logo


Aug-2016

A two stage approach to controlling fouling on preheat trains at OMV’s Schwechat refinery

Fouling in heat exchanger preheat trains is a well-known problem in the refining industry. Fouling reduces exchanger performance, increases furnace duty, reduces cooling capacity and increases pressure drop. The resulting increase in fuel costs and loss in throughput can significantly affect refinery profitability.

José Ignacio Beltrán Peral, Peter Reinberger OMV
Arjan Baks, Juan Gómez Prado, Anthony Waters, KBC Advanced Technologies

Viewed : 6424


Article Summary

This paper presents a two stage approach to this problem: Stage 1: Control the sources of fouling by improving crude management and storage, desalter operation, crude compatibility checks and other “housekeeping” issues. Stage 2: Monitor the remaining fouling and cleaning potential using a solution based on rigorous simulation.

In Stage 2, rigorous simulation is needed to account for complex network interactions and accurately calculate the cleaning potential. An example of the importance of this is shown in this paper.

In addition, the monitoring solution results should be accessible by a wide range of site functions, including Operations, Maintenance, Process and Planning.

A case study presents the use of this two-stage approach for a fouling mitigation study for a crude unit at OMV’s Schwechat Refinery. In Stage 1, engineering judgement and site interviews were used to compare the current operation against best practice. In addition, a crude compatibility study highlighted which crude mixtures should be avoided. In Stage 2, KBC’s Petro-SIM™ HX Monitor is used to track the fouling in the preheat train, and recommend the most cost effective cleaning strategy.

Introduction
Fouling in heat exchanger preheat trains is a well-known problem in the refining industry. Fouling will gradually reduce heat transfer coefficients and therefore heat recovery. This means that the furnace inlet temperature will drop over time, thus increasing the furnace duty and the operating cost. In addition, pump-around cooling duty can be affected by fouling, which can in turn affect the cooling capacity and throughput of the unit. A further effect is that fouling deposits effectively reduce tube diameters leading to an increase in the pressure drop.

All these issues will increase operating energy cost, but more importantly might cause a loss in throughput, thus revenue. Throughput loss due to fouling will have a significant impact on refinery profitability.

Fouling mitigation is therefore necessary to minimise these costs. It is typical that most refineries do some fouling mitigation, such as filtration, desalting, some cleaning outside of the main shutdowns and some monitoring. However, leading practices within the industry are moving towards a complete approach, firstly focussed on prevention, followed by early diagnosis of fouling and then ensuring that cleaning is the most cost-effective it can be.

Fouling Prevention
The types of fouling experienced in crude oil include:
1) Particulate fouling due to deposition of organic precursors (e.g. asphaltenes);
2) Corrosion fouling due to the presence of naphthenic acid, salts, sulphur and water;
3) Chemical reaction fouling due to polymerization of heavy hydrocarbons at high temperatures.

The fouling prevention mechanism would depend on the type of fouling. Some of the common techniques to prevent fouling are: filtration, desalting, fouling inhibition, crude compatibility and preheat revamp.

Regarding filtration, crude oil from the reservoir is generally contaminated with water, salt, wax, sand and mud, which contain ferric oxides and other particulates. These particulates deposit on the heat exchanger surfaces, promoting build-up of organic and inorganic fouling. It has been reported [2] that fouling at the cold end of the preheat train can be reduced by filtering the crude oil. After filtration, water and any residual salt can be removed through desalting. This will reduce corrosion fouling.

Fouling inhibition can be achieved by mixing hydrocarbons [3] and by anti-fouling chemicals. Leading refineries are now accompanying these with further fouling prevention, such as crude compatibility checks. In some cases revamp of the preheat train is also considered in order to reduce fouling. This is done to: (a) switch fluids placement, (b) avoid vaporisation, (c) avoid corrosion, and (d) ensure optimum tube and shell velocities.

Fouling Monitoring
Irrespective of the fouling prevention methods, it is standard practice for refiners to clean all the exchangers in the preheat train during shutdown. The period between start-up and shutdown is termed the run length for the preheat train and this is normally between four and six years.

The preheat train run length may be set by a regulatory body or determined by the refinery managers based on economics. In the second case, the typical objective is to maximise the run length, and wherever possible extend it, as a shutdown on the refinery has a significant financial implication. This will usually necessitate cleaning of some exchangers during the run, in order to maintain the throughput and operating performance of the preheat train.

The key questions are then “Which exchangers to clean, why, and when?” In order to answer this, refiners will typically use “experience”, in-house monitoring tools or rigorous simulation software.

Experience led cleaning can achieve savings, but such experience is hard to maintain and update with the latest plant conditions, so additional exchangers may be cleaned with little or no impact on the network performance. Also, experience is easily lost as people move on, and businesses cannot afford to have repeated cycles of trial and error learning.

The next step, an in-house monitoring tool, is typically a stand-alone spreadsheet model which simply calculates the “UA” in the exchangers. It may, or may not, be linked to real plant data. These are typically created by the plant engineering experts, and are difficult to update or maintain. They are usually limited to showing fouling trends, rather than simulating the benefit of cleaning. Also, they must make many simplifying assumptions about how to model exchangers and calculate fluid properties. One weakness of this is that they cannot distinguish between “UA” reductions caused by fouling and those caused by flow decreases. Consequently, the results can only be properly interpreted by expert users.

In-house models are labour intensive, due to the need for manual data reconciliation, crude property determination and the vital result analysis. The time taken for these tasks, as well as the potential loss of the expert resource or software changes, add further complications to creating an effective and streamlined business.

The final step, based on rigorous simulation, is described in a later section. The advantage of this approach is shown in Fig. 1.


Add your rating:

Current Rating: 3


Your rate: