Synthesis of refinery hydrogen networks based on compressor performance models

Abstract

Synthesis of refinery hydrogen network contributes to reductions in hydrogen consumption, production cost, and emissions. Hydrogen compressors are drivers for hydrogen distribution, which has a direct effect on the performance of hydrogen networks. The existing works focus on the hydrogen network synthesis and often use simple compressor models for easy modelling and solving, which may lead to overly ideal results. This work proposes a novel mathematical method for refinery hydrogen network synthesis considering compressor performances, incorporating both centrifugal and reciprocating compressors. The compression power vs. volumetric flowrate curves and the compression ratio vs. volumetric flowrate curves are considered for a centrifugal compressor, while the efficiency of a reciprocating compressor is correlated with the compression ratio. A mixed integer nonlinear programming model (MINLP) is formulated to minimize the total annualized cost. The developed MINLP model is examined based on two hydrogen network cases reported in literature. The results show that the minimum total annualized cost is a co-optimization of compressor type, operating points, compressor placement, and hydrogen network structure. In the two cases, compression efficiencies of centrifugal compressors range from 0.760 to 0.795 and that of reciprocating compressors are between 0.539 and 0.807. The consideration of centrifugal compressors rather than just reciprocating compressors not only improves the compression efficiencies but also simplify the hydrogen network structures. Moreover, it also brings reductions in the compression power by 2.23% and 1.97%, and decreases in the compressor investment by 7.12% and 6.87%.