Single-crystal double-row high-temperature alloy blade large-module casting simulation method

Abstract

The invention relates to the field of directional solidification simulation of single-crystal high-temperature alloy blades, in particular to a large-module casting simulation method for single-crystal double-row high-temperature alloy blades. The method comprises the following steps: constructing a single-crystal double-row large-module model, dividing a casting grid, and setting process parameters of analogue simulation: symmetrically setting castings, setting alloy physical property parameters, setting a directional solidification heat transfer coefficient, setting a casting drawing rate, and setting seed crystal method crystal orientation and nucleation parameters. The solidification process and the mixed crystal defect position of the large-module casting in the directional solidification process are obtained through numerical simulation, the change process of an isotherm and a pasty area in the directional solidification process is obtained, and a macrosegregation distribution diagram of any position of the blade is visually observed. By analyzing the analogue simulation result, various technological parameters can be efficiently and flexibly replaced for analogue simulation, the technological parameters suitable for single crystal large module casting are rapidly screened out, and formation of certain casting defects is avoided in advance.

Description

Technical Field

[0001] The present invention relates to the field of directional solidification simulation of single crystal high-temperature alloy blades, and in particular to a large-module casting simulation method for single crystal double-row high-temperature alloy blades. Background Art

[0002] With the continuous development of aero-engine technology, my country has gradually established a complete and mature specialized technical system. Among these, high-temperature alloy turbine blades are crucial components in aero-engines and a key factor influencing the performance of Chinese-made aero-engines. These blades must withstand extreme rotational motion and the impact of high-temperature combustion gases, placing new demands on the high-temperature service stability and complex stress-bearing capabilities of the blade material.

[0003] Directional solidification technology produces single-crystal superalloys without grain boundaries by controlling the temperature gradient during solidification. This eliminates alloy failure caused by crack initiation and propagation at grain boundaries, improving the mechanical properties and strength of crystalline materials. High-speed solidification (HRS), a commonly used directional solidification process, effectively controls the temperature gradient and solidification rate during solidification, resulting in an ideal alloy dendrite structure. This process is currently the most widely used directional solidification process for superalloys.

[0004] Large-module casting technology for single-crystal, double-row, high-temperature alloy blades is an iterative process based on a high-speed solidification process. It can produce 24 single-crystal, double-row, 12-blade blades in a single batch, significantly improving turbine blade production efficiency. However, due to the complex internal radiation heat transfer inherent in large-module casting, various process parameters can have varying degrees of impact on the as-cast structure of the casting, necessitating the determination of appropriate process parameters based on specific circumstances.

[0005] Directional solidification process simulation technology can effectively simulate the actual directional solidification casting process and calculate heat transfer and latent heat release during solidification. By setting appropriate process parameters, it can simulate heat transfer and solidification defects in the actual directional solidification process. Therefore, an effective large-scale casting simulation method is of great significance for pre-determining appropriate process parameters and realizing the mass production application of large-scale casting technology. Summary of the Invention

[0006] In view of this, the purpose of the present invention is to provide a large-module casting simulation method for single-crystal double-row high-temperature alloy blades, which uses simulation methods to determine appropriate process parameters in advance and effectively predict the temperature field, casting defects, macro-segregation, etc. during the production process.

[0007] In order to achieve the above object, the technical solution of the present invention is:

[0008] A large-module casting simulation method for single-crystal double-row high-temperature alloy blades comprises: constructing a single-crystal double-row large-module model, meshing the casting, and setting process parameters for the simulation: setting the casting symmetry axis, setting the alloy physical property parameters, setting the directional solidification heat transfer coefficient, setting the casting pulling rate, and setting the crystal orientation and nucleation parameters of the seed crystal method; through numerical simulation, the solidification process and the position of the impurity defects of the large-module casting during the directional solidification process are obtained, the change process of the isothermal line and the mushy zone during the directional solidification process are obtained, and the macro-segregation distribution diagram at any position of the blade is intuitively observed, providing a theoretical basis for the microstructure analysis during the experiment.

