All Issue

2026 Vol.45, Issue 1

Research Article

Optimization of centrifugal fan for clothes dryer using transfer learningbased multi fidelity modeling

전이 학습 기반 다중충실도 모델을 이용한 의류 건조기용 다익 원심팬의 최적 설계

Munseong Hwang1
,
Cheolung Cheong1*
,
Sungdae Cho2
,
Jinho Choi2
황 문성1
,
정 철웅1*
,
조 성대2
,
최 진호2
1부산대학교 기계공학부
2LG 전자
*Corresponding Author
31 January 2026. pp. 1-12
PDF XML Citation
Abstract

This study proposes a deep learning-based multi-fidelity optimization design process to improve the aerodynamic performance of a clothes dryer fan system. The optimization targets are the multi-blade centrifugal fan and scroll, which have a dominant influence on drying performance. The design space was expanded by adding the scroll cutoff angle and operating pressure to the impeller inlet and outlet angles. To minimize data generation costs within this expanded design space, transfer learning was employed. First, flow results obtained through 2D Computational Fluid Dynamics (CFD) simulations were used to pre-train the model as low-fidelity data. Subsequently, transfer learning was performed on datasets containing the new variables. Then, an optimal Deep Neural Network (DNN) surrogate model was constructed using Automated Machine Learning (Auto-ML). The results indicate that the proposed model achieved a 35 % reduction in the required training data while maintaining prediction accuracy comparable to conventional multi-fidelity models. Validation using the derived optimal design parameters confirmed that the flow rate improved by approximately 24 % compared to the baseline model. Furthermore, verification via 3D CFD revealed a prediction error of approximately 1.4 %. In conclusion, this study demonstrates that the proposed design process is effective for the aerodynamic optimization of fan systems while reducing data generation costs.

Keywords
Multi-blade centrifugal fan
Transfer learning
Multi-fidelity
Surrogate model
Optimization design

본 연구는 의류 건조기 팬 시스템의 공력 성능개선을 위해 딥러닝 기반 다중충실도 최적화 설계 과정을 제안한다. 최적화 대상은 건조 성능에 지배적인 영향을 미치는 다익 원심 팬과 스크롤이며, 임펠러 입·출구각에 스크롤 컷오프 각도과 운전 압력을 추가하여 설계 공간을 확장하였다. 본 연구는 전이 학습 기법을 도입하여 확장된 설계 공간에서의 데이터 구축 비용을 최소화하였다. 먼저, 2차원 전산유체역학(Computational Fluid Dynamics, CFD)기반의 전산 해석을 통해 유동 결과를 확보하였으며, 이러한 데이터를 저충실도 데이터로 사전 학습시켰다. 이후, 새로운 변수가 포함된 데이터에 대해 전이학습을 수행하였다. 이후, Automated Machine Learning(Auto-ML)을 통해 최적의 심층신경망(Deep Neural Network, DNN) 대리모델을 구축하였다. 연구 결과로서 제안된 모델은 기존의 다중충실도 모델 대비 유사한 예측 정확도를 유지하면서도 필요 학습 데이터 수를 약 35 % 절감하는 효율성을 보였다. 도출된 설계 인자의 최적값을 적용한 검증 결과에서는 기준 모델 대비 유량이 약 24 % 향상됨을 확인하였다. 3차원 CFD로 검증한 결과, 예측 모델의 오차는 약 1.4 %로 확인되었다. 결론적으로, 본 연구에서 제안한 설계 과정은 학습 데이터 구축 비용을 절감하면서도 팬 시스템의 공력 최적화에 효과적임을 입증하였다.

키워드
다익 원심팬
전이 학습
다중충실도
대리모델
최적설계
References
1

J. Choi, S.-Y. Ryu, C. Cheong, M.-K. Kim, and K. Lee, “Blade shape optimization of centrifugal fan for improving performance and reducing aerodynamic noise of clothes dryer” (in Korean), J. Acoust. Soc. Kr. 38, 321-327 (2019).

2

S. Cho, C. Cheong, J. Choi, and W. Choi, “Virtual product development of circulation centrifugal fan for clothes dryer using two-dimensional computational fluid dynamics techniques” (in Korean), J. Acoust. Soc. Kr. 44, 1-13 (2025).

3

S. Y. Han and J. S. Maeng, “Shape optimization of cut-off in a multi-blade fan/scroll system using neural network,” Int. J. Heat Mass Transf. 46, 2833-2839 (2003).

10.1016/S0017-9310(03)00066-8
4

J.-S. Kim, U.-C. Jeong, D.-W. Kim, S.-Y. Han, and J.-E. Oh, “Optimization of sirocco fan blade to reduce noise of air purifier using a metamodel and evolutionary algorithm,” Appl. Acoust. 89, 254-266 (2015).

10.1016/j.apacoust.2014.10.005
5

S. Zhou, K. Yang, W. Zhang, K. Zhang, C. Wang, and W. jin, “Optimization of multi-blade centrifugal fan blade design for ventilation and air-conditioning system based on disturbance CST function,” Appl. Sci. 11, 7784 (2021).

10.3390/app11177784
6

Z.-F. Liu, Z.-D. Wang, S.-H. Du, H. Yang, Y.-K. Wei, and W. Zhang, “Orthogonal optimization design of a Sirocco fan and numerical analysis on the internal flow characteristics,” Proc. Inst. Mech. Eng. A, 238, 90-110 (2024).

10.1177/09576509231195120
7

W. Choi, J. Choi, and C. Cheong, “Flow noise prediction test and numerical validation for clothes dryer fan systems considering the effect of heat exchangers” (in Korean), J. Acoust. Soc. Kr. 44, 303-311 (2025).

8

ANSI/AMCA Standard 210-07, Laboratory Methods of Testing Fans for Aerodynamic Performance Rating, ANSI/AMCA, 2007.

9

R. Storn and K. Price, “Differential evolution – A simple and efficient heuristic for global optimization over continuous spaces,” J. Glob. Optim. 11, 341-359 (1997).

10.1023/A:1008202821328
10

F. Hutter, L. Kotthoff, and J. Vanschoren, Automated Machine Learning: Methods, Systems, Challenges (Springer International Publishing, Cham, 2019), pp.1-219.

10.1007/978-3-030-05318-5
11

J. Bergstra, R. Bardenet, Y. Bengio, and B. Kégl, “Algorithms for hyper-parameter optimization,” Pocr. Adv. Neural Inf. Process. Syst. 24, 2546-2554 (2011).

12

T. Akiba, S. Sano, T. Yanase, T. Ohta, and M. Koyama, “Optuna: A next-generation hyperparameter optimization framework,” Proc. 25th ACM SIGKDD Int. Conf. Knowl. Discov. Data Min, 2623-2631 (2019).

10.1145/3292500.3330701
Information
  • Publisher :The Acoustical Society of Korea
  • Publisher(Ko) :한국음향학회
  • Journal Title :The Journal of the Acoustical Society of Korea
  • Journal Title(Ko) :한국음향학회지
  • Volume : 45
  • No :1
  • Pages :1-12
  • Received Date : 2025-12-23
  • Revised Date : 2026-01-20
  • Accepted Date : 2026-01-23
  • DOI :https://doi.org/10.7776/ASK.2026.45.1.001
Journal Informaiton The Journal of the Acoustical Society of Korea The Journal of the Acoustical Society of Korea
E-Book
  • scopus_JASK
  • ESCI_JASK
  • Google scholar_JASK
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • ccl
Journal Informaiton Journal Informaiton - close