Vidar Alstad

Reproducing slugging oscillations of a real oil well

This paper addresses the problem of reproducing oscillations generated by the well-known slugging phenomenon in multiphase flow. Reported investigations show how to determine the parameters of a recently proposed ordinary differential equations system, so that it captures the characteristics of actually observed slugging oscillations. A tuning procedure based on the mathematical properties of the model is presented. It is then applied to a test case consisting of a real oil well located in the North Sea. An observer using only topside measurements allows asymptotic reconstruction of critical variables such as the downhole pressure.

Model-based control of slugging flow: An experimental case study

In this paper, we compare two solutions for suppressing slug flow in pipes conveying a diphasic fluid by active feedback control of the outlet (production) choke. Conventionnaly, a non-collocated PI controller is used to stabilize the measured inlet pressure. Unfortunately, this method lacks robustness to changes in operating conditions and reveals difficult to calibrate. In this work, we propose a model-based approach to design a nonlinear state-feedback control law based on a first principles model of the slugging phenomenon. A theoretical proof of convergence of the closed-loop model is provided, as well as an experimental validation of these results on a mid-scale testbench.

Model-based control of slugging: advances and challenges

We review recent advances in the suppression of the slugging phenomenon by model-based control. We focus on three aspects of recent contributions: models, observers and control laws. For each category, we evaluate and compare existing solutions, and propose directions for improvement.

Slugging in multiphase flow as a mixed initial-boundary value problem for a quasilinear hyperbolic system

This paper studies the multiphasis slugging flow phenomenon occurring in oil wells and flow lines. The main contribution is a low-dimensional distributed parameters model, comprising the gas mass fraction, the pressure, and gas velocity as states. Along with appropriate boundary conditions, on the one-dimensional space domain, it constitutes a well-posed mixed initial-boundary value problem for a quasilinear hyperbolic system. Numerical simulation results obtained with a presented characteristics method solver stress the validity of the approach and the fair representativeness of the model. In particular, the period of simulated oscillations and their overall shape is in accordance with reference results from the literature. Controllability and observability open problems are formulated for future works.

Medium-Scale Experiments on Stabilizing Riser-Slug Flow

Original SPE manuscript received for review 24 July 2008. Revised manuscript received for review 11 May 2009. Paper (SPE 120040) peer approved 23 May 2009. Summary This is the second of two papers describing control experiments on a medium-scale slug rig. The first paper (Sivertsen et al. 2009) describes experiments performed on a small-scale laboratory rig built at the Norwegian University of Science and Technology (NTNU) Department of Chemical Engineering. These experiments showed that, despite noisy measurements, it is possible, with feedback control, to “stabilize the flow” (i.e., to achieve reasonably smooth flow in the normally riser-induced severe slug-flow region) using only topside measurements. The question to be answered is whether these results also apply for larger riser systems. In the present paper, we look at some results obtained from a 10-m-high, 3-in.-diameter medium-scale test rig located at the Statoil Research Centre in Porsgrunn, Norway. Several cascade control structures are tested and compared, both with each other and with the results obtained from the small-scale NTNU loop. The rig was also modeled and analyzed using a simple three-state dynamic model. The new experiments were successful and confirm the results of Sivertsen et al. (2009) from the small-scale rig. The valve opening with nonslug flow operation could be increased from approximately 12% with no control to almost 24% with control using topside measurements only. This makes it possible to produce with a larger production rate and increase the total recovery from the producing oil field. The valve opening with control could be further increased to approximately 28% using measurements from the bottom of the riser, but such measurements may not be available in many cases.

Combinations of Measurements as Controlled Variables: Application to a Petlyuk Distillation Column

Abstract A new simple approach for selecting controlleci variables, that give nearoptimal operation with a constant set-point feedback structure in the presence of uncertainty, is presented. The controlled variables are linear combinations of a subset of the available measurements. A method for selecting the best sub-set and the required number of measurements is derived. The method is illustrated on a Petlyuk (Divided wall) distillation column.

Attenuation of Slugging in Unstable Oil Wells by Nonlinear Control

Abstract This paper illustrates the potential of nonlinear model-based control applied for stabilization of unstable flow in oil wells. A simple empirical model is developed that describes the qualitative behavior of the downhole pressure during severe riser slugging. A nonlinear controller is designed by an integrator backstepping approach, and stabilization for open-loop unstable pressure setpoints is demonstrated. The proposed backstepping controller is shown in simulations to perform better than PI and PD controllers for low pressure setpoints, and is in addition easier to tune. Operation at a low pressure setpoint is desirable since it corresponds to a high production flow rate.

Combination of measurements as controlled variables for self-optimizing control

A new method for selecting controlled variables (c) as linear combination of measurements (y) is proposed based on the idea of self-optimizing control. The objective is to find controlled variables, such that a constant setpoint policy leads to near optimal operation in the presence of low frequency disturbances (d). We propose to combine as many measurements as there are unconstrained degrees of freedom (inputs, u) and major disturbances such that Δcopt(d)=0. To illustrate the ideas a gas-lift allocation example is included. The example show that the method proposed here give controlled variables with good self-optimizing properties.