Data-Driven Design of Cast Nickel-Based Superalloy and Precision Forming of Complex Castings
WANG Donghong1, SUN Feng1,2(), SHU Da1,2(), CHEN Jingyang3, XIAO Chengbo3, SUN Baode1,2
1. Shanghai Key Lab of Advanced High-Temperature Materials and Precision Forming and State Key Laboratory of Metal Matrix Composites, School of Materials Science and Engineering, Shanghai Jiao Tong University, Shanghai 200240, China 2. Materials Genome Initiative Center, Shanghai Jiao Tong University, Shanghai 200240, China 3. Science and Technology on Advanced High Temperature Structural Materials Laboratory, AECC Beijing Institute of Aeronautical Materials, Beijing 100095, China
Cite this article:
WANG Donghong, SUN Feng, SHU Da, CHEN Jingyang, XIAO Chengbo, SUN Baode. Data-Driven Design of Cast Nickel-Based Superalloy and Precision Forming of Complex Castings. Acta Metall Sin, 2022, 58(1): 89-102.
The development of material genomics engineering and intelligent material-processing technology provides new ideas for researching, developing, and manufacturing key thermal superalloy components of aeroengines. Based on the demand for superalloy materials and casting processing, a high-throughput dynamic simulation software system was developed. Combined with the screening criteria of nickel-based casting superalloy, a new nickel-based casting superalloy was selected and developed from more than 5.2 million-component combinations. High-temperature durability at 815oC and 400 MPa is better than foreign Inconel 939 superalloy. For the precision molding of complex superalloy casting, the data-driven process of the casting deformation is integrated, which reveals the correlation between the process parameters and size precision during solidification deformation. Thus, a data-driven process parameter optimization method is proposed herein. A data-driven casting outlet design method based on the model and algorithm, combined with the test design and multi-target genetic algorithm, which optimized the casting process parameters, was established, and the production rate of the casting process increased by 13.39% after the test verification. The combination of data-driven component design and data model-based process design will accelerate the development and application of aviation materials and components.
Fund: National Science and Technology Major Project of China(J2019-VI-0004-0117);National Natural Science Foundation of China(51821001);National Key Research and Development Program of China(2016YFB0701405);Fund of the Science and Technology on Advanced High Temperature Structural Materials Laboratory(6142903200105)
About author: SUN Feng, associate professor, Tel: (021)54748974, E-mail: fsun@sjtu.edu.cn
Fig.1 Framework of thermodynamic/kinetic simulation software for novel cast nickel-based superalloys
Fig.2 Typical property diagram and illustrated feature parameters (Tm—melting point, TCP—topologically closed-packed, ΔT—solidification temperature range)
Element
Min.
Max.
Step
Cr
10
25
1
Al
0
6
1
Ti
0
6
1
Co
15
30
1
Mo
0
11
1
W
0
15
1
Nb
0
6
1
Ni
40
70
-
Al + Ti
0
10
1
Mo + W
0
15
1
Table 1 Composition range for high-throughput computation
Fig.4 Data-driven framework for prediction of distortion of investment casting (PA—pattern allowance, RSM—response surface methodology, RBF—radial basis function, DOE—design of experiment, AlloyTemp and ShellTemp—temperatures of alloy and shell)
Parameter
Packing pressure / MPa
Packing time / s
Injection speed / (cm3·s-1)
Injection temperature / oC
Upper bound
0.5
10
30
62
Lower bound
5
45
300
70
Table 2 Design variables for dimensional error simulation of wax pattern and each range
Parameter
AlloyTemp / oC
ShellTemp / oC
PA
Upper bound
1600.00
1100.00
0.02
Lower bound
1450.00
900.00
0
