Recursive Subspace Least Squares Estimation for Data-Driven Model Predictive Control and Its Application to Aeroengines

Model predictive control (MPC) is subject to lim-itations, including stringent model accuracy requirements and time-consuming modeling, which impede its broader application. To address these challenges, this article proposes a predic-tive controller based on Data-Driven Recursive Subspace Least Squa...

Full description

Saved in:
Bibliographic Details
Published inIEEE International Conference on Industrial Informatics (INDIN) pp. 1 - 7
Main Authors Wang, Na, Wen, Si-Xin, Wu, Yuhu
Format Conference Proceeding
LanguageEnglish
Published IEEE 18.08.2024
Subjects
Online AccessGet full text
ISSN2378-363X
DOI10.1109/INDIN58382.2024.10774346

Cover

More Information
Summary:Model predictive control (MPC) is subject to lim-itations, including stringent model accuracy requirements and time-consuming modeling, which impede its broader application. To address these challenges, this article proposes a predic-tive controller based on Data-Driven Recursive Subspace Least Squares (DD-RSPC), obviating the need for any prior model identification. Unlike traditional MPC, DD-RSPC dynamically updates the prediction model using real-time measured input-output data, thereby significantly enhancing system adaptability. Furthermore, this article introduces a rigorous stability theorem that employs the Lyapunov function to ensure the stability of the DD-RSPC approach. Finally, the effectiveness of DD-RSPC is validated through aeroengine simulation experiments, effectively substantiating its adaptability, stability, and practical applicability.
ISSN:2378-363X
DOI:10.1109/INDIN58382.2024.10774346