A thin-walled hollow turbine blade size precision control method based on variable shrinkage factor

Abstract

This invention relates to the field of turbine blade dimensional control technology, and particularly to a method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on a variable shrinkage factor. The method involves numerical simulation of the solidification stage of turbine blade precision casting to obtain a design model; numerical simulation of the unconstrained stage of turbine blade precision casting to obtain a casting model; performing inverse deformation iterative calculations on the obtained design model and casting model to obtain an optimized mold cavity and determine the deformation amount; repeating steps 1-3 using the controlled variable method until the dimensional deformation accuracy of the thin-walled hollow turbine blade during investment casting is controlled. The method successfully predicts the deformation distribution of the turbine blade, calculates the shrinkage rate of each node of the turbine blade in the XYZ directions, obtains a continuously changing shrinkage factor, and substitutes it into the improved inverse deformation iterative formula to obtain a new mold. Experiments verify that the proposed method controls the deformation amount within the design requirement of ±0.15mm, indicating that the dimensional accuracy of the thin-walled hollow turbine blade during investment casting is effectively controlled.

Description

Technical Field

[0001] This invention relates to the field of turbine blade size control technology, and in particular to a method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on a variable shrinkage factor. Background Technology

[0002] As a critical hot-end component of aero-engines, turbine blades operate in extreme environments of high temperature, high pressure, and high load, and their quality directly affects and determines engine performance. Investment casting is a near-net-shape forming technology with high raw material utilization and minimal subsequent machining, making it particularly suitable for workpieces that cannot be machined or where machining would be too wasteful of material. Furthermore, castings produced through investment casting do not have sharp edges or burrs like machined parts. Turbine blades, with their complex serpentine internal cavities, uneven blade thickness, high curvature, and trailing edge cleavage, are therefore typically produced using investment casting.

[0003] In domestic and international industrial production, the Bridgman process is commonly used to produce high-temperature alloy or single-crystal turbine blades. During directional solidification, the different cooling rates and materials of different parts of the casting result in varying shrinkage, leading to non-uniform thermal stress and making the blades highly susceptible to deformation. Furthermore, the removal of constraints from the mold shell, ceramic core, and gating system causes the mechanical resistance stress to redistribute through deformation, resulting in non-linear and non-uniform blade deformation. Currently, the yield rate of hollow turbine blade precision casting is low, with approximately half of the defects due to dimensional deviations. Therefore, to improve the manufacturing yield of hollow turbine blades, it is essential to strictly control the dimensional accuracy of the turbine blades during the investment casting process. Thus, a precision control method is urgently needed to address the aforementioned technical problems. Summary of the Invention

[0004] The purpose of this invention is to improve the dimensional accuracy of turbine blades during investment casting. This invention proposes a method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on a variable shrinkage factor. First, numerical simulations were performed on the directional solidification and constraint removal processes. The deformation of the blade body and shroud was relatively large, and the deformation distribution of the turbine blade was successfully predicted. Then, the deformation distribution at each node of the turbine blade, P = {P1, P2, ..., P...}, was calculated. n The contraction rate λ in the XYZ direction is λ = [λ x λ y λ z ], thus obtaining the continuously varying contractile factor K = [K x K y K z Substitute into the improved inverse deformation iterative formula Obtain the new mold P′ M(x, y, z). Finally, cross-sectional data of the blades were collected using a coordinate measuring machine (CMM). The experiment verified that the proposed method successfully controlled the deformation within the design requirement of ±0.15 mm, indicating that the dimensional accuracy of the thin-walled hollow turbine was effectively controlled during the investment casting process.