[0009] The single crystal double row high temperature alloy blade large module casting simulation method, the single crystal double row large module model is constructed as follows: the model includes a pouring cup, a blade runner, an outer blade, an inner blade, a seed crystal segment, an intermediate support, and a water-cooled copper plate; each of the above components is separately established in the modeling software and then assembled, so as to ensure that it is identified as a separate individual in the simulation software; the entire large module model includes 24 single crystal blades and 24 seed crystal segments, the blades are divided into two layers, the inner and outer layers, the closest distance between the two layers is 20 mm, each layer is provided with 12 single crystal blades and 12 seed crystal segments, each blade is 90 mm high and 7 mm thick; the seed crystal segment connects the blade and the water-cooled copper plate, the bottom of the seed crystal segment is in direct contact with the water-cooled copper plate, the seed crystal segment is 20 mm high, and the crystal orientation is [0 01], the diameter of the water-cooled copper plate is 300 mm; the pouring cup and the water-cooled copper plate are connected by an intermediate support, and the diameter of the pouring cup is 150 mm; the bottom of the pouring cup is connected to the top of the blade through a blade runner, and the diameter of the blade runner is 5 mm.

[0010] The described large module casting simulation method for single crystal double-row high-temperature alloy blades, the casting mesh division refers to: first, a two-dimensional mesh is established for the large module casting, water-cooled copper plate, casting mold shell, and furnace part; after checking that the mesh has no errors, a three-dimensional mesh is established for the above-mentioned part except the furnace and checked for errors.

[0011] The described large-module casting simulation method for single-crystal double-row high-temperature alloy blades and the two-dimensional grid establishment method are as follows: first, the overlapping areas of the large-module casting are eliminated to ensure that there is no overlap between the parts, and then the parts are assembled to connect the parts to each other; after completing the above steps, the seed crystal segment and the blade part are divided into a two-dimensional grid with a unit length of 1; the gate cup and the blade runner part are divided into a two-dimensional grid with a unit length of 4; the furnace part is divided into a two-dimensional grid with a unit length of 20; after completing the above steps, each is grid-checked separately.

[0012] The single-crystal double-row high-temperature alloy blade large-module casting simulation method, when setting the process parameters of the simulation, needs to set the alloy physical properties parameters, directional solidification heat transfer coefficient, casting pulling rate, and nucleation parameters in sequence; after ensuring that the values ​​are consistent with the experimental or production process to the greatest extent, the simulation is carried out to obtain the most practical simulation results.

[0013] The single crystal double-row high-temperature alloy blade large module casting simulation method, the alloy physical property parameters are: the alloy grade selected in the simulation process is DD5, and the density, Young's modulus, and Poisson's ratio of the alloy used for simulation are calculated in advance using thermodynamic simulation software and brought into the simulation software.

[0014] The single crystal double-row high-temperature alloy blade large module casting simulation method described above, the directional solidification heat transfer coefficient refers to: the heat transfer coefficient includes the casting and the mold shell, the casting and the water-cooled copper plate, the mold shell and the water-cooled copper plate, the mold shell outer surface and the furnace, the water-cooled copper plate and the furnace, and the corresponding coefficient is set according to the literature.

[0015] The single-crystal double-row high-temperature alloy blade large-module casting simulation method, the casting pulling rate refers to: during the simulation process, due to the large amount of calculation of the casting parts, the casting is usually fixed and the furnace moves in the opposite direction during the pulling process to simulate the directional solidification process; wherein, the furnace is set to move at a speed of 5mm / min along the -Z axis direction, thereby simulating the actual directional solidification process.

[0016] In the large-module casting simulation method for single-crystal double-row high-temperature alloy blades, the nucleation parameters are as follows: the surface of the seed crystal segment contacting the water-cooled copper disk is defined as a single crystal nucleation, and the orientation is set to [0 0 1]. To achieve a higher degree of consistency with reality, the four outer surfaces of the seed crystal segment are also defined as single crystal nucleations; the blade portion is defined as a body nucleation, and the average nucleation undercooling is 10-25k. Then, the position and amount of stray crystal formation are observed under different process parameters under the same nucleation parameters.

[0017] The single-crystal double-row high-temperature alloy blade large-module casting simulation method can intuitively observe the isotherm distribution map, mushy zone distribution map, macro-segregation distribution map, and crystal orientation distribution map of the blade after completing the simulation, and then analyze the causes of casting defects and increase the yield of large-module blades.

[0018] The design idea of ​​the present invention is:

[0019] The present invention constructs a single crystal double-row large module model through software, imports it into casting simulation software for assembly and elimination of overlapping parts, performs grid division and detects error, sets the casting symmetrically, sets physical properties, heat transfer parameters, pulling rate, seed crystal orientation, etc., runs simulation, and obtains the temperature distribution of the mushy zone during the solidification process, casting defect distribution, macro-segregation distribution, etc.