Table 3 Design variables for solidification deformation simulation and each range
Fig.5 Data-driven framework for prediction of porosity and process design in investment casting (HTC—heat transfer coefficient)
Parameter
Diameter
Length
AlloyTemp
ShellTemp
HTC
mm
mm
oC
oC
W·m-2·K-1
Upper bound
40.00
50.00
1600.00
1100.00
900
Lower bound
20.00
20.00
1450.00
900.00
750
Table 4 Design variables for prediction of porosity and each range
Fig.6 Master alloy ingot (a), testing specimens (b), and shape and dimensions of a typical casting (unit: mm) (c)
Alloy
Ni
Co
Cr
Al
Ti
Nb
W
Mo
1
44
30
15
4
1
2
4
0
2
54
20
15
3
3
0
5
0
Table 5 Screened basic alloy compositions
Fig.7 Creep performance of screened alloys and several typical cast superalloys (evaluated by JMatPro)
Fig.8 SEM images of Alloy 1 (a, b) and Alloy 2 (c, d) (a, c) as-cast (b, d) after solid solution
Fig.9 Interdendritic eutectics of Alloy 1 (a) and Alloy 2 (b), and EDS analyses (atomic fraction, %) of γ/γ' eutectics (c) and MC carbides of Alloy 1 (d) and Alloy 2 (e)
Fig.10 Typical mechanical property curves of SJTU-1 alloy (Rm—ultimate tensile strength, Rp0.2—yield strength, A—elongation) (a) tensile tests at room temperature (RT) and 815oC (b) creep at 815oC and 400 MPa
Fig.11 Optimization results of the injection process of wax pattern (unit: mm)(a) optimization process (b) minimum warpage optimization
Fig.12 The relation matrix among the process parameters of wax pattern
Fig.13 Two-dimensional response surface model among injection process parameters of wax pattern (a) volume shrinkage response surface of injection speed combine with packing time (b) volume shrinkage response surface of injection temperature combine with packing time (c) volume shrinkage response surface of injection speed combine with packing pressure (d) warpage response surface of injection speed combine with packing time (e) warpage response surface of injection temperature combine with packing time (f) warpage response surface of injection speed combine with packing time
Fig.14 The effects of the inputs on the outputs of dimension quality of casting (D_Ave—aveage diameter)
Fig.15 2D interaction effects of the process parameters for average diameter and ovality (a) average diameter response surface of shell temperature combine with alloy temperature (b) average diameter response surface of PA combine with shell temperature (c) average diameter response surface of alloy temperature combine with PA (d) ovality response surface of shell temperature combine with alloy temperature (e) ovality response surface of PA combine with shell temperature (f) ovality response surface of alloy temperature combine with PA
Fig.16 3D interaction effects of the inputs for average diameter (a~c) and ovality of investment casting (d~f)
Fig.17 Effects of the input on the casting yield and safe margin of riser
Fig.18 The measured specimens (a, c) cross-sectional morphologies of the riser by optimal solution (a) and modulus method (c) (b) castings by the two methods
1
Su Y J , Fu H D , Bai Y , et al . Progress in materials genome engineering in China [J]. Acta Metall. Sin., 2020, 56: 1313
Wang H , Xiang X D , Zhang L T . On the Data-driven materials innovation infrastructure [J]. Engineering, 2020, 6: 609
3
Qin Z J , Wang Z , Wang Y Q , et al . Phase prediction of Ni-base superalloys via high-throughput experiments and machine learning [J]. Mater. Res. Lett., 2021, 9: 32
4
Kirchdoerfer T , Ortiz M . Data-driven computational mechanics [J]. Comput. Meth. Appl. Mech. Eng., 2016, 304: 81
5
Sabau A S , Porter W D . Alloy shrinkage factors for the investment casting of 17-4PH stainless steel parts [J]. Metall. Mater. Trans., 2008, 39B: 317
6
Gebelin J C , Jolly M R . Modelling of the investment casting process [J]. J. Mater. Process. Technol., 2003, 135: 291