[0005] To achieve the above objectives, the present invention provides a technical solution:

[0006] A method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on variable shrinkage factor includes the following steps:

[0007] Step 1: Numerical simulation of the solidification stage of turbine blade precision casting to obtain the design model, denoted as P. D (x,y,z);

[0008] Step 2: Perform numerical simulation of the constraint removal stage of turbine blade precision casting to obtain the casting model, denoted as P. C (x,y,z);

[0009] Step 3: Perform inverse deformation iterative calculations on the design model and casting model obtained in Steps 1 and 2 to obtain the optimized mold cavity, denoted as P′. M (x,y,z), determine the deformation ΔD=max{ΔD1,ΔD2,...,ΔD n};

[0010] Step 4: Repeat steps 1-3 using the controlled variable method until the dimensional deformation accuracy of the thin-walled hollow turbine blades during the investment casting process is controlled.

[0011] A further technical solution of the present invention is: in step 1, the numerical simulation process of the solidification stage of the turbine blade precision casting is as follows:

[0012] Step 1.1: Convert the thin-walled hollow turbine blade model into a finite element model;

[0013] Step 1.2: Set the key parameters for investment casting to meet the first preset conditions, complete the numerical simulation of the solidification stage of the turbine blade precision casting, and derive the initial design model P required for anti-deformation. D (x,y,z).

[0014] A further technical solution of the present invention is: in step 2, the process of numerical simulation for calculating the constraint removal stage of precision casting is as follows:

[0015] Step 2.1: Provide the initial temperature field, stress field, and displacement field for the unconstraint stage;

[0016] Step 2.2: Set displacement constraints for the casting to meet the second preset condition, and complete the numerical simulation of the constraint removal stage of precision casting.

[0017] Step 2.3: Remove the constraints of the mold shell, ceramic core, and gating system, and export the deformed casting model P. C (x,y,z).

[0018] A further technical solution of the present invention is as follows: In step 1.1, the process of converting the hollow turbine blade model into a finite element model is as follows: the model is split using software tools to obtain multiple volume files, which are then imported into HyperMesh to generate a mesh. The model is discretized into Tetra elements of different sizes. The model is divided into trailing edge, leading edge, blade body, gating system, and furnace body. Different mesh sizes are set, and the turbine blade is discretized into a set of n nodes, where n represents the total number of turbine blade nodes.

[0019] A further technical solution of the present invention is as follows: In step 1.2, the key parameters for investment casting include: material type, mold shell thickness h, drawing speed v, and cold copper temperature T. ch , casting temperature T pr Mold shell preheating temperature T po The first preset condition is that the calculation will terminate when the temperature drops to 300℃.

[0020] A further technical solution of the present invention is: in step 2.1, the post-processing result of the final step of the mapping solidification stage is extracted as the initial temperature field, stress field and displacement field of the unconstrained stage.

[0021] A further technical solution of the present invention is: in step 2.2, the "four-point constraint method" is used as the displacement constraint of the casting; the second preset condition refers to the temperature dropping from 300℃ to 25℃ as the calculation termination condition.

[0022] A further technical solution of the present invention is: in step 3, the inverse deformation iterative calculation process is as follows:

[0023] The design model P obtained in steps 1 and 2 D (x,y,z) and casting model P C Substituting (x,y,z) into the inverse deformation iteration formulas (1) and (2):

[0024]

[0025]

[0026] Among them, P′ M (x,y,z) represents the mold cavity, P M (x,y,z) represents the inverse deformation optimization model; λ represents the shrinkage rate; K represents the shrinkage factor;

[0027] The optimized mold cavity P′ is obtained M (x,y,z), error is adopted Determine the deformation amount ΔD = max{ΔD1, ΔD2, ..., ΔD} at all discrete points. n Does it meet the design requirements?