[0020] First, use software to model the water-cooled copper plate, seed crystal segment, single crystal blade, runner, pouring cup, intermediate support, and furnace respectively, and assemble them. During the modeling process, ensure that there is no interference between the inner and outer blades. A distance greater than the thickness of the mold shell must be reserved in advance between each layer of blades. Set the thickness of the furnace to zero. Export the file after completing the modeling.

[0021] Next, import the above file into simulation software. The software eliminates the overlap between the runner and the pouring cup and sets the mesh. Considering computer computing power and to minimize computational effort, the mesh size for the seed crystal and blade sections is set to 1 unit length, the mesh size for the rest of the casting is set to 4 units length, and the mesh size for the furnace section is set to 20 units length. After meshing, ensure that there are no errors before proceeding to the next simulation step.

[0022] Select the furnace mesh in the statistics tree and convert it to a button box in the software. Define this as the furnace section and set the correct radiation direction. After confirming that the mesh is correct, prepare the mold shell for the casting. Set the mold shell thickness to 5mm and select the gate and symmetry planes without a mold shell. After generating the mold shell, check the mesh to ensure that there are no errors. Generate the volume mesh and then proceed to the next simulation setting.

[0023] After completing the mesh part, continue to set the parameters of the casting process: (1) Set the gravity direction to the +Z axis direction, select the water-cooled copper disk part, set its material to the mold shell-water-cooled copper disk material in the software database, and define its temperature to 20℃; (2) Select the casting part, set its material to the experimental alloy with pre-set parameters, and define its temperature to 1400~1600℃; (3) Select the mold shell part, set its material to the mold shell-refractory alumina material in the software database, and define its temperature to 1400~1600℃; (4) Select the buckle box part, without setting its material, set its upper heating zone temperature to 1500~1600℃, the lower heating zone temperature to 1450~1600℃, and set the cooling zone temperature to 20℃; (5) Select the seed crystal section part, and set the seed crystal orientation to [0 0 1]; (6) Select the blade casting part, input the pre-set body shape core parameters through the body shape core option in the software, select the blade part for grain structure analysis, and after checking that the data is correct, perform simulation.

[0024] Based on the above technical solution, it can be seen that the large-module casting simulation method for single-crystal double-row high-temperature alloy blades provided by the present invention has at least one of the following advantages and beneficial effects:

[0025] 1. The present invention's large-module casting simulation method for single-crystal, double-row, high-temperature alloy blades effectively simulates the production process of single-crystal, large-module blades using a directional solidification process. By adjusting process parameters such as the withdrawal rate, casting temperature, and mold shell thickness in the simulation software and observing their effects on solidification defect formation and temperature field changes, analysis of the simulation results allows optimization of process parameters before experiments or production, preventing the formation of certain casting defects.

[0026] 2. This invention can predict the locations of certain casting defects during experiments or production. Through the large-scale casting simulation method of the present invention, it is possible to visually observe the locations where stray crystals are likely to form, such as at the sharp corners of the blade edge, the air intake edge, and the junction between the blade body and the shroud. Compared to complex and costly on-site experiments, this invention can economically and efficiently analyze the relationship between various process parameters and specific casting defects, allowing for rapid adjustments and optimization of process parameters.

[0027] 3. The simulation results of this invention allow for intuitive observation of the macrosegregation distribution of the as-cast structure, providing a theoretical basis for microstructural analysis during the experimental process. This method is applicable to most mature commercial superalloys. Because the composition of these superalloys is relatively fixed, the simulated macrosegregation results are highly reliable. Combining the mushy zone distribution and isotherm distribution during solidification allows for intuitive analysis of the causes of certain solidification defects during production and for appropriate adjustments. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] Figure 1 This is a large die casting model of a high-temperature alloy blade of the present invention.

[0029] Figure 2 This is a schematic diagram of the simulation of the large mold casting, mold shell, and furnace of the present invention.

[0030] Figure 3 This is a temperature field distribution diagram of the large mold casting of the present invention; among them, (a) color and temperature color card, (b) outer blade cross section, (c) outer blade longitudinal section, (d) inner blade cross section, (e) inner blade longitudinal section.

