Evaluation of Prediction Tool for Sour Water Corrosion Quantification and Management in Refineries

The Sour Water Joint Industry Project (JIP) conducted by Honeywell led to the development of new data and insights related to corrosion in H₂S-dominated Ammonium bi-sulfide sour water systems, commonly found in a number of refinery processes.  The JIP provided a new methodology and software prediction tool targeted towards predicting and quantifying corrosion in hydro-processing units, specifically reactor effluent air cooler (REAC) equipment and associated piping.  Based on comprehenive laboratory data, multiphase flow-modeling and numerical data interpolation, the Predict-SW Prediction System facilitates corrosion prediction, failure prevention, optimized material selection and safe operational practice as a basis for compelling cost saving and improved maintenance planning in refinery operations.
This paper evaluates the efficacy of the JIP-based Prediction System when applied to real plant evaluation situations in a variety of refinery conditions.  The paper also details appropriate methods of utilizing the data and functionality within the tool to ensure optimized prediction accuracy.  Case studies identifying pitfalls to avoid while using the tool are also provided along side comparison of predicted corrosion rates with measured rates from inspection.  The case stuides demonstrate the importance of rigorous flow modeling and need for accurate NH₄HS concentration and environmental parametric data variables in assessment for corrosiveness of alkaline sour water systems.