Real time simulation enhancement of drilling operation measurements

En kommersiell versjon av modellen er ferdig og er blir kjørt som en digital tvilling i MPD opersioner. Modellen har da som formål å estimere trykket i bunnen av et borehull samt å sende styresignaler til utstyr på toppen slik at trykket nede i brønnen også kunne kontrolleres. Resulateter fra disse kjøreingen er presenteret i papert « Using an Advanced Digital Twin to Improve Downhole Pressure Control» Modellen er også videreutviklet til å kunne simulere gass-kick, til bruk i training. Denne delen av modellen er nå godkjent av IWCF(International Well Control Forum) Det er også skrevet et paper som omhandler hvordan modellen kan brukes for å forbedre estimatet for vekten på borekronen. Erfaring fra boringen viste behov for en temperatur modell. Denne er ferdig utvilket og skal verifiseres med data fra NOV-testrig. Vi har også publisert er journal artikkel «A study of the heave-induced surge and swap pressures when a drillstring is in the slips of a floating vessel».

-

A.1.1 Background: Measuring what is going on downhole during a drilling operation has proven to be very difficult. Today, the most commonly used method is to add a few sensors downhole and send the information up via mud pulse telemetry at a bit rate of a few bits per second. More advanced telemetry systems is knocking on the door to the industry, but is struggling to gain ground due to cost and lack of reliability. Today the industry is looking for ways to cut cost and there is an eagerness to advance from manual or remote controlled to more automated or autonomous systems. A key to achieve more automation is to understand what is going on downhole and fill the gap between the available sensor measurements. A.1.2 Objective The objective of the study is to look at what is available today of topside and downhole measurements and see if these can be utilized better to understand what is going on downhole A key to achieve this is to combine the measurements with a mathematical model of the downhole environment, and predict unmeasured states and parameters, or to fill in prediction in-between slow rate measurements. Different techniques such as Least Square methods or Kalman filters should be evaluated for this purpose. The study should look into utilizing existing models and measurement. In addition, and maybe most important, the study should investigate how this can be integrated real-time in the control systems. The thesis should also imply how the operation can benefit from these measurements. Finally, but not the least, the thesis should investigate the practical feasibility of the solution.