[0031] Figure 4 This is a macrosegregation distribution diagram of the large mold casting of the present invention; among them, (a) color and C element content color card, (b) outer blade cross section, (c) outer blade longitudinal section, (d) inner blade cross section, (e) inner blade longitudinal section.

[0032] Figure 5 This is a distribution diagram of the mushy zone during the solidification process of the large mold casting of the present invention; among them, (a) color and temperature color card, (b) outer blade and inner blade at 1245s, (c) outer blade and inner blade at 1434s.

[0033] Figure 6 This is a distribution diagram of the miscellaneous crystals in the edge plate part of the large mold casting of the present invention; among them, (a) outer blade longitudinal section, (b) outer blade edge plate part cross section, (c) inner blade longitudinal section, (d) inner blade edge plate part cross section, (e) outer blade longitudinal section, (f) outer blade edge plate part cross section, (g) inner blade longitudinal section, (h) inner blade edge plate part cross section.

[0034] Figure 7 Flowchart of the simulation software in the embodiment.

[0035] Figure 8 It is a simulation grid diagram established by the simulation software in the embodiment.

[0036] Figure 1 Center: 1. Gate cup; 2. Blade runner; 3. Outer blade; 4. Inner blade; 5. Seed crystal segment; 6. Water-cooled copper plate; 7. Middle support.

[0037] Figure 2 Middle: 8. Casting mold shell; 9. Axis of symmetry; 10. Water-cooled copper plate; 11. Furnace heating zone; 12. Furnace drawing chamber. DETAILED DESCRIPTION

[0038] like Figures 1-6 As shown, a large-module casting simulation method for single-crystal double-row high-temperature alloy blades includes the following steps:

[0039] like Figure 1As shown, commercial modeling software is used to model the actual production size. The single crystal double-row large module casting is connected by a pouring cup 1, a blade runner 2, an outer blade 3, an inner blade 4 and a seed crystal segment 5, and is placed on a water-cooled copper plate 6. Among them, the blade runner 2 connects the pouring cup 1 and the upper part of the outer blade 3 and the inner blade 4, the lower side of the seed crystal segment 5 is in direct contact with the water-cooled copper plate 6, and the pouring cup 1 and the water-cooled copper plate 6 are connected by an intermediate support 7. The diameter of the pouring cup top is 150mm, and it is connected to the outer blade 3 and the inner blade 4 through the blade runner 2. The diameter of the blade runner 2 is 5mm. Such a design can reduce the generation of return material in this part without affecting the microstructure and mechanical properties of the casting, saving production costs for large module technology. The outer blades 3 and inner blades 4 are placed parallel to each other, with the closest distance between the blade edges being 20 mm and the distance between the blade bodies being 43 mm. This placement effectively avoids defects during the casting process of large-module castings and effectively improves the yield rate of high-temperature alloy blades. The blades shown in this invention are experimental models, with a height of 90 mm and a thickness of 7 mm. This design can better understand the location and control factors of defects during the casting process of large-module castings. The seed crystal segment 5 is 20 mm high and has a crystal orientation of [0 01]. It is attached to the underside of each blade and directly contacts the water-cooled copper plate 6. The water-cooled copper plate has a diameter of 300 mm and is filled with 20°C circulating cooling water.

[0040] After creating a single crystal double-row large module casting model using commercial modeling software, it was imported into simulation software to simulate the casting process. After importing the model, any areas of part overlap in the modeling software must be eliminated. This creates two overlapping meshes in the simulation software, which can cause errors and prevent subsequent simulations from proceeding. Using the software's built-in tools to remove these overlapping areas, meshing was performed. Considering computing power, the mesh size for the seed crystal and blade sections was set to 1 unit length, the mesh size for the pouring cup and runner sections was set to 4 units length, and the mesh size for the furnace section was set to 20 units length. The mesh was checked for errors. If there were fewer than two errors and they occurred in areas other than blades or seed crystals, the simulation software's built-in functions could be used to remove them. If there were more than two errors or they occurred in the blades or seed crystals, the mesh in the erroneous areas needed to be repaired using manual modification tools. Only after ensuring that the mesh was error-free could the next step be carried out. After meshing the casting and the buckle box, the mold shell needed to be created, with a thickness of 4-5 mm. When setting up the mold, the top surface of the pouring cup, the sides of the water-cooled copper plate, and the bottom surface must be adjusted to achieve a simulation effect that is completely consistent with the on-site experiment. After the mold is set up, a mesh check is also required. If there are fewer than five errors in this part, the software's built-in function can be used to repair them. If there are more than five errors, the mold must be deleted and re-established.

[0041] After the grid is established and no errors are ensured, the casting is processed symmetrically. Due to the limitation of computer computing power, it is necessary to set the large module casting to be axisymmetric. The large module casting has 24 blades in total. 1 / 12 water-cooled copper plate is selected for axisymmetric mirroring to generate the entire large module casting. Figure 2 As shown, with the vertical axis of symmetry 9 as the reference, the upper parts of a set of outer blades and inner blades are connected to the 1 / 12 pouring cup through the blade runner, and the lower parts of the outer blades and inner blades are connected to the 1 / 12 water-cooled copper plate 10 through the seed crystal segment, forming the casting mold shell 8. The outer side of the casting mold shell 8 is the furnace heating zone 11, and the bottom of the furnace heating zone 11 is the furnace drawing chamber 12. The buckle box part also needs to be modeled in advance and the 1 / 12 part is selected for assembly with the casting to ensure that the center line of the buckle box coincides with the center line of the casting.

[0042] After completing the symmetry setting, it is necessary to establish a three-dimensional mesh. After establishment, it is also necessary to check all three-dimensional meshes except the buckle box part to ensure that there are no errors before proceeding to the next simulation.

[0043] After the mesh is established, the physical properties and heat transfer coefficients for the simulation need to be set. The casting alloy is DD5. The density, Young's modulus, Poisson's ratio, and other parameters for the simulation alloy were pre-calculated using thermodynamic simulation software and imported into the simulation software. The mold shell material uses refractory aluminum from the simulation software's material library. The heat transfer coefficient is set according to the actual site conditions, and different temperatures are set for different locations of the buckle box to match the site conditions as closely as possible. After defining the physical properties and heat transfer coefficients, the nucleation parameters for the outer blades, inner blades, and seed crystal segment need to be defined. To match the actual casting process, the surface of the seed crystal segment contacting the water-cooled copper plate is defined as a single crystal nucleation with an orientation of [0 0 1]. To achieve a higher degree of consistency with the actual casting process, the four outer surfaces of the seed crystal segment are also defined as single crystal nucleations with the same orientation as the bottom surface. This ensures that the seed crystal segment is highly consistent with the actual seed crystal during the simulation. The outer and inner blades are defined as body nuclei with an average nucleation undercooling of 10 to 25 K. After setting the nucleation parameters, the casting's pull-out process needs to be defined. The buckle box is set to move in the -Z direction at a speed of 5mm / min, simulating the downward pull-out process during directional solidification. After completing this process, the casting process is checked for errors. Once these errors are correct, simulation can be performed. The computer calculates the directional solidification process of the large-module casting. The simulation results are used to analyze the temperature field, mushy zone distribution, stray crystal defects, and macrosegregation of the single-crystal double-row large-module blade.

[0044] The present invention is further described in detail below through simulation examples.

[0045] Example 1

[0046] This embodiment specifically describes the process of casting single crystal double-row large module castings through simulation, verifies the feasibility and simplicity of the present invention, and successfully predicts the stray crystal defects and macrosegregation of large module blades. The specific implementation method is as follows:

[0047] Use modeling software to create parts: pouring cup, blade runner, outer blade, inner blade, seed crystal segment, water-cooled copper plate, buckle box. The above parts are 1 / 12 of the overall large module casting. The circular array function in the modeling software can reproduce the complete large module casting. The buckle box part of all parts is consistent with the actual size. Since the buckle box part is considered to be a completely vacuum environment in the simulation software, it does not need to be consistent with the actual size, resulting in an increase in the amount of calculation. The buckle box part parts also need to be set to zero thickness in advance to prepare for the subsequent introduction into the simulation software. After the establishment of each part, assemble them and bring them into the simulation software as an igs format file. Each part can be identified separately in the simulation software.

