PID gain tuning method for oil refining controller based on neural networks

Yoshihiro Abe, Masami Konishi, Jun Imai, Ryuusaku Hasagawa, Masanori Watanabe, Hiroaki Kamijo

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In these years, plant control systems are being highly automated. But the control performances change with the passage of time, so it is necessary to tune them. This is why human experts tune the control system to improve the total plant performances. In this study, PID control system for the oil refining chemical plant process is treated. In the oil refining controller, there are thousands of the control loops in the plant to keep the product quality desired value and to secure the safety of the plant operation. According to the ambiguity of the interference between the control loops, it is difficult to estimate the plant dynamic model accurately. Neuro emulator is employed to model the plant characteristics. Combining neuro emulator and RNN model, auto tuning system of PID control gains has been constructed. Through numerical experiments using actual plant data, the effect of the proposed method was ascertained.

Original languageEnglish
Title of host publicationSecond International Conference on Innovative Computing, Information and Control, ICICIC 2007
PublisherIEEE Computer Society
ISBN (Print)0769528821, 9780769528823
DOIs
Publication statusPublished - Jan 1 2007
Event2nd International Conference on Innovative Computing, Information and Control, ICICIC 2007 - Kumamoto, Japan
Duration: Sept 5 2007Sept 7 2007

Publication series

NameSecond International Conference on Innovative Computing, Information and Control, ICICIC 2007

Other

Other2nd International Conference on Innovative Computing, Information and Control, ICICIC 2007
Country/TerritoryJapan
CityKumamoto
Period9/5/079/7/07

ASJC Scopus subject areas

  • Computer Science(all)
  • Mechanical Engineering

Fingerprint

Dive into the research topics of 'PID gain tuning method for oil refining controller based on neural networks'. Together they form a unique fingerprint.

Cite this