Refinery operators face increasingly complex challenges in managing integrity of process units and assets – driven by the goal to achieve operational excellence and maximize asset performance while minimizing costs and maintaining the highest safety and environmental standards. Such challenges stem from multiple operational factors, including changing process feed, driven by opportunity crudes and throughput, need to work with more complex process dynamics and difficulties associated with correlating potential corrosion damage to process parameters across different refinery units in a real-time framework.

A novel software framework incorporating intelligent predictive analytics linked to process historian has been created to provide refinery operators with the ability to see corrosion damage in real time. Built on decades of intense experimental corrosion data and predictive model development through Joint Industry Programs (JIP), this framework offers first-of-its-kind insights to proactively manage unit operations, optimize throughput management and derive enhanced mechanical integrity through real-time linkages to process data. The system also enables easy implementation of industry-best practices through support for digitized API RP584 (Integrity Operating Windows) and API RP970 (Corrosion Control Documents).1, 2 

This paper provides an overview of the predictive analytics technology foundational to the software and describes the implementation of software based corrosion prediction framework at multiple refinery sites as well as the benefits of pre-emptive, proactive corrosion monitoring through a real time ‘software sensor’ framework. This prediction framework leverages information already existing within the DCS and process historian infrastructure and provides the reliability group with a real-time dashboard presenting corrosion rates, piping circuit wall thicknesses and inspection planning information. Such an approach has quantified profitability benefits stemming from flexibility in processing opportunity crude slates while enhancing real time mechanical integrity and reliability management of critical refinery units.

Subject
Sensors, Corrosion rate, Predicted corrosion rate, Corrosion prediction, Inspection, Refineries, Integrity operating windows, Experimental data, Corrosion modeling, Corrosion data, Integrity management, Corrosion monitoring, Types of corrosion
© 2019 Association for Materials Protection and Performance (AMPP). All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording, or otherwise) without the prior written permission of AMPP. Positions and opinions advanced in this work are those of the author(s) and not necessarily those of AMPP. Responsibility for the content of the work lies solely with the author(s).
2019
Association for Materials Protection and Performance (AMPP)
You do not currently have access to this content.