stress and maximum shear strain in each layer are plotted as a function of the structure.

[0051] Figure 5 The following are statistical cloud diagrams showing the maximum and minimum tangential stresses and normal stresses in the YSZ-01 layer and the thickness variations of the YSZ-01 and YSZ-02 layers (the fourth figure shows the statistical cloud diagrams of the maximum and minimum stresses in the YSZ-01 layer under different structures). Upper left: YSZ-01S22 Tens Stress Distribution is a cloud diagram showing the maximum tensile stress variation of the YSZ-01 layer in different structures. Upper right: YSZ-01S22 Comp Stress Distribution is a cloud diagram showing the maximum compressive stress variation of the YSZ-01 layer in different structures. Lower left: YSZ-01S12 Shearing Stress Distribution is a statistical cloud diagram showing the maximum negative shear stress of the YSZ-01 layer in different structures. Lower right: YSZ-01S12 Shearing Stress Distribution is a statistical cloud diagram showing the maximum positive shear stress distribution of the YSZ-01 layer in different structures. S22 refers to Figure 4 The positive sign indicates tensile stress and the negative sign indicates compressive stress in the longitudinal direction. S12 refers to the tangential stress. According to the shear stress equivalence theorem, the positive and negative signs indicate the direction and the numerical value indicates the size.

[0052] Figure 6 The following are statistical cloud diagrams showing the maximum and minimum tangential stresses and normal stresses in the YSZ-02 layer and the thickness variations of the YSZ-01 and YSZ-02 layers (the fourth figure shows the statistical cloud diagrams of the maximum and minimum stresses in the YSZ-02 layer under different structures). Upper left: YSZ-02S22 Tens Stress Distribution is a cloud diagram showing the maximum tensile stress variation of the YSZ-02 layer in different structures. Upper right: YSZ-02S22 Comp Stress Distribution is a cloud diagram showing the maximum compressive stress variation of the YSZ-02 layer in different structures. Lower left: YSZ-02S12 Shearing Stress Distribution is a statistical cloud diagram showing the maximum negative shear stress of the YSZ-02 layer in different structures. Lower right: YSZ-02S12 Shearing Stress Distribution is a statistical cloud diagram showing the maximum positive shear stress distribution of the YSZ-02 layer in different structures. S22 refers to Figure 4 The positive sign indicates tensile stress and the negative sign indicates compressive stress in the longitudinal direction. S12 refers to the tangential stress. According to the shear stress equivalence theorem, the positive and negative signs indicate the direction and the numerical value indicates the size.

[0053] Figure 7The following are statistical cloud diagrams showing the maximum and minimum tangential stresses and normal stresses in the YSZ-03 layer and the thickness variations of the YSZ-01 and YSZ-02 layers (the fourth figure shows the statistical cloud diagrams of the maximum and minimum stresses in the YSZ-03 layer under different structures). Upper left: YSZ-03S22 Tens Stress Distribution is a cloud diagram showing the maximum tensile stress variation of the YSZ-03 layer in different structures. Upper right: YSZ-03S22 Comp Stress Distribution is a cloud diagram showing the maximum compressive stress variation of the YSZ-03 layer in different structures. Lower left: YSZ-03S12 Shearing Stress Distribution is a statistical cloud diagram showing the maximum negative shear stress of the YSZ-03 layer in different structures. Lower right: YSZ-03S12 Shearing Stress Distribution is a statistical cloud diagram showing the maximum positive shear stress distribution of the YSZ-03 layer in different structures. S22 refers to Figure 4 The positive sign indicates tensile stress and the negative sign indicates compressive stress in the longitudinal direction. S12 refers to the tangential stress. According to the shear stress equivalence theorem, the positive and negative signs indicate the direction and the numerical value indicates the size.

