Improving Refinery Reliability and Profitability Using Connected Data

The transition from board-mounted, pneumatic refinery controls to electronic DCS-centered control and data historian systems that started to become popular in the 1970’s has helped refinery operations improve production, reliability, and product yield optimization. However, still today much of the data originating from the many third-party service companies that support refinery operations (for evaluation of specialty chemical programs, vibration monitoring, corrosion and inspection readings, etc) remains locked away in proprietary systems that are shared typically only as periodic snapshots of performance and largely separate from the refinery unit operating data – collectively referred to as “dark data”. Over the last several years, as part of the movement to create an “industrial internet of things” (also called the 4^th industrial revolution), new sensors, controllers, and data handling infrastructures have been developed to connect this once dark data with the traditional refinery operating data. This connection allows more rapid, more transparent, and remote evaluation of system performance, including automated application of proprietary analytic algorithms, that was impossible to do previously – thus providing significant opportunities to better optimize reliability and profitability for these once isolated systems.

This paper will show how connecting this dark data can be accomplished, how this changes the method frequency with which performance can be measured, and how more dynamic corrective action plans are developed and implemented. Case study examples will be included to highlight improvements in operations, reliability, and profitability to refinery facilities.