Browse Prior Art Database

LNG Plant Advisory System -- A Knowledge-Based Approach

IP.com Disclosure Number: IPCOM000218653D
Publication Date: 2012-Jun-06

Publishing Venue

The IP.com Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 21% of the total text.

Page 01 of 10

Session and Paper Number: 42C

LNG Plant Advisory System - A Knowledge-Based Approach

Felix F. de la Vega

Jatin T. Shah

The M.W. Kellogg Company

P. 0. Box 4557

Houston, Texas 77210-4557

"Preparedfor presentation at

AlChE 1993 Spring National Meeting 31 March 1993

Applications of Artificial Intelligence I: Presentations"

"Copyright The M. W. Kellogg Company"

"Date: March 1993"

" UNPUBLISHED"

"AIChE shall not be responsible for statements or opinions contained Inpapers or printed in itspublications."


Page 02 of 10

ABSTRACT

    A real-time knowledge-based system known as the LNG Plant Advisory System (LNGPASTM) has been developed at The M. W. Kellogg Company to guide and assist LNG (liquefied natural gas) plant operators in control and operation of the LNG storage facilities.

    LNGPASTM combines knowledge-based systems methodologies and conventional programming techniques to provide a unique approach in assisting plant operators in simplifying the logical decision process of monitoring and controlling the operation of the LNG storage facility. The system incorporates expert process knowledge, operating procedures, real-time sensor data and other information to providevital, timely, on-line information, diagnostics, and advice.

   The system is designed to receive on-line process data from existing sensors, analyze them, detect abnormalities, and advise LNG plant operators on any corrective actions to minimize or prevent potentially dangerous situations, such as "rollover". The system also assists operators in analyzing events which may have led to a potential rollover situation and/or predict the effects of any corrective actions to prevent it.

   LNGPASTM was developed using the M/Power real-time process management software environment by Mitech Corporation; it runs on DEC VAX computers. The system also uses Kellogg's specialized process simulator to perform various analyses and heat and material balance calculations.

INTRODUCTION

    In today's control room environment, plant operators are faced with a significant challenge in maintaining control and operation of process plants. They are required to analyze critical process information in real time, diagnose potential plant upset conditions, take appropriate corrective actions to improve product quality and plant efficiency, prevent occurrences of potentially hazardous situations and/or costly

Page 2

8



Page 03 of 10

shutdowns, etc. They use their operating experience, knowledge, and judgment to make various decisions.

    Real-time knowledge-based systems are ideally suited for the development of advisory or diagnostic systems for the operation and control of a processing facility. These systems provide a means to capture "best" plant operating philosophy in the form of rule-based knowledge. They use this knowledge to simplify the task of analyzing large amounts of real-time data, identify trends, and diagnose abnormalities. In addition, they also provide advice on any potential process upset conditi...