Control Oriented Modeling of a De-oiling Hydrocyclone

Deoiling hydrocyclones are an important part of the downstream water treatment in offshore oil & gas production, they ensure a low discharge of oil and thus a higher yield of produced oil. This work investigates the possibility of developing a simple control-oriented model of a de-oiling hydrocyclone based on experimental data that can support systematic analysis and control design of hydrocyclone systems. The most widely used control solution of a hydrocyclone is a Pressure Drop Ratio (PDR) control strategy, which is often empirically designed and experimentally tuned in a case-by-case manner. There is a lack of a systematic and deep-insight analysis of the capability, stability and limitations of these control solutions, as there are few control-oriented models available for de-oiling hydrocyclone systems. This paper proposes a method of retrieving a set of simple 1st-order transfer function models from a set of designed experiments based on a lab-scaled hydrocyclone system. Some preliminary results are also illustrated and discussed. The conclusion of this preliminary study is that the models obtained can emulate the dynamics of the system in a reasonable manner subject to the trade-off between simplicity and accuracy. In addition, higher order state space models are introduced and their relative advantage of depicting the complicated dynamics of the hydrocyclone's pressures is investigated.