[0054] Figure 8 This is a structural block diagram of the high-throughput coating structure numerical analysis system. DETAILED DESCRIPTION

[0055] The algorithm flow of the present invention is explained below in conjunction with the accompanying drawings, and the technical solutions in the implementation examples of the present invention are described in detail and completely. Obviously, the implementation examples described are only part of the implementation examples of the present invention, rather than all the implementation examples.

[0056] During the specific implementation process, high-throughput calculations of the coating gradient structure were performed based on the Abaqus finite element simulation software. The present invention developed a high-throughput calculation method for analyzing the thermal stress of different structures at high temperatures and screening the optimal coating structure design parameters for different coating materials, thicknesses and substrates. The mapping relationship between structural evolution and coating performance was obtained, which facilitates the subsequent screening of customized coating structure parameters.

[0057] like Figure 1 As shown, the present invention provides a high-throughput design method for a gradient structure of a thermal barrier coating resistant to ultra-high temperature thermal shock, comprising the following steps:

[0058] 1) Based on the input parameters (total thickness and width of the specimen, substrate thickness, bonding layer thickness, total number of gradient ceramic layers, and minimum ceramic layer thickness), batch generate all coating structure parameter designs under the current conditions. Each structure parameter design includes the total thickness and width of the coating specimen, substrate thickness, bonding layer thickness, and the thickness of the i-th ceramic layer, where i = 1, 2, 3, ..., n (i ≤ n), where n is the total number of gradient ceramic layers.

[0059] 2) Take a piece of coating structure data from step 1) and use the Abaqus Python API interface to build a two-dimensional model, which includes a substrate, a bonding layer, and ceramic gradient layers.

[0060] 3) Obtaining the material properties of the substrate, bonding layer, and each ceramic gradient layer, and assigning the material properties of the substrate, bonding layer, and each ceramic gradient layer to the substrate, bonding layer, and each ceramic gradient layer of the two-dimensional model in step 2) to obtain a two-dimensional numerical model.

[0061] 4) Establish analysis steps for the numerical model, set boundary conditions and apply loads.

[0062] 5) Divide the coating structures into grids.

[0063] 6) Submit numerical models for calculation.

[0064] 7) Analyze the odb file in the result data and extract the stress field, strain field and temperature field data of each layer of the coating specimen for subsequent analysis.

[0065] 8) Execute steps 2) to 7) repeatedly to complete the numerical analysis and result extraction of the high-throughput designed coating structure.

[0066] In step 2), the specific modeling method is as follows: first, a rectangle is drawn according to the total thickness and total width of the test piece, the horizontal length of the rectangle is half of the total width of the test piece, and the vertical height of the rectangle is the total height of the test piece. Then, the horizontal line is used to divide the area of ​​each layer in turn according to the structural parameters of each coating to form a gradient coating geometric structure, such as Figure 2 shown.

[0067] In step 4), the specific steps are:

[0068] 401, the analysis step created is steady state, temperature and displacement coupling, field quantity output and history output;

[0069] 402, set the boundary conditions. A simply supported boundary condition was applied to the edge of the substrate to avoid infinite displacement of the model. An axisymmetric constraint was applied to the left boundary of the specimen. Only the half-section of the cylindrical specimen was analyzed to reduce the numerical solution time.

[0070] 403, a temperature load is applied to the two-dimensional numerical model to simulate the thermal stress changes of the coating during the temperature increase process.

[0071] In step 5), in order to facilitate programming operations, a unified seeding method is used. The seeding spacing is determined according to the thickness of the thinnest layer to ensure that the thinnest layer has at least three layers of units. The unit shape uses quadrilateral units as the main division method. The unit cell types are CPE4T and CPE3T. CPE3T is a triangular unit type required to prevent uneven seeding. The meshing effect is shown in Figure 3 .

[0072] The grid method is a method used to determine the distribution of grid nodes when meshing a model in fields such as finite element analysis. Its purpose is to divide the model into appropriate units to improve calculation accuracy and efficiency.

[0073] In step 6), in order to speed up the calculation process and shorten the calculation time, the parallel computing settings need to be modified according to the computer hardware settings.

