Cascade Control of Unstable Systems with Application to Stabilization of Slug Flow

Abstract

The topic of this paper is the effect of stabilizing control on the remaining control problem. In many cases there is no effect. However, stabilization requires the active use of inputs, and the underlying unstable pole will appear as an undesirable unstable zero if we are concerned with input performance. The implications of this are clearly demonstrated on the application to stabilization of severe slugging in two-phase pipeline-riser systems. We find that a controllability analysis gives important information for measurement selection and performance limitations.

Selection of variables for stabilizing control using pole vectors

“…Nevertheless, the pole vectors have proven themselves useful in several applications with more than one unstable mode, including the stabilizing control of the Teneessee-Eastman process (Havre, 1998) (Havre and Skogestad, 1998) with 6 unstable modes, and the selection of pressure sensor location for stabilization of desired two-phase flow regimes in pipelines (Havre et al, 2000) (Storkaas et al, 2001) which has a pair of complex RHP-poles. For such applications the pole vectors need to be interpreted with care and the results need to be checked, for example, by designing controllers.…”

Model-Free Predictive Anti-Slug Control of a Well-Pipeline-Riser

“…Several active control strategies have been addressed for stabilizing the slugging phenomena, some of them are mentioned in Godhavn et al (2005), Ogazi AI (2010), Di Meglio et al (2010a, Storkaas and Skogestad (2003a) and Jahanshahi and Skogestad (2015), Dalen et al (2015).…”

Model-free optimal anti-slug control of a well-pipeline-riser in the K-Spice/LedaFlow simulator

“…• Feedback PID control strategy of a pipeline-riser, controlling the riser base pressure with the topside choke as control input, in Ogazi AI (2010), Jahanshahi and Skogestad (2015), Storkaas and Skogestad (2007), Storkaas et al (2001) and Skogestad (2009).…”