[0048] like Figure 7 As shown in the figure, the flowchart of the simulation software in the embodiment will be described in detail below: First, all parts are checked for interference, and it is found that there is no interference. After that, all parts are eliminated from overlapping surfaces, and the results show that a total of 8 overlapping surfaces are eliminated. Select the furnace part alone, hide the rest, mesh the furnace part, and select a grid size of 20. Since the meshing is a dimensionless calculation, there is no need to consider the grid size unit here, only the number of grids needs to reach the expected value. After meshing, a grid check is performed, which shows that there is a problem with the edge of the furnace, because this edge line is the subsequent symmetry plane, so the error here does not need to be changed. Select the mesh of the furnace and define it as a buckle box, that is, to simulate the furnace in the directional solidification process. After the definition is completed, the next step of meshing can be continued. Select all parts except the furnace parts, and divide the grids for different parts by manual meshing. Divide the blade and seed crystal parts into a grid size of 1 unit length, and the rest of the parts into 4 unit lengths, and perform a grid check. The inspection results show that there are 2 errors. The error location is the attachment of the connection between the seed crystal and the water-cooled copper disk. The error is caused by the sharp angle of the grid. Use the built-in repair function to repair it. Check again and there are no errors. Select the mold shell function in the two-dimensional grid to establish the mold shell. When establishing the mold shell, select the upper surface of the pouring cup. The side and bottom surfaces of the water-cooled copper disk cannot be set as a shell to achieve a simulation effect that is completely consistent with the on-site experiment. The mold shell thickness is 4mm. After establishment, apply it and perform a mesh check on the mold shell. The results show that there are no errors. Select all parts for body mesh establishment. After establishment, perform a body mesh check. The results show that there are no errors. The specific results are as follows Figure 8 shown.

[0049] After successfully creating the mesh, symmetry needs to be set. Since the parts created using the modeling software are all 1 / 12 of the overall casting, it is necessary to use the symmetry function in the simulation software to achieve the effect of simulating the entire casting while reducing the amount of calculation. This is also a common method of reducing the amount of calculation in current simulation technology. Select the axisymmetry function, select the center point of the upper surface of the buckle box, the center point of the casting gate cup, and the center point of the lower surface of the buckle box, and perform axisymmetry. The result is displayed as 12.

[0050] Before running the simulation software, it is also necessary to set the gravity direction, physical parameters, heat transfer coefficient, nucleation coefficient, etc. In this embodiment, the specific settings are as follows: the gravity direction is set to -Z. Set the physical parameters of the casting part to DD5; set the physical parameters of the water-cooling plate part to Copper; set the physical parameters of the mold shell part to Refractory Alumina. Set the heat transfer coefficient between the casting and the water-cooling plate to 900~1200; set the heat transfer coefficient between the casting and the mold shell to 600~900; set the heat transfer coefficient between the water-cooling plate and the mold shell to 100~200. Set the average supercooling degree of surface nucleation to 12-20k; set the average supercooling degree of body nucleation to 15-20k.

[0051] First, the alloy DD5 was selected for the casting, and its simulated composition is shown in Table 1. The remaining parameters of the alloy were pre-calculated using thermodynamic simulation software, including density, Young's modulus, Poisson's ratio, etc. The calculation results of this embodiment are as follows: density is 7120-8588 kg / m 3 , Young's modulus ranged from 0 to 179,868 MPa, and Poisson's ratio ranged from 0.36 to 0.51. Because these parameters are temperature-dependent, the data above are all within the pouring temperature range. These data were then entered into the simulation software. The mold shell material used was refractory aluminum from the simulation software's material library, and the water-cooling copper plate material used was also from the simulation software's material library.

[0052] Table 1 Chemical composition of alloys used for simulation (wt.%)

[0053] Alloy B C Al Cr Co Mo Hf Ta W Re Ni Example 1 0.002 0.02 7.5 8.1 8.7 1.7 0.15 6.8 5.0 2.3 margin

[0054] After completing the physical property parameter settings, the corresponding heat transfer coefficients were set based on actual site conditions and relevant literature. The outer blades, inner blades, and seed crystal segments were selected to define the nucleation parameters. The surface of the seed crystal segment contacting the water-cooled copper disk was defined as single crystal nucleation in the surface nucleation method. The Euler angles were set to zero, and the nucleation orientation was set to [0 0 1]. The four outer surfaces of the seed crystal segment were also selected according to the above parameters. The blade segment was defined as a bulk nucleation, with an average nucleation undercooling of 18k and a standard deviation of 5k. After completing the nucleation parameter settings, the casting's extraction process was also defined. The buckle box was set to move in the -Z direction at a speed of 5.5mm / min, starting 300s after pouring. The maximum processor speed of 24 was selected for simulation. The directional solidification process of the large-module casting was simulated by computer. The simulation results analyzed the temperature field, mushy zone distribution, stray crystal defects, and macrosegregation of the single crystal double-row large-module blade.