[0028] A further technical solution of the present invention is: in step 3, the turbine blade shrinkage rate λ={λ1,λ2,...,λ n}, decompose it along the XYZ direction into λ1=[λ 1x λ 1y λ 1z ], and thus obtain the variable shrinkage factor K = [K] related to the shrinkage rate λ. x K y K z Then, the shrinkage factor K of the turbine blade node in different directions is given by formula (3), which reduces the number of reverse compensation iterations:

[0029]

[0030] A further technical solution of the present invention is: in step 3, the mold cavity P′ M The calculation process for (x, y, z) is as follows:

[0031] P D P C Substituting K into the inverse deformation iteration formula (1), we get:

[0032]

[0033] Where j represents the number of inverse deformation iterations; then the final mold shape that meets the requirements is:

[0034] P′ M (x,y,z)=P′ M (x j+1 ,y j+1 ,z j+1 (5)

[0035] Using the new mold P′ M (x,y,z) Perform numerical simulations again for the directional solidification and unconstraint stages, and obtain the optimized deformation amount by registering the deformed model with the design model.

[0036] A further technical solution of the present invention is: the turbine blade structure is divided into constrained structure and unconstrained structure according to the constraint method, wherein the shrinkage rate of the constrained structure is calculated by formula (6):

[0037]

[0038] Where point O is the shrinkage center of the blade during investment casting, point B is on the mold outline and becomes point C after deformation, B1 and B2 represent the nodes on the blade base and blade back side of the casting, respectively, B3 and B4 represent the nodes near the ceramic core on the blade base and blade back side, respectively, and C1, C2, C3 and C4 are the corresponding points after deformation, λ n,(x,y,z) This represents the contraction rate at the nth node with coordinates (x, y, z);

[0039] The shrinkage rate of an unconstrained structure is calculated using formula (7):

[0040]

[0041] Beneficial effects:

[0042] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0043] (1) This invention proposes a method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on variable shrinkage factor. The directional solidification process is simulated to predict the deformation distribution of the turbine blade. After removing constraints, the deformation of the blade body and edge plate further increases. Then, by calculating the shrinkage rate of each finite element node of the turbine blade, it is decomposed along the three directions XYZ to obtain the variable shrinkage factor related to the shrinkage rate. By improving the traditional inverse deformation iteration formula, the mold design is combined with the variable shrinkage factor and shrinkage rate to reduce the number of reverse compensation iterations and reduce the error accumulation caused by two-dimensional surface reconstruction. The mold cavity that meets the requirements can be obtained through one or a few iterations. This method can effectively control the dimensional accuracy of hollow turbine blades in the investment casting process and improve the blade qualification rate.

[0044] (2) This invention proposes a method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on variable shrinkage factor. The cross-sectional data of the blades are collected using a coordinate measuring machine (CMM). Experiments have verified that the proposed method successfully controls the deformation within the design requirement of ±0.15mm, indicating that the dimensional accuracy of the thin-walled hollow turbine blades in the investment casting process is effectively controlled, thereby improving the dimensional accuracy of the turbine blades in the investment casting process. Attached Figure Description

[0045] Figure 1 This is a flowchart of the method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on variable shrinkage factor according to the present invention;

[0046] Figure 2 This is a schematic diagram of the deformation of a turbine blade after directional solidification and unconstraint in one embodiment of the present invention.

[0047] Figure 3 This is a schematic diagram of the contraction factors of three sections of turbine blades II, V, and IX according to one embodiment of the present invention;

[0048] Figure 4 This is a schematic diagram of the deformation of a casting after anti-deformation optimization according to one embodiment of the present invention;

[0049] Figure 5 This is a schematic diagram of a coordinate measuring experiment according to one embodiment of the present invention;

[0050] Figure 6 This is the actual deformation of a turbine blade as measured in one embodiment of the present invention.

[0051] The reference numerals in the figure are:

[0052] 1—Turbine blade; 2—Anti-vibration platform; 3—Computer; 4—Drive unit; 5—Probe system; 6—Positioning fixture. Detailed Implementation

[0053] The present invention will be further described in detail below with reference to specific embodiments, but this does not constitute any limitation on the present invention.

[0054] Example 1

[0055] A method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on variable shrinkage factor includes the following steps:

[0056] Step 1: Numerical simulation of the solidification stage of turbine blade precision casting to obtain the design model, denoted as P. D (x,y,z);

[0057] Step 2: Perform numerical simulation of the constraint removal stage of turbine blade precision casting to obtain the casting model, denoted as P. C (x,y,z);

[0058] Step 3: Perform inverse deformation iterative calculations on the design model and casting model obtained in Steps 1 and 2 to obtain the optimized mold cavity, denoted as P′. M (x,y,z), determine the deformation ΔD=max{ΔD1,ΔD2,...,ΔD n};

[0059] Step 4: Repeat steps 1-3 using the controlled variable method until the dimensional deformation accuracy of the thin-walled hollow turbine blades during the investment casting process is controlled.

[0060] Example 2

[0061] Taking a certain type of aero-turbine blade as an example, the specific implementation process of the present invention's method for controlling the dimensional accuracy of a thin-walled hollow turbine blade based on a variable contraction factor is as follows: Figure 1 As shown:

[0062] Step 1:

[0063] Using tools such as UG software, the CAD model was split into multiple volume files, which were then imported into HyperMesh for mesh generation. To approximate the blade geometry and improve computational efficiency, the model was discretized into Tetra elements of different sizes. The mesh sizes for the trailing edge, leading edge, blade body, gating system, and furnace body were 0.1, 0.2, 2, 5, and 10 mm, respectively. The turbine blade was discretized into a set of n nodes, P = {P1, P2, ..., P...}. 26966 Simulation parameters are set as follows: mold shell thickness h = 5mm, material type SAND_Silica, pulling speed v = 6mm / min, and cold copper temperature T. ch =20℃, casting temperature T pr =1550℃, mold shell preheating temperature T po =1550℃, cooling to 300℃ is the calculation termination condition, completing the numerical simulation of the solidification stage of the turbine blade precision casting, and deriving the initial CAD design model P. D ={P D1 ,P D2 ,...,P D,26966}

[0064] Step 2:

[0065] Figure 2 This is a schematic diagram of the turbine blade deformation after directional solidification and deconstraint. The *.vdb and p.dat simulation files from the solidification stage are copied, and the post-processing result *.unf from the final solidification step is extracted / mapped as the initial temperature, stress, and displacement fields for the deconstraint stage. The casting is constrained using a "four-point constraint method": the XYZ displacement of node 28837 is constrained to 0; point 28793 is selected along the X direction from this point; the Y displacement is constrained to 0; point 28773 is selected along the Y direction from this point; the Z displacement is constrained to 0; and point 18671 is selected along the Z direction from this point, constraining its X displacement to 0. The calculation terminates when the temperature drops from 300℃ to 25℃. The simulation is then submitted, completing the deconstraint stage simulation of the precision casting process. The temperatures of 300℃ and 25℃ are derived from data from actual turbine blade production; these values ​​are kept consistent with actual production data to improve simulation accuracy.

[0066] After removing constraints such as the mold shell, crystal guide segment, ceramic core, and gating system, the turbine blades undergo further deformation. The deformed mesh P is then derived. C ={P C1 ,P C2 ,...,P C,26966 The deformed model is then registered with the CAD model to obtain the deformation diagrams of sections II, V, and IX.

[0067] Step 3:

[0068] Inverse deformation optimization is performed using the pre- and post-processing files from steps 1 and 2 to control the deformation of turbine blades during investment casting. The traditional inverse deformation iterative formula is:

[0069] P′ M =P M -K(P C -P D (1)

[0070] Where P D (x,y,z), P C (x,y,z) and P M (x, y, z) represent the CAD design model, casting, and anti-deformation optimization model, respectively, and K is the shrinkage factor, typically an empirical constant. This paper introduces a structural shrinkage rate to transform K into a continuously varying quantity related to the turbine blade structure, which better reflects the deformation of the blade during investment casting. The transformed equation is:

[0071]

[0072] To accurately describe blade contraction, K is decomposed along the XYZ direction into K = [K x K y K z Then the contraction factor of the turbine blade node in different directions is:

[0073]

[0074] Turbine blade structures can be classified into constrained and unconstrained structures according to the constraint method. Assuming point O is the shrinkage center of the blade during investment casting, and point B is on the mold profile, becoming point C after deformation, the shrinkage rate of the constrained structure can be calculated as follows:

[0075]

[0076] B1 and B2 represent the nodes on the blade basin and blade back side of the casting, respectively; B3 and B4 represent the nodes near the ceramic core on the blade basin and blade back side, respectively; and C1, C2, C3 and C4 are the corresponding points after deformation.

