New Experimental Results Show the Application of Fiber Optic to Detect and to Track Gas Position in Marine Risers and Shed Lights on the Gas Migration Phenomenon Inside a Closed Well

Abstract

Abstract The main objective of this manuscript is to present and to discuss the results and significant observations gathered during 13 experimental runs conducted in a full-scale test well at Louisiana State University (LSU). The other two objectives of this manuscript are to show the use of distributed fiber optic sensing and downhole pressure sensors data to detect and to track the gas position inside the test well during the experiments; and to discuss experimental and simulated data of the gas migration phenomenon in a closed well. An existing test well at LSU research facilities was recompleted and instrumented with fiber optic sensors to continuously collect downhole data and with four pressure and temperature downhole gauges at four discrete depths within an annulus formed 9 5/8″ casing and 2-7/8″ to a depth of 5025′. A chemical line was attached to the tubing allowing the nitrogen injection at the bottom of the hole. The research facilities were also equipped with a surface data acquisition system. The experiments consisted in injecting nitrogen into the test well filled with water by two means: either injecting it down through the chemical line or down through the tubing to be subsequently bullheaded to the annulus. Afterwards, either the nitrogen was circulated out of the well with a backpressure being applied at surface to mimic an MPD operation or left to migrate to the surface with the test well closed. During the runs, the three acquisition systems (fiber optic, downhole gauges, and surface data acquisition) recorded all relevant well control parameter for a variety of gas injected volumes (2.0-15.1 bbl), circulation rates (100-300 GPM) and applied backpressures (100-300 psi). The experimental results gathered by the acquisition systems were very consistent in measuring gas velocities inside the well. The numerical model predictions matched very close the pressure behavior observed in the experimental trials. In the gas migration experiments, it was observed that the bottomhole pressure is not carried to the surface and that this pressure is a function of the volume of gas injected in the well. These facts are supported by the numerical simulation results. The manuscript shows the possibility of the use of fiber optic and downhole pressure sensors information to detect and to track the gas position inside a well or the marine riser during normal or MPD operations. Additionally, the vast amount of experimental data gathered during the experiments in which the nitrogen was left in the closed well to migrate to surface helped shed lights on the controversial issue concerning the surface pressure build-up while the gas migrates to surface in a closed well. Numerical simulations were all instrumental for supporting the findings.