Oil-Water Separation in Liquid-Liquid Hydrocyclones (LLHC) –Experiment and Modeling

摘要

The liquid-liquid Hydrocyclone (LLHC) has been widely usedby the Petroleum Industry for the past several decades. Alarge quantity of information on the LLHC available in theliterature includes experimental data, computational fluiddynamic simulations and field applications. The design ofLLHCs has been based in the past mainly on empiricalexperience. However, no simple and overall designmechanistic model has been developed to date for the LLHC.The objective of this study is to develop a mechanistic modelfor the de-oiling LLHCs, and test it against available and newexperimental data. This model will enable the prediction ofthe hydrodynamic flow behavior in the LLHC, providing adesign tool for LLHC field applications.A simple mechanistic model is developed for the LLHC.The required input for the model is: LLHC geometry, fluidproperties, inlet droplet size distribution and operationalconditions. The model is capable of predicting the LLHChydrodynamic flow field, namely, the axial, tangential andradial velocity distributions of the continuous-phase. Theseparation efficiency and migration probability are determinedbased on swirl intensity prediction and droplet trajectoryanalysis. The flow capacity, namely, the inlet-to-underflowpressure drop is predicted utilizing an energy balance analysis.An extensive experimental program has been conductedduring this study, utilizing a 2” MQ Hydroswirl hydrocyclone.The inlet flow conditions are: total flow rates between 27 to18 gpm, oil-cut up to 10%, median droplet size distributionsfrom 50 to 500 ìm, and inlet pressures between 60 to 90 psia.The acquired data include the flow rate, oil-cut and dropletsize distribution in the inlet and in the underflow, the rejectflow rate and oil concentration in the overflow and theseparation efficiency. Additional data for velocity profileswere taken from the literature, especially from the Colman andThew (1980) study. Excellent agreement is observed betweenthe model prediction and the experimental data with respect toboth separation efficiency (average absolute relative error of3%) and pressure drop (average absolute relative error of1.6%).