[0055] like Figure 3 As shown, the outer blades ( Figure 3 b, 3c), inner blade ( Figure 3 d, 3e) isothermal diagrams of the cross section and longitudinal section. The liquidus temperature and solidus temperature of the DD5 alloy in the figure are calculated in advance by thermodynamic simulation software. By analyzing the temperature field of the outer blades and the inner blades, the temperature changes and heat transfer states at different positions of the blades during the large-module casting of single-crystal double-row blades can be clearly obtained, and the key positions and possible causes of casting defects can be analyzed. Since the internal temperature field of the casting is extremely complex during the large-module casting process, it is easy to produce a shadow effect in the center of the casting, resulting in a large temperature difference on both sides of the casting. Therefore, it is particularly important to analyze the isotherms of the blades at different casting times. By observing the temperature change trend, the internal heat transfer can be analyzed, and the process parameters can be improved to avoid possible casting defects in advance.

[0056] like Figure 4 As shown in the simulation results, it can be observed that the outer blades ( Figure 4 b, 4c), inner blade ( Figure 4 d, 4e) cross-sectional and longitudinal C element macrosegregation diagrams. As the casting is continuously pulled downward, the large module blades are lowered from the higher temperature furnace heating zone to the lower temperature furnace pulling chamber and gradually solidify. During the solidification process of the large module blades, the uneven distribution of the internal heat field of the casting leads to different solidification rates and temperature gradients at different locations of the blades. Figure 4As shown, the macrosegregation at the inner blade edge is significantly smaller than at other locations. This is likely due to the higher temperature gradient and faster solidification rate at the edge, which results in smaller macrosegregation than at other locations. Therefore, the simulation method of the present invention can intuitively observe the macrosegregation of single-crystal blades, thereby adjusting process parameters and casting geometry to prevent casting defects in advance, increase the blade yield of single-crystal large-module casting technology, and reduce costs and increase efficiency.

[0057] like Figure 5 As shown in the figure, the simulation results show the distribution of the mushy zone at different times for the outer and inner blades. By observing the mushy zone, the solidification sequence of the outer and inner blades can be intuitively determined. Combining the isotherm diagram and macrosegregation diagram at the same time, the solidification state at specific locations at a specific time can be analyzed. This can then be combined with solidification defect results or experimental microstructure analysis to form theoretical support.

[0058] like Figure 6 As shown in the figure, the simulation results show that the outer blades ( Figure 6 a, 6e), inner blade ( Figure 6 c, 6g) crystal orientation distribution diagram. The nucleation function of the simulation software is used to effectively simulate the actual seed crystal solidification process. It can be observed that the platform with a sudden change in the cross section is prone to miscellaneous crystal defects. Through the simulation results ( Figure 6 b, 6d) can intuitively observe the location and size of the stray crystals formed. By changing the process parameters such as pouring temperature, mold shell thickness, pulling rate, etc., combined with the isotherm diagram and macrosegregation diagram, the occurrence of stray crystal defects can be effectively avoided.

[0059] The results of the embodiments show that the present invention uses simulation software to input the same process parameters as those in the experiment or production to calculate the isotherm distribution map, mushy zone distribution map, macro segregation distribution map, crystal orientation distribution map, etc. of the target casting. It can accurately, efficiently and flexibly change various process parameters for simulation, and can more accurately simulate the entire process of directional solidification of single crystal blades, quickly find the ideal single crystal large module casting process parameters, estimate the risk of casting defects, save time and operating costs, and has important application value for the casting defect estimation and actual production application of single crystal large module casting technology.

Claims

1. A large-module casting simulation method for single-crystal double-row high-temperature alloy blades, characterized in that: The method includes: constructing a single crystal double-row large module model, meshing the casting, and setting the process parameters of the simulation: setting the symmetry axis of the casting, setting the physical property parameters of the alloy, setting the directional solidification heat transfer coefficient, setting the casting pulling rate, and setting the crystal orientation and nucleation parameters of the seed crystal method; through numerical simulation, the solidification process and the position of the impurity defects of the large module casting during the directional solidification process are obtained, the change process of the isotherm and the mushy zone during the directional solidification process is obtained, and the macro-segregation distribution map at any position of the blade is intuitively observed, providing a theoretical basis for the microstructure analysis during the experiment.

2. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 1 is characterized in that: The single crystal double-row large module model is constructed as follows: the model includes a pouring cup, a blade runner, an outer blade, an inner blade, a seed crystal segment, an intermediate support, and a water-cooled copper plate; each of the above components is separately established in the modeling software and then assembled to ensure that it is recognized as a separate individual in the simulation software; the entire large module model includes 24 single crystal blades and 24 seed crystal segments. The blades are divided into two layers, the inner and outer layers, with the closest distance between the two layers being 20 mm. Each layer is equipped with 12 single crystal blades and 12 seed crystal segments, each blade is 90 mm high and 7 mm thick; the seed crystal segment connects the blade to the water-cooled copper plate, and the bottom of the seed crystal segment is in direct contact with the water-cooled copper plate. The seed crystal segment is 20 mm high, the crystal orientation is [0 0 1], and the diameter of the water-cooled copper plate is 300 mm; the pouring cup and the water-cooled copper plate are connected by an intermediate support, and the diameter of the pouring cup is 150 mm; the bottom of the pouring cup is connected to the top of the blade through the blade runner, and the diameter of the blade runner is 5 mm.

3. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 1 is characterized in that: Casting mesh division means: first, two-dimensional meshes are established for the large module casting, water-cooled copper plate, casting mold shell, and furnace part. After checking that there are no errors in the meshes, three-dimensional meshes are established and checked for the above parts except the furnace.

4. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 3 is characterized in that: The two-dimensional grid establishment method is as follows: first, the overlapping areas of the large mold casting are eliminated to ensure that there is no overlap between the parts, and then the parts are assembled to connect with each other; after completing the above steps, the seed crystal segment and the blade part are divided into a two-dimensional grid with a unit length of 1; the gate cup and the blade runner part are divided into a two-dimensional grid with a unit length of 4; the furnace part is divided into a two-dimensional grid with a unit length of 20; after completing the above steps, each part is grid checked separately.

5. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 1 is characterized in that: When setting the process parameters for simulation, it is necessary to set the alloy physical properties, directional solidification heat transfer coefficient, casting pulling rate, and nucleation parameters in sequence; after ensuring that the values ​​are consistent with the experimental or production process to the greatest extent possible, simulation can be performed to obtain the most realistic simulation results.

6. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 5 is characterized in that: The physical properties of the alloy refer to: the alloy grade selected for the simulation process is DD5, and the density, Young's modulus, and Poisson's ratio of the alloy used for the simulation are calculated in advance using thermodynamic simulation software and brought into the simulation software.

7. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 5 is characterized in that: The heat transfer coefficient of directional solidification refers to: the heat transfer coefficient includes the heat transfer coefficient between the casting and the mold shell, the casting and the water-cooled copper plate, the mold shell and the water-cooled copper plate, the outer surface of the mold shell and the furnace, and the water-cooled copper plate and the furnace. The corresponding coefficient is set according to the literature.

8. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 5 is characterized in that: The casting withdrawal rate refers to the following: During the simulation process, due to the large amount of calculation required for the casting parts, the casting is usually fixed and the furnace moves in the opposite direction during the withdrawal process to simulate the directional solidification process; among them, the furnace is set to move at a speed of 5mm / min along the -Z axis to simulate the actual directional solidification process.

9. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 5, characterized in that: The nucleation parameters are as follows: the surface of the seed crystal segment contacting the water-cooled copper disk is defined as a single crystal nucleation, and the orientation is set to [0 0 1]. In order to better match the actual situation, the four outer surfaces of the seed crystal segment are also defined as single crystal nucleations; the blade part is defined as a body nucleation, and the average nucleation undercooling is 10~25k. Then, the position and amount of stray crystal formation under different process parameters under the same nucleation parameters are observed.

10. The large-module casting simulation method for single-crystal double-row high-temperature alloy blades according to claim 1, characterized in that: After completing the simulation, the isotherm distribution map, mushy zone distribution map, macro segregation distribution map, and crystal orientation distribution map of the blade are visually observed, and the causes of the casting defects are analyzed to increase the yield of large-module blades.

Images