Research Papers

Machine Learning Enabled Adaptive Optimization of a Transonic Compressor Rotor With Precompression

[+] Author and Article Information
Michael Joly

Thermal and Fluid Sciences,
United Technologies Research Center,
East Hartford, CT 06108
e-mail: jolymm@utrc.utc.com

Soumalya Sarkar

Autonomous and Intelligent Systems,
United Technologies Research Center,
East Hartford, CT 06108
e-mail: sarkars@utrc.utc.com

Dhagash Mehta

Autonomous and Intelligent Systems,
United Technologies Research Center,
East Hartford, CT 06108
e-mail: mehtadb@utrc.utc.com

1Corresponding author.

Contributed by the International Gas Turbine Institute (IGTI) of ASME for publication in the JOURNAL OF TURBOMACHINERY. Manuscript received September 21, 2018; final manuscript received October 17, 2018; published online January 25, 2019. Editor: Kenneth Hall.

J. Turbomach 141(5), 051011 (Jan 25, 2019) (9 pages) Paper No: TURBO-18-1262; doi: 10.1115/1.4041808 History: Received September 21, 2018; Revised October 17, 2018

In aerodynamic design, accurate and robust surrogate models are important to accelerate computationally expensive computational fluid dynamics (CFD)-based optimization. In this paper, a machine learning framework is presented to speed-up the design optimization of a highly loaded transonic compressor rotor. The approach is threefold: (1) dynamic selection and self-tuning among several surrogate models; (2) classification to anticipate failure of the performance evaluation; and (3) adaptive selection of new candidates to perform CFD evaluation for updating the surrogate, which facilitates design space exploration and reduces surrogate uncertainty. The framework is demonstrated with a multipoint optimization of the transonic NASA rotor 37, yielding increased compressor efficiency in less than 48 h on 100 central processing unit cores. The optimized rotor geometry features precompression that relocates and attenuates the shock, without the stability penalty or undesired reacceleration usually observed in the literature.

Copyright © 2019 by ASME
Your Session has timed out. Please sign back in to continue.