[0077] The calculation of the shrinkage rate of an unconstrained structure can be expressed as:

[0078]

[0079] The CAD design model P obtained in steps 1 and 2 D ={P D1 ,P D2 ,...,P Dn} and casting model P C ={P C1 ,P C2,...,P Cn Substituting into the inverse deformation iteration formula (2), we obtain the optimized mold cavity P′. M (x,y,z), error is adopted Determine the deformation amount ΔD = max{ΔD1, ΔD2, ..., ΔD} at all discrete points. n Does it meet the design requirements?

[0080] Figure 3 This is a schematic diagram of the shrinkage factors for three sections of the turbine blade: II, V, and IX. By combining the shrinkage rate with mold design to reduce the number of reverse compensation iterations, the turbine blade shrinkage rate is calculated as λ = {λ1, λ2, ..., λ...}. 26966}, decompose it along the XYZ direction into λ1=[λ 1x λ 1y λ 1z ], thus obtaining the variable shrinkage factor K = [K x K y K z ], P D P C Substituting K into the inverse deformation iterative formula

[0081]

[0082] Where j represents the number of inverse deformation iterations. The final mold shape that meets the requirements is then...

[0083] P′ M (x,y,z)=P′ M (x j+1 ,y j+1 ,z j+1 (7)

[0084] Using the new mold P′ M (x,y,z) Numerical simulation of the directional solidification and unconstraint stages is performed again. The deformed model is registered with the CAD model to obtain the optimized deformation of the three sections II, V and IX. Figure 4 This is a schematic diagram of the deformation of the casting after anti-deformation optimization.

[0085] Step 4:

[0086] Figure 5This is a schematic diagram of a coordinate measuring machine (CMM) experiment. The experimental measuring device is a CMM manufactured by Hexagon (Qingdao) Co., Ltd., model GLOBAL STATUS 121510. The drive unit 4 controls the movement of the guide rail, while the anti-vibration platform 2 and positioning fixture 6 improve the stability of the turbine blade 1 under test. The measurement results are transmitted to the computer 3 via the probe system 5 and displayed in real time. The probe radius is 1 mm, and the laboratory temperature is maintained at approximately 25℃. After acquiring the cross-sectional point cloud data using the CMM, the measured point cloud is registered with the CAD theoretical model to obtain the actual deformation of the turbine blade 1 during the entire precision casting process.

[0087] Figure 6 It is the actual deformation of the turbine blade measured in practice; due to the presence of slits and large curvature structures on the blade, the deformation at its trailing edge is relatively large, resulting in a large amount of deformation during investment casting. Figure 6 As can be seen, the simulation and experimental results show the same trend, with errors within ±0.15mm of the design requirements. After optimization, the average maximum deformation of different cross-sections changed from 0.3736mm to 0.1154mm, and the maximum deformation actually measured by the coordinate measuring machine was 0.1272mm. The results show that the proposed method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on variable shrinkage factor successfully controls the deformation of turbine blades during investment casting.

[0088] Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of the present invention without departing from the principles and spirit of the present invention.

