Dynamic Pre-Modeling of MPD Operations Enabled Optimal Procedures and Operations

Abstract

Outline: An advanced dynamic flow and temperature model was used to optimize and control MPD operations at Kvitebjørn. This paper describes in detail pre-simulations done to prepare for operations and develop detailed procedures, and post-analysis of real time data done to improve performance with respect to upcoming MPD operations at Kvitebjørn. The paper gives details on:Pre-simulations of drilling operations including temperature effectsExtra offline simulations during operationsPost-simulations with analysis of model-data deviations Results and observations: The paper discusses in details critical operations with main focus on sequences with multiple fluids in the well. These include setting of a balanced mud pill, tripping out and in with the pill in the hole, circulating out the pill, and cementing. Comparison with data taken when testing in cased hole is discussed, and it is described how good pressure control was obtained in real operations after having drilled into the reservoir. Applications and conclusions: The paper gives useful input to the use of advanced dynamic models for MPD operations. It gives a good impression of current status as well as possibilities and limitations for future operations. Introduction The Kvitebjørn field in the North Sea, close to Statfjord and Gullfaks, started production in 2004. Conventional drilling continued after this, but the increasing pressure depletion caused severe losses on the ninth well. This event put an end to conventional drilling at Kvitebjørn, and the development of a comprehensive managed pressure drilling setup started. Due to an operational window between pore and fracture pressure that was very small and uncertain, good control of all dynamic effects was addressed by using and advanced dynamic flow and temperature model both before, during and after operations on each well. By using calibrated dynamic software and detailed measurements of mud and fluid parameters as a function of pressure and temperature, it was possible to make procedures based on the best available data and the most reliable computations. A general overview of the first MPD operations at Kvitebjørn is given in Ref. 1, while more details on parts of the setup is given in Refs. 2–4. Earlier UBD/MPD experience offshore Norway is described in Refs. 5–8. Model overview The numerical model used in these computations has been thoroughly discussed in Petersen et al. (Ref. 9). Only a very brief discussion is included here, please go to Ref. 9 for a more detailed discussion.

Simplified Hydraulics Model Used for Intelligent Estimation of Downhole Pressure for a Managed-Pressure-Drilling Control System

“…Assuming the cross section of the flow A(x) is piecewise constant, Eq. 9 is similar to the equation used in advanced hydraulics models (see, for example, Petersen et al 2008a). Note also that, by using Eq.…”

“…Much effort, therefore, has been put into developing advanced hydraulics models in order to capture all aspects of the drilling hydraulics (Rommetveit and Vefring 1991;Petersen et al 1998Petersen et al , 2001Petersen et al , 2008aLage et al 1999;Lage and Time 2000;Bjørkevoll et al 2000Bjørkevoll et al , 2003. These models are able to reproduce a wide range of drilling-specific effects to an impressingly high degree of detail.…”

Adaptive estimation of downhole pressure for Managed Pressure Drilling operations

“…The adaptive observer (10) guarantees that, for any initial conditions, the solutions q bit (t) , e q (t) andθ (t) to (12), (13) and (14) are globally uniformly bounded and satisfy lim t→∞ (q bit (t) , e q (t)) = 0. Proof: Consider the function V q bit , e q ,θ = 1 2…”

“…This has triggered a lot of research into developing advanced hydraulic models in order to capture all aspects of the drilling hydraulics, see e.g. [11], [12], [13], [14]. These models are based on distributed parameter models of multiphase flow and are able to reproduce a wide range of drilling-specific events, such as pipe connections, cementing, gelling and multi-fluid scenarios to an impressingly high degree of detail [15], [16].…”