[0074] For example, using 4-thread or Hybrid multi-thread mode to speed up the calculation process. When simulating the pore structure and gradient structure of thermal barrier coatings, the calculation amount is huge. Using 4-thread or Hybrid multi-thread mode can significantly speed up the calculation.

[0075] In step 7), the odbAccess interface in Abaqus is used to read the data and export the node information, unit information, and stress information in the numerical result model, and then save it in an easy-to-read encoding format.

[0076] like Figure 8 As shown, the present invention provides a complete set of high-throughput coating structure numerical analysis system, specifically including:

[0077] High-throughput coating structure generation module, used to batch generate coating structure design parameters;

[0078] Model building module, used to intelligently build coating geometric models based on various coating structural parameters;

[0079] Material property conversion module, which is used to convert the input material properties into a format that can be used by the solver and automatically assign the material properties to each coating structure;

[0080] The core processing module is used to automatically establish analysis steps, apply loads, set boundary conditions, mesh, and submit jobs;

[0081] The post-processing module is used to read the odb result file and format the required information and save it into a data file in an easy-to-read encoding format.

[0082] The present invention also provides a computer-readable storage medium on which a computer program for implementing the above-mentioned module is stored. Executing the computer program can realize the simulation of the above-mentioned high-throughput coating numerical analysis method.

[0083] like Figure 8 As shown, the connection relationship between the various modules of the high-throughput coating structure numerical analysis system and the computer-readable storage medium is as follows: in the storage medium, a boot program is saved, through which the main thread of the method can be called, and then the calculation data is stored in the readable storage medium through the high-throughput coating structure generation module, the model establishment module, the material property conversion module, the core processing module, the post-processing module and the storage database module.

[0084] The present invention is further described in detail below through examples and drawings.

[0085] Example 1

[0086] In this example, a high-throughput design for a thermal barrier coating (TBC) structure consisting of a GH3536 substrate, a NiCrAlY bond layer, and three layers of ceramic coating composed of various rare earth oxide co-stabilized zirconias (YSZ-01, YSZ-02, and YSZ-03). The total coating thickness is 3.8 mm, consisting of a 3 mm substrate, a 150 μm bond layer, and a total of 650 μm ceramic layers.

[0087] A high-throughput design method for a gradient structure of a thermal barrier coating resistant to ultra-high temperature thermal shock comprises the following steps:

[0088] 1) High-throughput calculation of coating structure: the total thickness of the three ceramic layers (650 μm) is divided into five parts, each with a thickness of 130 μm. Then all coating combinations for high-throughput design are (0.13, 0.13, 0.39), (0.13, 0.26, 0.26), (0.13, 0.39, 0.13), (0.26, 0.13, 0.26), (0.26, 0.26, 0.13), (0.39, 0.13, 0.13), (unit: mm), where the format is (ceramic layer I, ceramic layer II, and ceramic layer III), for a total of 6 coating structures.

[0089] 2) One of the six coating structures was selected and the Abaqus Python API was used to build a geometric model, including a 3 mm thick GH3536 substrate, a 150 μm bonding layer, and a thickness distribution of the ceramic coating structure.

[0090] 3) Material property design for each coating structure. This invention considers the temperature-dependent changes in the thermophysical properties of each coating structure. This includes thermal conductivity, density, specific heat capacity, elastic modulus, Poisson's ratio, and thermal expansion coefficient. First, the table of input material thermophysical parameters is converted into a document format that Abaqus can recognize and loaded into Abaqus. The temperature-dependent thermophysical parameters of the GH3536 substrate, NiCrAlY, YSZ-01, YSZ-02, and YSZ-03 materials are shown in Table 1:

[0091] Table 1 Thermophysical parameters of materials required for calculation

[0092]

[0093] 4) Establish analysis steps for the model in 3), set boundary conditions and apply loads, including:

[0094] 401, the analysis step created is steady state, temperature and displacement coupling; create field output, and the output variables are stress, strain, displacement and node temperature;

[0095] 402, setting boundary conditions. A simply supported boundary condition was applied to the ground edge of the GH3536 substrate to avoid infinite displacement of the model. An axisymmetric constraint was applied to the left boundary of the specimen. Only the half-section of the cylindrical specimen was analyzed to reduce the numerical solution time.

