Natural gas valve modeling study based on long and short-term memory neural network

Tao Cui; Hongrui Wang; Yao Liu

doi:10.1117/12.3046876

27 September 2024 Natural gas valve modeling study based on long and short-term memory neural network

Tao Cui, Hongrui Wang, Yao Liu

Proceedings Volume 13261, Tenth International Conference on Mechanical Engineering, Materials, and Automation Technology (MMEAT 2024); 132614Z (2024) https://doi.org/10.1117/12.3046876
Event: 10th International Conference on Mechanical Engineering, Materials, and Automation Technology (MMEAT 2024), 2024, Wuhan, China

Abstract

The paper proposed a modeling approach for a natural gas compressor valve using a long and short-term memory neural network. The main objective is to simulate the dynamic response characteristics of the valve and offer support for optimizing the anti-surge control strategy. Using the Damadics Actuator Benchmark Library generated samples to train the model. The results indicate that the model's performance gradually improves on the training set. Despite some overfitting phenomena occurring when the excitation signals change, the trained network demonstrates a good fit with the actual data. This study presents an effective modeling and identification method for compressor anti-surge control, which serves as a valuable reference for ensuring the safe and stable operation of the system.

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

Citation Download Citation

Tao Cui, Hongrui Wang, and Yao Liu "Natural gas valve modeling study based on long and short-term memory neural network", Proc. SPIE 13261, Tenth International Conference on Mechanical Engineering, Materials, and Automation Technology (MMEAT 2024), 132614Z (27 September 2024); https://doi.org/10.1117/12.3046876

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

Includes PDF, HTML & Video, when available