Fouling Modelling in Crude Oil Preheat Systems

Abstract

Fouling in the pre-heat system for crude oil distillation has become one of the most challenging issues within the refinery industry. For a single crude oil distillation unit, the cost due to fouling can reach magnitudes of millions of dollars per year. Given a fouling model, deposition can be mitigated through the manipulation of heat exchanger tube wall temperatures, wall shear stress and cleaning of heat exchangers. However, the implementation of such strategies requires a mathematical model. Fouling models can be developed from laboratory tests, but such experimental work involves a significant amount of time and the controlled conditions during a test cannot be extrapolated to field processes with confidence. Fouling threshold modelling also presents drawbacks; since each fouling rate model is developed for a specific mechanism, and each parameter within these models can change significantly when the type of crude oil is changed. To overcome these problems, a new methodology for determining fouling models is proposed from on-line data, eliminating the need for laboratory experiments. Heat transfer coefficients coupled with different fouling mechanism models for individual heat exchangers are used to predict thermal and fouling behavior within a heat exchanger network, using reconciled measured data and parametric fitting. The model is able to split the fouling contributions of both shell and tube-side. Also, the reconciled data present no systematic and random errors. Each fitted model can be used for prediction of fouling conditions for operating decisions or optimization of cleaning schedules or retrofit.