Fast Model Based Approximation of the Closed-loop Performance Limits of Gas/Liquid Inline Separators for Accelerated Design

Abstract

A current trend in the oil and gas industry is to use compact so called inline separators (ILS). Unlike their large conventional counterparts, the operation of these separators is very sensitive towards variations in the multiphase flow to be separated. This sensitivity easily results in operational problems and economic loss and may prohibit the application of ILS, in particular as many current production operations are facing large slug flow type of variations. One way to reduce the ILS sensitivity towards flow variations is via improved control. Here, motivated by the industrial need for cost-effective compact separators with sufficient flow variation handling capabilities, a model based approach is pursued to obtain this improvement. More specific, as a first main contribution, a new approach to control oriented modeling of gas/liquid (G/L) ILS is proposed which, in contrast to currently available such modeling approaches, allows for a comprehensive evaluation and design of G/L ILS control strategies. As an example application of the models resulting from this approach and as a second main contribution of this paper, a new model and feedforward control based method for fastly approximating the closed-loop performance limits of a G/L ILS is proposed. The motivation for pursuing this method is an acceleration in overall G/L ILS design speed. The merits of the method are demonstrated through a simulation based application on a commercially available G/L ILS.

State and Parameter Estimation for a Gas-Liquid Cylindrical Cyclone

“…The last group of (advanced) controllers typically require a model of the plant and therefore, they achieve both (feedback and feedforward) control of the measured variables and optimization based on unmeasured variables (e.g. [7], [8], [9], [10] and [11]). Information of these unmeasured variables are only available through estimation and require the development of estimators.…”

Adaptive feedback linearizing control of a gas liquid cylindrical cyclone

“…More recently, a control-oriented model of a GLCC including the phenomena of liquid carry over (LCO) and gas carry under (GCU) has been developed [9]. This model used empirical data for GCU and LCO calculations and thus gave a more correct representation of the dynamics of a GLCC.…”