Systems engineering based advanced optimization for sustainable water management in refineries

Abstract

Refineries are increasingly compelled to play an imperative role in achieving sustainable management of water as mandated by sustainable development goals (SDGs). Existing optimization models lack the ability to addresses this challenge of obtaining optimal refinery water network configuration for problems of industrial scale due to difficulty in application of complex solution algorithms of mixed integer non-linear problem (MINLP) formulations to large dimensions. Therefore, the optimization formulation in this work exploits process engineering derived constraints to build an integrated superstructure, which results in a simplified non-linear problem (NLP), and thus reduces computational complexity. Further, to substantially minimize fresh water consumption, the proposed model incorporates novel features of a modern refinery complex viz. distributed effluent treatment plant based on total dissolved solids (TDS) levels in waste water, segregation of sour water treatment based on phenolics content in condensate stream, and minimizing cooling tower blowdown rates by optimizing the cycles of concentration in cooling tower operation. The formulation is demonstrated for optimal water routing and economic benefits of the proposed approach through a representative case study of a real refinery complex. Towards the goal of zero liquid discharge, an efficient new scheme of recycling treated water to the boiler feed water system and cooling tower make up water requirements in optimal ratios, has been suggested from the results of the proposed model. Sensitivity of water reallocation in optimal refinery water network with respect to crude quality and desalter performance is also presented through application of proposed optimization framework in the case studies.