7
Gayda J . The effect of heat treatment on residual stress and machining distortions in advanced nickel base disk alloys [R]. NASA/TM-2001-210717. Springfield, VA: NASA, 2001: 1
8
Allison J , Backman D , Christodoulou L . Integrated computational materials engineering: A new paradigm for the global materials profession [J]. JOM, 2006, 58(11): 25
9
Chen J Y , Li Q , Xiao C B , et al . The second generation hot corrosion resistant Ni-based single crystal superalloy DD489 and its typical properties [J]. Heat Treat. Met., 2019, 44(6): 65
Liu X J , Chen Y C , Lu Y , et al . Present research situation and prospect of multi-scale design in novel Co-based superalloys: A review [J]. Acta Metall. Sin., 2020, 56: 1
Hui X , Fang H Z , Chen G L , et al . Atomic structure of Zr41.2Ti13.8Cu12.5Ni10Be22.5 bulk metallic glass alloy [J]. Acta Mater., 2009, 57: 376
12
Wang W Y , Li J S , Liu W M , et al . Integrated computational materials engineering for advanced materials: A brief review [J]. Comput. Mater. Sci., 2019, 158: 42
13
Wang W Y , Li P X , Lin D Y , et al . Did code: A bridge connecting the materials genome engineering database with inheritable integrated intelligent manufacturing [J]. Engineering, 2020, 6: 612
14
Wang Y , Sun F , Dong X P , et al . Thermodynamic analysis in the design of several typical nickel-based single-crystal superalloys [J]. Acta Metall. Sin., 2010, 46: 334
Wang J , Sun F , Dong X P , et al . Thermodynamic analysis in the design of several typical nickel-based single-crystal superalloys [J]. Shanghai Nonferrous Met., 2011, 32(2): 49
Zacherl C L , Shang S L , Kim D E , et al . Effects of alloying elements on elastic, stacking fault, and diffusion properties of Fcc Ni from first-principles: Implications for tailoring the creep rate of Ni-Base superalloys [A]. Superalloys 2012 [C]. Warrendale: TMS, 2012: 455
17
Menou E , Ramstein G , Bertrand E , et al . Multi-objective constrained design of nickel-base superalloys using data mining- and thermodynamics-driven genetic algorithms [J]. Modell. Simul. Mater. Sci. Eng., 2016, 24: 055001
18
Wang D H , He B , Liu S M , et al . Dimensional shrinkage prediction based on displacement field in investment casting [J]. Int. J. Adv. Manuf. Technol., 2016, 85: 201
19
Tavakoli R , Davami P . Optimal feeder design in sand casting process by growth method [J]. Int. J. Cast Met. Res., 2007, 20: 288
20
Liu C H , Jin S , Lai X M , et al . Influence of complex structure on the shrinkage of part in investment casting process [J]. Int. J. Adv. Manuf. Technol., 2015, 77: 1191
21
Gebelin J C , Jolly M R , Cendrowicz A M , et al . Simulation of die filling for the wax injection process: Part I. Models for material behavior [J]. Metall. Mater. Trans., 2004, 35B: 755
22
Wang D H , Dong A P , Zhu G L , et al . The propagation and accumulation of dimensional shrinkage for ring-to-ring structure in investment casting [J]. Int. J. Adv. Manuf. Technol, 2018, 96: 623
23
Kumar S , Karunakar D B . Development of wax blend pattern and optimization of injection process parameters by grey-fuzzy logic in investment casting process [J]. Int. J. Metalcast., 2021: https://doi.org/10.1007/s40962-021-00655-y
24
Yu J P , Wang D H , Li D Y , et al . Engineering computing and data-driven for gating system design in investment casting [J]. Int. J. Adv. Manuf. Technol., 2020, 111: 829
25
Chabbi A , Yallese M A , Nouioua M , et al . Modeling and optimization of turning process parameters during the cutting of polymer (POM C) based on RSM, ANN, and DF methods [J]. Int. J. Adv. Manuf. Technol., 2017, 91: 2267
26
Hardin R A , Choi K K , Gaul N J , et al . Reliability based casting process design optimisation [J]. Int. J. Cast Met. Res., 2015, 28: 181
