Controle e otimização de uma unidade de hidrodessulfurização de múltiplas correntes de diesel

Abstract

The presence of contaminants such as sulfur, nitrogen and aromatics in diesel is strictly controlled, following specific regulations of the fuel sector. These components affect diesel quality, impacting pollutant emissions and product marketing. The most widely used process to meet diesel specifications is catalytic hydrogenation, specifically hydrotreatment (HDT). Brazilian legislation imposes the use of S10 diesel. In this thesis, a HDT unit of multiple diesel streams was studied using a phenomenological mathematical model of a trickle-bed reactor (TBR). Predictive control strategies (MPC) were implemented, searching for the product in the S10 diesel specification. A problem of optimization of the HDT process was proposed, solved by the interior point method, considering the hydrodesulfurization reaction. The optimization sought to minimize operating costs by keeping the sulfur content in the product below the maximum allowable value. A kinetic model of catalytic deactivation was tested in order to evaluate its effects in the diesel HDT process. With the control and optimization strategies, the contaminant levels reached values that meet the S10 diesel specification. Thus, the MPC controller presented satisfactory performance and, with the optimization, reached the optimized values of 404.09 US$/h of the cost function, 699 K of the TBR feed temperature, 0.21 cm/s of the superficial velocity of the hydrogen gas. In terms of optimization, the flow rates of oils A, B and C were 0.018 cm3 /s, 0.022 cm3 /s and 0.06 cm3 /s respectively, and the concentration of sulfur compounds resulting from mixing the oils in the feed was 302 ppm.