Multi-objective optimization considering cost-benefit ratio for the placement of gas detectors in oil refinery installations

Abstract

It is of great importance for an oil refinery installation to have an effectively designed gas detection system. The majority of current research on the optimization of gas detector placement focuses on a single-objective stochastic programming (SP) approach. However, it is actually a multi-objective optimization problem, trying to obtain optimal solutions with conflicting criteria. This paper proposes a multi-objective optimization method for detector placement, achieving a balance between the detection efficiency improvement and the required investment. Based on the leakage scenario and the consequence modelling, the cumulative death probability (CDP) in the whole unit is quantitatively represented. The reduced fatality can also be evaluated according to the detection time. Then, a cost-benefit ratio (CBR) model of the gas detector investment and benefit from fatality reduction can be established. With the objective of minimizing both the CDP and CBR, a multi-objective layout optimization model considering the cost-benefit (MLOM-CB) is proposed. A non-dominated sorting genetic algorithm-II (NSGA-II) is used to obtain acceptable results. The optimization of hydrogen sulfide gas detector placement is taken as a case study. Results demonstrate that this approach can improve the detection efficiency of gas detector networks and the flexibility of decision-making for gas detector investment.