Abstract
In composition and behaviour, bitumen and heavy oil are more complex than conventional oil; hence they pose greater challenges for reservoir and process simulations. This notwithstanding, one practical approach for constructing efficient fluid-property models is the lumping technique, entailing the use of pseudo-components (PC). However, this is not so straight-forward, especially for these unconventional hydrocarbons whose compositions are typically available as chemical aggregates (saturates, aromatics, resins and asphaltenes) rather than the common and more convenient pure components.
This paper highlights key issues affecting construction of PC’s in general, and limits them to heavy oil and bitumen. It argues that the number and characters of PC’s should be influenced by prospective method of development, processing as well as the nature of fluids the crude would contact. Hence, process-based PC schemes are proposed for the Nigerian unconventional crudes. As an example, the number of components required for a thermal process using asphaltene-precipitating injectant (e.g. CO2) and that of a process based on steam, a non-precipitant, is shown to differ. Similar distinctions are indicated between thermal and solvent-based processes.
For some reference conditions, it is demonstrated that the Nigerian heavy crude can, in its simplest form, be reasonably represented as a single-component and up to five components where it is likely to undergo more complex processes or when greater details are necessary. With minor modifications, the proposed schemes should be adaptable to heavy crudes elsewhere.