Claims

1. A method for controlling the dimensional accuracy of thin-walled hollow turbine blades based on variable shrinkage factors, characterized by: The methods and steps include the following: Step 1: Numerical simulation of the solidification stage of the turbine blade investment casting, and the design model is obtained, denoted as ; Step 2: Calculate the numerical simulation of the turbine blade investment casting de-constraint stage to obtain a casting model, denoted as ; Step 3: Perform inverse deformation iterative calculations on the design model and casting model obtained in Steps 1 and 2 to obtain the optimized mold cavity, denoted as... Determine the amount of deformation ; Step 4: Repeat steps 1-3 using the controlled variable method until the dimensional deformation accuracy of the thin-walled hollow turbine blades during the investment casting process is controlled. In step 3, the inverse deformation iterative calculation process is as follows: Substitute the design model and casting model obtained in steps 1 and 2 into the inverse deformation iteration formulas (1) and (2): （1） （2） wherein, i.e. the mold cavity, is a counter-deformation optimization model; is the shrinkage factor; Optimized mold cavity , the error is , the deformation of all discrete points is determined whether it meets the design requirements; Calculate the shrinkage rate of turbine blades Decompose it along the XYZ direction into Thus, the shrinkage rate is obtained. Related variable contractility factor Then the shrinkage factor of the turbine blade node in different directions is given by formula (3), which reduces the number of reverse compensation iterations: （3）。 2. The precision control method of claim 1, wherein: In step 1, the numerical simulation process for the solidification stage of the turbine blade precision casting is as follows: Step 1.1: Convert the thin-walled hollow turbine blade model into a finite element model; Step 1.2: Set the key parameters for investment casting to meet the first preset conditions, complete the numerical simulation of the solidification stage of turbine blade precision casting, and derive the initial design model required for anti-deformation. The first preset condition refers to the calculation termination condition being a temperature drop to 300℃.

3. The precision control method of claim 2, wherein: In step 1.1, the process of converting the hollow turbine blade model into a finite element model is as follows: the model is split using a tool to obtain multiple volume files, which are then imported into HyperMesh to generate a mesh. The model is discretized into Tetra elements of different sizes. The model is divided into trailing edge, leading edge, blade body, gating system, and furnace body. Different mesh sizes are set, and the turbine blade is discretized into a set of n nodes, where n represents the total number of turbine blade nodes.

4. The precision control method of claim 2, wherein: The key parameters in step 1.2 include: material type, shell thickness h, drawing rate v, cold copper temperature T ch , pouring temperature T pr , shell preheating temperature T po .

5. The precision control method according to claim 1, characterized in that: In step 2, the numerical simulation process for calculating the constraint removal stage of precision casting is as follows: Step 2.1: Provide the initial temperature field, stress field, and displacement field for the unconstraint stage; Step 2.2: Set displacement constraints for the casting to meet the second preset condition and complete the numerical simulation of the precision casting deconstraint stage; the second preset condition refers to the calculation termination condition being a temperature drop from 300℃ to 25℃. Step 2.3: Remove the constraints of the mold shell, ceramic core, and gating system, and export the deformed casting model. .

6. The precision control method of claim 5, wherein: In steps 2.1 and 2.2, the post-processing results of the final step of the mapping solidification stage are extracted as the initial temperature field, stress field, and displacement field of the deconstraint stage; the "four-point constraint method" is used as the displacement constraint of the casting.

7. The precision control method of claim 1, wherein: Mold cavity The calculation process is: Substituting , and into the inverse deformation iteration formula (1), we have: （4） in, j Let represent the number of inverse deformation iterations; then the final mold shape that meets the requirements is... （5） Using new molds The numerical simulation of the directional solidification and the de-constraint stage is carried out again, and the optimized deformation is obtained by matching the deformed model with the design model.

8. The precision control method of claim 1, wherein: Turbine blade structures are classified into constrained structures and unconstrained structures according to the constraint method. The formula for calculating the shrinkage rate of the constrained structure is (6): （6） Point O is the shrinkage center of the blade during investment casting, and point B is on the mold outline, which becomes point C after deformation. These represent the nodes on the blade base and blade back of the casting, respectively. These represent the nodes near the leaf basin and the ceramic core on the back of the leaf, respectively. These are the corresponding points after deformation; This indicates that the coordinates of the nth node are Shrinkage rate at the location; The formula (7) for calculating the shrinkage rate of an unconstrained structure is: （7）。

Images