Enhancing refinery heavy oil fractions analytical performance through real-time predicative modeling

Area of Science:

• Petroleum Engineering
• Analytical Chemistry
• Process Control

Background:

• Heavy oil refining requires accurate real-time monitoring of physiochemical properties.
• Existing methods for determining properties like sulfur content, MCR, and metal concentrations can be time-consuming.
• Optimizing heavy oil processing necessitates advanced analytical techniques for immediate feedback.

Purpose of the Study:

• To develop and validate a suite of robust models for real-time physiochemical property determination in heavy oil refinery fractions.
• To integrate a single analyzer employing partial least square regression for enhanced refinery analysis.
• To provide precise, real-time predictions of critical petroleum properties for heavy oil feeds.

Main Methods:

• Development of six partial least square regression equations integrated into a single analyzer.
• Application of real-time analytical techniques within refinery operations.
• Validation of model predictions against American Society for Testing and Materials (ASTM) standards.

Main Results:

• Successful real-time prediction of sulfur content, micro carbon residues (MCR), asphaltene content, heating value, and nickel and vanadium concentrations.
• Models are tailored for heavy oil feeds with API gravity (3°-32°) and sulfur content (2.8-5.5 wt%).
• Demonstrated accuracy and precision of the physiochemical prediction models.

Conclusions:

• The developed models significantly enhance the efficiency of refinery analysis for heavy oil.
• This approach establishes a new standard for real-time monitoring and optimization in heavy oil processing.
• The integrated analyzer streamlines analysis, bridging gaps between different refinery processes.