[0096] 403, a temperature load of 1400℃ is applied to the model to simulate the changes in stress, strain, displacement and node temperature in the coating system when the coating is heated to 1400℃.

[0097] 5) Mesh the geometric structure of the coating design. The cell types are CPE4T and CPE3T. The main cell shape is quadrilateral, and the backup cell shape is triangle. Since the minimum thickness of the ceramic layer is 0.13 mm, the mesh size is 0.04 mm to ensure that each ceramic layer has at least 3 layers of cells.

[0098] 6) Use 4-thread or Hybrid multi-thread mode to accelerate the calculation process, establish the job analysis work, and complete this round of coating structure numerical analysis process.

[0099] 7) Use the Abaqus Python API to read the input inp file and the result odb file to extract the node, element, stress, and interface information for each coating structure. Obtain the next coating structure design and repeat steps 2) to 6) to complete the numerical analysis of the six coating structures.

[0100] like Figure 4The following figure shows the calculation results for a ceramic layer structure of (0.13, 0.13, 0.39) uniformly heated in a furnace. The figure shows that the maximum and minimum tangential stresses occur at the coating edges, while the maximum and minimum normal stresses are mainly distributed at the coating edges and between layers.

[0101] like Figure 5-Figure 7 As shown in the figure, through the above high-throughput coating structure design method, the influence of the thickness change of each layer on the coating cloud distribution can be obtained. As an example, the maximum and minimum tangential stress and normal stress of the YSZ-01, YSZ-02 and YSZ-03 layers are extracted. Figure 5 The maximum and minimum tangential stress and normal stress statistics of the YSZ-01 layer are: Figure 6 The maximum and minimum tangential stress and normal stress statistics of the YSZ-02 layer are: Figure 7 The maximum and minimum tangential stresses and normal stress statistics of the YSZ-03 layer are shown in the figure. According to the rules of this figure, the comprehensive performance of the coating can be optimized according to the failure position of the coating.

[0102] Implementation results show that the present invention avoids missing some structural designs through high-throughput coating structure design, automates modeling, accelerates the efficiency of coating structure optimization, reduces errors introduced by manual modeling, and is deeply integrated with Abaqus' Python API to achieve unsupervised completion of calculation tasks. This high-throughput calculation can describe the influence of coating structure on the comprehensive performance of the coating, provide theoretical support for slowing the initiation time of coating cracks and improving the thermal cycle life of the coating, and is of great significance to coating structure optimization.

Claims

1. A high-throughput design method for a gradient structure of a thermal barrier coating resistant to ultra-high temperature thermal shock, characterized in that: High-throughput calculation of the coating gradient structure was performed based on ABAQUS finite element simulation software, and the finite element modeling process was implemented in sequence using Python code. The specific steps are as follows: S1. Generate coating structure by high-throughput enumeration or grid method according to the number of coating gradient layers and total coating thickness input by the script; S2. Using the Python API in Abaqus, a 2D geometric model of the thermal barrier coating and its substrate was created based on high-throughput structural design geometry parameters. The geometric model parameters included the coating substrate thickness, the bond layer thickness, the thermally grown oxide layer thickness, and the thickness distribution of each gradient ceramic layer. S3. Read the material properties of the substrate, bonding layer, thermally grown oxide layer, and ceramic layer, and assign the material properties of the substrate, bonding layer, thermally grown oxide layer, and ceramic layer to the substrate, bonding layer, thermally grown oxide layer, and ceramic layer in the coating geometric model established in step S2 to obtain a geometric numerical model; S4. Assemble the coating geometric numerical model to the instance, establish an analysis step, set boundary conditions for the geometric numerical model, and apply a temperature load to the geometric numerical model; S5. Mesh the substrate and coating separately; S6. Submit the geometric numerical model for analysis to obtain the stress and strain distribution of each layer of the structural coating, extract and save it in the database; S7. Repeat steps S2 to S6, use the Python API of Abaqus to perform modeling and calculation analysis on the generated high-throughput coating structural parameters, and extract and save the data into the database.

