Rigorous Refinery-wide Optimisation: A Case Study For Petronor

Nowadays, refinery operations must be optimized in order to cope with the increasing costs of producing within specifications and with variable feedstock. Simulation software is a fundamental tool for optimizing the refinery and increasing its margins. Traditionally, in order to define the optimal raw material to be purchased, refinery margins are maximised using Linear Programming (LP) tools. Nevertheless, daily refinery optimization requires the use of more rigorous models that take into account operating constraints and non-linear behavior of the process units. Hence, rigorous simulation software complements the LP and plays a significant role in improving the efficiency of the refinery. This paper addresses the use of rigorous simulation software (Petro-SIM) for optimizing the daily operations in Petronor, subsidiary company of Repsol YPF, near Bilbao in northern Spain. Petro-SIM is a comprehensive rigorous simulator that allows modeling of an entire refinery from crude feed to finished product blending. The use of global models that integrate different refinery units, allows analysis of their inter-dependencies, which provides a strong foundation to support the daily planning of refinery operations. Furthermore, Petro-SIM is also used to improve the accuracy of the LP model and to optimize the operations of refinery units. The rigorous refinery-wide model of the Petronor refinery predicts yields and qualities from refinery units for each crude change. It also calculates other important variables such as column cut points or blending recipes. Results of those predictions are sent to Operations, and used to set initial targets until actual product quality data comes from the laboratory. Since the crude basket changes every few days, the availability of the simulation results enhances operation and significantly increases the refinery margin.