(61b) Modeling Crude Units with Rigorously Calculated Preheat Trains

A typical crude unit consists of a distillation tower operating at low positive pressure (the atmospheric column) and one that operates at very deep vacuum (the vacuum tower). Simulation programs are ideally suited to solve distillation problems because of their ability to handle vapor-liquid equilibrium and enthalpy calculations. The distillation columns work by feeding crude oil at a high temperature. The crude oil is separated into various products by removing heat from the vaporized crude. Heat is physically removed from the crude by reflux generated externally by condensing overhead vapor and internally by liquid pump around circulation.

Efficient crude oil fractionation results from recovering as much of the heat from the crude and vacuum tower and putting it back into the crude. This is accomplished in a complex of heat exchangers known as the crude preheat train. Typically, these calculations were handled by assuming a heat transfer coefficient for each exchanger in the train and calculating the outlet temperatures. The heat transfer coefficient is verified by downloading the operating parameters from the simulation model into a heat exchanger program. Getting results from both programs to agree creates an iterative calculation.

In a recent crude unit expansion project the feed blend was shifted to lighter crudes. The preheat train calculations were critical to calculating the inlet temperature and duty of the crude unit furnaces. The project team realized early on that only operating scenarios that were within the firing limit of the existing furnaces would be economical. The preheat train consisted of about 35 heat exchange services. A conventional, iterative approach of calculating heat transfer coefficients in a separate program would be inefficient and time consuming. This presentation will report on the results of integrating an Exchanger Design and Rating Program within the simulation so that rigorous exchanger calculations could be done rapidly and consistently.