Intelligent steam turbine start-up control based on deep reinforcement learning

Guangya Zhu, Ding Guo, JinXing Li, Yonghui Xie and Di Zhang

Energy, 2025, vol. 320, issue C

Abstract: The requirement for frequent start-ups and shutdowns is prevalent in turbo-generator units to accommodate fluctuating loads during flexible operations. These cause drastic changes in temperature and stress, leading to instantaneous structural deformations. Hence, research on intelligent start-up control is essential for ensuring safety. In this work, a rotor stress field reconstruction model based on a deep convolutional neural network was first designed. The accuracy of predicting the stress distribution in the critical area reaches 99.7 %. The time cost of the trained neural network model is 0.11s in a single case, shortened by 99.8 % with comparison to finite element analysis. Then, a Twin Delayed Deep Deterministic Policy Gradient-based Main Steam Temperature Controller for the Rotor Start-up was proposed and developed.

Keywords: Steam turbine rotor; Start-up control; Stress field reconstruction; Deep reinforcement learning (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://www.sciencedirect.com/science/article/pii/S0360544225009776
Full text for ScienceDirect subscribers only

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:eee:energy:v:320:y:2025:i:c:s0360544225009776

DOI: 10.1016/j.energy.2025.135335

Access Statistics for this article

Energy is currently edited by Henrik Lund and Mark J. Kaiser

More articles in Energy from Elsevier
Bibliographic data for series maintained by Catherine Liu ().