Denton, J. , 1997, “ Lessons From Rotor 37,” J. Therm. Sci., 6(1), pp. 1–13. [CrossRef]
Oyama, A. , Liou, M.-S. , and Obayashi, S. , 2004, “ Transonic Axial-Flow Blade Optimization: Evolutionary Algorithms/Three-Dimensional Navier-Stokes Solver,” J. Propul. Power, 20(4), pp. 612–619. [CrossRef]
Reid, L. , and Moore, R. D. , 1978, “ Design and Overall Performance of Four Highly Loaded, High Speed Inlet Stages for an Advanced High-Pressure-Ratio Core Compressor,” Report No. NASA-TP-1337, E-9302.
John, A. , Shahpar, S. , and Qin, N. , 2017, “ Novel Compressor Blade Shaping Through a Free-Form Method,” ASME J. Turbomach., 139(8), p. 081002. [CrossRef]
Luo, J. , Zhou, C. , and Liu, F. , 2014, “ Multipoint Design Optimization of a Transonic Compressor Blade by Using an Adjoint Method,” ASME J. Turbomach., 136(5), p. 051005. [CrossRef]
Chima, R. V. , 2009, “ Swift Code Assessment for Two Similar Transonic Compressors,” AIAA Paper No. AIAA-2009-1058.
Wang, G. G. , and Shan, S. , 2007, “ Review of Metamodeling Techniques in Support of Engineering Design Optimization,” ASME J. Mech. Des., 129(4), pp. 370–380. [CrossRef]
Abraham, A. , and Jain, L. , 2005, Evolutionary Multiobjective Optimization, Springer, London, pp. 1–6.
Jin, Y. , and Sendhoff, B. , 2009, “ A Systems Approach to Evolutionary Multiobjective Structural Optimization and Beyond,” IEEE Comput. Intell. Mag., 4(3), pp. 62–76.
Jin, Y. , 2011, “ Surrogate-Assisted Evolutionary Computation: Recent Advances and Future Challenges,” Swarm Evol. Comput., 1(2), pp. 61–70. [CrossRef]
Doherty, J. , and Christensen, S. , 2011, “ Use of Paired Simple and Complex Models to Reduce Predictive Bias and Quantify Uncertainty,” Water Resour. Res., 47(12), p. 21.
Young, P. C. , and Ratto, M. , 2011, “ Statistical Emulation of Large Linear Dynamic Models,” Technometrics, 53(1), pp. 29–43. [CrossRef]
Vuik, C. , Segal, A. , and Meijerink, J. , 1999, “ An Efficient Preconditioned cg Method for the Solution of a Class of Layered Problems With Extreme Contrasts in the Coefficients,” J. Comput. Phys., 152(1), pp. 385–403. [CrossRef]
Díaz-Manríquez, A. , Toscano, G. , Barron-Zambrano, J. H. , and Tello-Leal, E. , 2016, “ A Review of Surrogate Assisted Multiobjective Evolutionary Algorithms,” Comput. Intell. Neurosci., 2016, p. 14.
Myers, R. H. , Montgomery, D. C. , and Anderson-Cook, C. M. , 2016, Response Surface Methodology: Process and Product Optimization Using Designed Experiments, Wiley, New York.
Hardy, R. L. , 1971, “ Multiquadric Equations of Topography and Other Irregular Surfaces,” J. Geophys. Res., 76(8), pp. 1905–1915. [CrossRef]
Rosales-Pérez, A. , Coello, C. A. C. , Gonzalez, J. A. , Reyes-Garcia, C. A. , and Escalante, H. J. , 2013, “ A Hybrid Surrogate-Based Approach for Evolutionary Multi-Objective Optimization,” IEEE Congress on Evolutionary Computation (CEC), Cancun, Mexico, June 20–23, pp. 2548–2555.
Schalkoff, R. J. , 1997, Artificial Neural Networks, Vol. 1, McGraw-Hill, New York.
Verstraete, T. , Coletti, F. , Bulle, J. , Vanderwielen, T. , and Arts, T. , 2013, “ Optimization of a u-Bend for Minimal Pressure Loss in Internal Cooling Channels—Part I: Numerical Method,” ASME J. Turbomach., 135(5), p. 051015. [CrossRef]
Goel, T. , Haftka, R. T. , Shyy, W. , and Queipo, N. V. , 2007, “ Ensemble of Surrogates,” Struct. Multidiscip. Optim., 33(3), pp. 199–216. [CrossRef]
Karakasis, M. K. , and Giannakoglou, K. C. , 2006, “ On the Use of Metamodel-Assisted, Multi-Objective Evolutionary Algorithms,” Eng. Optim., 38(8), pp. 941–957. [CrossRef]
Storn, R. , and Price, K. , 1997, “ Differential Evolution—A Simple and Efficient Heuristic for Global Optimization Over Continuous Spaces,” J. Global Optim., 11(4), pp. 341–359. [CrossRef]
Madavan, N. K. , 2002, “ Multiobjective Optimization Using a Pareto Differential Evolution Approach,” IEEE Congress on Evolutionary Computation (CEC'02), Honolulu, HI, May 12–17, pp. 1145–1150.
Deb, K. , Agrawal, S. , Pratap, A. , and Meyarivan, T. , 2000, “ A Fast Elitist Non-Dominated Sorting Genetic Algorithm for Multi-Objective Optimization: NSGA-II,” International Conference on Parallel Problem Solving From Nature, pp. 849–858.
Konak, A. , Coit, D. W. , and Smith, A. E. , 2006, “ Multi-Objective Optimization Using Genetic Algorithms: A Tutorial,” Reliab. Eng. Syst. Saf., 91(9), pp. 992–1007. [CrossRef]
Breiman, L. , 2001, “ Random Forests,” Mach. Learn., 45(1), pp. 5–32. [CrossRef]
Liaw, A. , and Wiener, M. , 2002, “ Classification and Regression by RandomForest,” R News, 2(3), pp. 18–22. https://www.r-project.org/doc/Rnews/Rnews_2002-3.pdf
Kohavi, R. , 1995, “ A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection,” 14th International Joint Conference on Artificial Intelligence (IJCAI), Montreal, Quebec, Canada, Aug. 20–25, pp. 1137–1143. http://ai.stanford.edu/~ronnyk/accEst.pdf
Feurer, M. , Springenberg, J. T. , and Hutter, F. , 2015, “ Initializing Bayesian Hyperparameter Optimization Via Meta-Learning,” 29th AAAI Conference on Artificial Intelligence (AAAI'15), Austin, TX, Jan. 25–30, pp. 1128–1135.
Jones, D. R. , Schonlau, M. , and Welch, W. J. , 1998, “ Efficient Global Optimization of Expensive Black-Box Functions,” J. Global Optim., 13(4)Dec, pp. 455–492. [CrossRef]
Brochu, E. , Cora, V. M. , and De Freitas, N. , 2010, “ A Tutorial on Bayesian Optimization of Expensive Cost Functions, With Application to Active User Modeling and Hierarchical Reinforcement Learning,” e-print arXiv:1012.2599. https://arxiv.org/abs/1012.2599
Saar-Tsechansky, M. , and Provost, F. , 2007, “ Handling Missing Values When Applying Classification Models,” J. Mach. Learn. Res., 8, pp. 1623–1657. http://jmlr.csail.mit.edu/papers/volume8/saar-tsechansky07a/saar-tsechansky07a.pdf


Grahic Jump Location
Fig. 1

Performance of NASA rotor 37, including CFD validation, grid independency study, and baseline for optimization

Grahic Jump Location
Fig. 2

Surrogate-assisted optimization

Grahic Jump Location
Fig. 3

Schematic of ML4SAO framework

Grahic Jump Location
Fig. 4

Receiver-operating characteristic curves of the response related to the prediction of efficiency on DOE of 150 (a) and 300 (b) candidates intended for rotor 37 optimization

Grahic Jump Location
Fig. 5

Surrogate-assisted optimization on Kursawe function ((a) mean and variance of distance over ten optimizations; (b) mean and variance of spread over ten optimizations)

Grahic Jump Location
Fig. 6

Surrogate-assisted optimization results on Kursawe function with most representative case of mean over ten optimization with different surrogate models ((a) Matern kernel; (b) RBF kernel GP; (c) ML4SAO); (d) Best Pareto front with 100 generations on ML4SAO)

Grahic Jump Location
Fig. 7

History of single-point optimization

Grahic Jump Location
Fig. 8

Performance of baseline and optima of single-point and multipoint optimizations

Grahic Jump Location
Fig. 9

History of multipoint optimization

Grahic Jump Location
Fig. 10

History of surrogate models selection

Grahic Jump Location
Fig. 11

Turbulent kinetic energy downstream of the rotor: baseline (a) and optimized (b)

Grahic Jump Location
Fig. 12

Isentropic Mach number loading at 70% span

Grahic Jump Location
Fig. 13

Relative Mach number filed at 70% span: baseline (a) and optimized (b)



Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In