| Summary: | In off-shore oil and gas platform efficiency, the reliability and fuel flexibility are the major concerns when selecting the gas turbine to support the electrical and mechanical demand on the platform. In order to fulfill these requirements, turbine inlet temperature and pressure ratio are not increased up to the optimal values and one or more redundant gas turbines may be employed. With increasing incentives for reducing the CO2 emissions off-shore, improving the thermal efficiency has become a focus area. Due to the peculiar low turbine outlet temperature and due to space and weight constraints, a steam bottoming cycle is not a convenient solution. On the contrary, organic Rankine cycles (ORCs) present the benefits of high simplicity and compactness. Furthermore, the working fluid can be selected considering the temperature profile at which the heat is supplied; hence the heat transfer process and the thermal efficiency of the cycle can be maximized. This paper is aimed at finding the most optimal ORC tailored for off-shore applications using an optimization procedure based on the genetic algorithm. Numerous working fluids are screened through, considering mainly thermal efficiency, but also other characteristics of the fluids, e.g. stability, environmental and human health impacts, and safety issues. Both supercritical and subcritical ORCs are included in the analysis. The optimization procedure is first applied to a conservative ORC where the maximum pressure is limited to 20 bar. Subsequently the optimal working fluid is identified by removing the restriction on the maximum pressure. Different limits on hazards and global warming potential (GWP) are also set. The study is focused on the SGT-500 gas turbine installed on the Draugen platform in the Norwegian Sea. The simulations suggest that, when a high hazard is accepted, cyclohexane is the best solution. With a turbine inlet pressure limit of 20 bar, the combined gas turbine-ORC system presents an efficiency of 43.7%, corresponding to an improvement of ...
|
|---|