Fouling Modelling in Crude Oil Heat Exchanger Networks using Data Reconciliation and Estimation of Unmeasured Process Variables

Abstract

Abstract In crude oil refineries, fouling has a significant impact on operating and maintenance costs. The growth of fouling layers on the surface of heat exchangers in a crude oil heat exchanger network (HEN) also increases both energy use and CO2 emissions. Fouling models facilitate process design and integration while systematically accounting for the impact of fouling. The most common models use a series of parameters that are specific to the type of crude oil, as well as the operating conditions of the pre-heat train. These models are of great importance as they can indicate the set of operating conditions (stream velocities and temperatures) at which the occurrence of fouling might be avoided. A key challenge is to define an accurate method for determining fouling models for each side of a heat exchanger within a network. To achieve this goal, plant-measured data are used to monitor the thermal performance of a pre-heat train, and to obtain specific insights related to the impact of fouling. The main concern related to these data and their interpretation is the effect of measurement error and the limited number of measurements available in the process. This work presents a new methodology for calculating fouling model parameters for both sides of shell-and-tube heat exchangers using simulated plant-data along with data reconciliation and gross error detection. A heat exchanger network model coupled with fouling dynamics is used for simulating a crude oil pre-heat train and for the prediction of fouling behaviour. The effect of full and partial instrumentation in the estimation of reconciled measurements is accounted for in a case study, where the prediction of fouling deposition is assessed.