Nov-2016

Simulating crude units with preflash

Process simulation offers a cost effective way of testing ideas to improve yield and energy performance.

JUAN GOMEZ-PRADO, RALPH GOODRICH and DON HOPPENS
KBC Process Technology

Article Summary

Acrude distillation unit (CDU) preflash drum is a vessel designed to flash a portion of the lighter crude components and water upstream of the main column charge heater. The wide-cut flashed vapour from the preflash drum is typically routed to the CDU main column flash zone to avoid contaminating any side stream products with heavy components that could be carried over with the flashed vapour. The liquid from the preflash drum is normally routed through additional preheat exchange before entering the charge heater.

Meanwhile, a CDU preflash tower is designed to have fractionating trays with overhead condensing capability and reflux to produce a more defined overhead product stream than a preflash drum. Typically, the preflash tower overhead product consists of naphtha boiling range material that can be routed directly to further naphtha processing. Preflash tower arrangements can also include multiple feeds, side cut products, a reboiler, and/or stripping steam. Similar to the preflash drum, the bottoms from the preflash tower are usually routed through additional preheat prior to entering the charge heater.

Preflash facilities are typically installed in CDUs to enable higher heater inlet temperatures by removing lighter components from the crude prior to entering the hotter parts of the preheat train. Removal of the light components avoids crude vaporisation at the charge heater pass controllers, which can cause control issues. Preflash towers are also often used to hydraulically unload the CDU main column to allow additional and/or lighter crude processing.

It will not come as a surprise that the way CDUs are operated can affect greatly refinery profit. On the one hand, CDUs typically account for approximately 25% of a refinery’s total energy consumption, thus their operation will affect significantly the overall site energy efficiency and energy bill. On the other hand, and perhaps more important from an economic point of view, if operating correctly they can help to shift refinery yields toward more valuable products without increasing existing processing/conversion.

In grassroots designs, preflash drums are likely to bring some of the following advantages: increased reliability and flexibility; lower operating cost; safeguard against water entrainment/slugs in the crude; and possible improved energy efficiency. These potential benefits should be weighed against the capital cost implications of the additional equipment required for these facilities.

A previous article covered the use of pinch and process simulations to determine investment ideas to improve the energy efficiency of CDU preheat trains with preflash.¹ In this article we will show how process simulations of the CDU as well as the entire refinery, built in Petro-SIM, have been used to evaluate operational changes to preflash facilities that result in more efficient CDUs from a yield and energy point of view as well as reducing capital investment in the grassroots design of a crude preheat train.

Case study 1: yield
(North American refinery)
This North American refinery heat integrated its crude preheat train with a downstream unit with the objective of improving the refinery’s energy efficiency. Thus heat from a hot stream of this downstream unit was used to preheat the crude further. Due to the new preheat train configuration, the crude temperature to the preflash tower was increased significantly. So much more, in fact, that the overhead cooling system became a bottleneck (reflux was kept at or near maximum). Control of the preflash overhead naphtha final boiling point (FBP) became difficult. The modified preheat resulted in 96% of the naphtha produced in the unit coming from the preflash tower.

A model of the entire refinery was developed in Petro-SIM and used to evaluate the effect on refinery yields of these preheat train changes, which were implemented prior to this study. The Petro-SIM refinery model showed that while the changes reduced the duty on the crude heaters, they also potentially decreased the amount of more valuable distillate product that could be recovered in the CDU. For this refinery, it appeared that the lost yield opportunity may be more valuable than the energy savings resulting from the improved preheat.

Due to the overhead cooling being a bottleneck, the only way to manipulate the overhead naphtha FBP is to adjust the crude temperature to the preflash tower. Bypasses exist on most exchangers in the preheat train, thus adjusting these bypasses would reduce the preflash tower inlet temperature and therefore the overhead naphtha FBP but also increase the duty on the crude heaters. The sensitivity analysis shown in Figure 1, performed using the model of the entire refinery, showed that the economic optimum for the overhead naphtha D86 FBP is approximately 18°C lower than the base case value, and to reach this FBP the crude temperature to the preflash must be reduced by approximately 12°C.

The refinery model shows that the lower temperature allows ~5000 b/d of the preflash naphtha production to be shifted to the crude tower. However, only ~50% of this material is recovered as crude naphtha with the remaining material drawn as kerosene. Less than 500 b/d of the additional crude kerosene is expected to be stripped out of the kerosene in the hydrotreating unit, resulting in a net increase of kerosene production of about 2000 b/d. Kerosene production at this facility is fixed, and the additional material is eventually blended into the diesel product (priced at ~$2.0/bbl higher than gasoline).

The shifts in refinery yields, material from gasoline to diesel, for the optimum case is worth approximately $6.0 million/y. However, due to the lower reformate production additional MTBE will have to be purchased to replace the lost reformate octane. There are also some minor costs associated with reduced fuel gas and LPG production. It should be pointed out that naphtha/kerosene barrels that are allowed to remain with the naphtha leaving the crude unit will suffer liquid volume loss across the reformer. If they are routed to diesel, the volume loss will be minimal.

In terms of energy, the firing rate on the crude unit increases by 14.4 Gcal/h, which is equivalent to $2.2 million/y. The economics associated with this case are only valid if the crude heater does not become a limit on the amount of crude that can be processed. Thus, the exchangers should not be bypassed if the result is a lower crude throughput due to heater firing limits.