2. The high-throughput design method for gradient structure of ultra-high temperature thermal shock resistant thermal barrier coating according to claim 1, characterized in that: In step S1, the high-throughput generation process of the coating structure is as follows: (1) According to the total thickness requirement of the coating to be analyzed and the number of gradient coatings, the coating thickness of the portion of the coating that requires high-throughput optimization is divided into multiple parts, and the fraction is required to be ≥ the number of gradient coating layers; (2) All possible coating structures are screened out using an ordered combination method to form structural parameter data for high-throughput design.

3. The high-throughput design method for gradient structure of ultra-high temperature thermal shock resistant thermal barrier coating according to claim 1, characterized in that: In step S2, a coating structure design is sequentially taken out from the high-throughput data generated in step S1 to construct a two-dimensional geometric model. The width of the two-dimensional geometric model is set according to actual conditions. The parameters of the two-dimensional geometric model include the coating substrate thickness, the bonding layer thickness, the thermally grown oxide layer thickness and the gradient ceramic layer thickness distribution.

4. The high-throughput design method for gradient structure of ultra-high temperature thermal shock resistant thermal barrier coating according to claim 1, characterized in that: In step S3, the material properties of the substrate, the bonding layer, the thermally grown oxide, and the ceramic layer are read from the material database. The material properties include thermal conductivity, density, Young's modulus, Poisson's ratio, thermal expansion coefficient, and density.

5. The high-throughput design method for gradient structure of ultra-high temperature thermal shock resistant thermal barrier coating according to claim 1, characterized in that: In step S4, the process of setting the boundary conditions of the geometric numerical model is as follows: (1) The analysis step created is steady state, temperature and displacement coupling; the field output is created, and the output variables are stress, strain, displacement, action and reaction force, concentrated force and bending moment, contact stress, node temperature, heat flux vector and reaction flux; the process output request variables are total energy and thermal; (2) Create thermal radiation parameters for the coating edge, create the left boundary as a symmetrical boundary, and the lower left corner as a fixed point; (3) According to the experimental conditions, the temperature load applied to the coating specimens subjected to furnace thermal shock is the entire two-dimensional numerical model, and the temperature load applied to the coating specimens subjected to frontal flame thermal shock is the heat flux density on the front side of the coating.

6. The high-throughput design method for gradient structure of ultra-high temperature thermal shock resistant thermal barrier coating according to claim 1, characterized in that: In step S5, the implementation details of establishing the mesh division are as follows: the size of the mesh division seed is designed according to the thickness of the thinnest coating, the unit shape adopts quadrilateral units as the main division method, the unit cell types are CPE4T and CPE3T, and the analysis type is plane strain; among them, CPE4T is a four-node thermally coupled plane strain quadrilateral unit, with bilinear displacement and temperature; CPE3T is a three-node plane strain thermally coupled triangular unit, with linear displacement and temperature.

7. The high-throughput design method for gradient structure of ultra-high temperature thermal shock resistant thermal barrier coating according to claim 1, characterized in that: In step S6, the node information, unit information, stress information and interface information of each layer of the coating structure are obtained by reading the current structure data file odb and saving them in the database.

8. The high-throughput design method for gradient structure of ultra-high temperature thermal shock resistant thermal barrier coating according to claim 1, characterized in that: In order to realize high-throughput model analysis, data are sequentially retrieved from the high-throughput structure design parameter database, and steps S2 to S6 are sequentially executed to obtain numerical calculation results of the high-throughput structure design.

Images