Risk-Informed Operations and Maintenance Decision Making Using Deep Reinforcement Learning

Ryan Spangler

Risk-Informed Operations and Maintenance Decision Making Using Deep Reinforcement Learning

Ryan Spangler

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

The goal of this research is to improve nuclear operations and maintenance (O&M) decision-making by integrating component reliability, condition monitoring, and deep reinforcement learning to reduce overall life-cycle costs. By using deep reinforcement learning, we can train a neural network to identify the optimal maintenance decision given the current state of the plant. Preliminary studies have shown that an optimized condition-based decision-maker can reduce O&M costs by over 50%.

Original language	American English
Title of host publication	Proceedings of the Annual Conference of the PHM Society 2022
Volume	14
Edition	1
State	Published - Oct 28 2022
Externally published	Yes

INL Publication Number

INL/EXP-22-68457
138309

Access to Document

https://papers.phmsociety.org/index.php/phmconf/article/view/3397

Cite this

@inproceedings{62bb6816da5147099a60cd191531af73,

title = "Risk-Informed Operations and Maintenance Decision Making Using Deep Reinforcement Learning",

abstract = "The goal of this research is to improve nuclear operations and maintenance (O\&M) decision-making by integrating component reliability, condition monitoring, and deep reinforcement learning to reduce overall life-cycle costs. By using deep reinforcement learning, we can train a neural network to identify the optimal maintenance decision given the current state of the plant. Preliminary studies have shown that an optimized condition-based decision-maker can reduce O\&M costs by over 50\%.",

author = "Ryan Spangler",

year = "2022",

month = oct,

day = "28",

language = "American English",

volume = "14",

booktitle = "Proceedings of the Annual Conference of the PHM Society 2022",

edition = "1",

}

TY - GEN

T1 - Risk-Informed Operations and Maintenance Decision Making Using Deep Reinforcement Learning

AU - Spangler, Ryan

PY - 2022/10/28

Y1 - 2022/10/28

N2 - The goal of this research is to improve nuclear operations and maintenance (O&M) decision-making by integrating component reliability, condition monitoring, and deep reinforcement learning to reduce overall life-cycle costs. By using deep reinforcement learning, we can train a neural network to identify the optimal maintenance decision given the current state of the plant. Preliminary studies have shown that an optimized condition-based decision-maker can reduce O&M costs by over 50%.

AB - The goal of this research is to improve nuclear operations and maintenance (O&M) decision-making by integrating component reliability, condition monitoring, and deep reinforcement learning to reduce overall life-cycle costs. By using deep reinforcement learning, we can train a neural network to identify the optimal maintenance decision given the current state of the plant. Preliminary studies have shown that an optimized condition-based decision-maker can reduce O&M costs by over 50%.

M3 - Conference contribution

VL - 14

BT - Proceedings of the Annual Conference of the PHM Society 2022

ER -

Risk-Informed Operations and Maintenance Decision Making Using Deep Reinforcement Learning

Abstract

INL Publication Number

Access to Document

Fingerprint

Cite this