A Design Optimization of Organic Rankine Cycle Turbine Blades with Radial Basis Neural Network-Reference-Cited by-同舟云学术

A Design Optimization of Organic Rankine Cycle Turbine Blades with Radial Basis Neural Network

Published:2023-12-20 Issue:1 Volume:17 Page:26
ISSN:1996-1073
Container-title:Energies
language:en
Short-container-title:Energies

Author:

Seo Jong-Beom¹^ORCID,Lee Hosaeng¹,Han Sang-Jo²

Affiliation:

1. Korea Research Institute of Ships & Ocean Engineering, Daejeon 34103, Republic of Korea

2. Department of Mechanical and Automotive Engineering, Seoul National University of Science and Technology, Seoul 01811, Republic of Korea

Abstract

In the present study, a 100 kW organic Rankine cycle is suggested to recover heat energy from commercial ships. A radial-type turbine is employed with R1233zd(E) and back-to-back layout. To improve the performance of an organic Rankine power system, the efficiency of the turbine is significant. With the conventional approach, the optimization of a turbine requires a considerable amount of time and involves substantial costs. By combining design of experiments, an artificial neural network, and Latin hypercube sampling, it becomes possible to reduce costs and achieve rapid optimization. A radial basis neural network with machine learning technique, known for its advantages of being fast and easily applicable, has been implemented. Using such an approach, an increase in efficiency greater than 1% was achieved with minimal design changes at the first and second turbines.

Funder

Korea Institute of Marine Science & Technology Promotion

Publisher

MDPI AG

Subject

Energy (miscellaneous),Energy Engineering and Power Technology,Renewable Energy, Sustainability and the Environment,Electrical and Electronic Engineering,Control and Optimization,Engineering (miscellaneous),Building and Construction

Link

https://www.mdpi.com/1996-1073/17/1/26/pdf

Reference24 articles.

1. Greenman, R.M. (1998). 2-D High-Lift Aerodynamic Optimization Using Neural Networks, NASA. NASA TM-1998-112233.

2. Rai, M.M., and Madavan, N.K. (2000, January 10–13). Improving the unsteady aerodynamic performance of transonic turbines using neural networks. Proceedings of the 38th AIAA Aerospace Sciences Meeting and Exhibit, Reno, Nevada. AIAA 2000-0169.

3. Norgaard, M., Jorgenson, C.C., and Ross, J.C. (1997). Neural Network Prediction of New Aircraft Design Coefficients, NASA. NASA TM-112197.

4. Vaidyanathan, R., Papila, N., Shyy, W., Tucker, P.K., Griffin, L.W., Fitz-Coy, N., and Haftka, R.T. (2000, January 6–8). Neural network-based and response surface-based optimization strategies for rocket engine component design. Proceedings of the Eighth AIAA/USAF/NASA/ISSMO Symposium on Multi-Disciplinary Analysis and Optimization, Long Beach, CA, USA.

5. Papila, N., Shyy, W., Fitz-Coy, N., and Haftka, R.T. (July, January 28). Assessment of neural net and polynomial-based techniques for aerodynamic applications. Proceedings of the 17th Applied Aerodynamics Conference, Norfolk, VA, USA. AIAA 99-3167.