Modeling of heavy-duty gas turbine based on machine learning algorithms

Wenjiang Huang; Jie Wang; Yingjie Li; Fang Zhang; Haiming Jiang

doi:10.1117/12.3052346

30 December 2024 Modeling of heavy-duty gas turbine based on machine learning algorithms

Wenjiang Huang, Jie Wang, Yingjie Li, Fang Zhang, Haiming Jiang

Author Affiliations +

Proceedings Volume 13394, International Workshop on Automation, Control, and Communication Engineering (IWACCE 2024); 133940L (2024) https://doi.org/10.1117/12.3052346
Event: International Workshop on Automation, Control, and Communication Engineering (IWACCE 2024), 2024, Hohhot, China

Abstract

This paper investigates gas turbine modeling techniques based on machine learning algorithms. Utilizing polynomial regression (PR), K-Nearest Neighbor (KNN), decision trees (DT), and multi-layer perceptrons (MLP), it constructs models from one year's operational data from a power plant's heavy-duty gas turbine. Predictions encompass key parameters such as compressor outlet temperature, compressor outlet pressure, power, and turbine exhaust temperature. Experimental results demonstrate that the MLP algorithm achieves high prediction accuracy for compressor outlet temperature (MAPE: 0.112%), compressor outlet pressure (0.201%), power (0.421%), and turbine outlet temperature (0.070%), highlighting its effectiveness in gas turbine modeling.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Wenjiang Huang, Jie Wang, Yingjie Li, Fang Zhang, and Haiming Jiang "Modeling of heavy-duty gas turbine based on machine learning algorithms", Proc. SPIE 13394, International Workshop on Automation, Control, and Communication Engineering (IWACCE 2024), 133940L (30 December 2024); https://doi.org/10.1117/12.3052346

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE