Novel Analyses and Design of Relief Wells Considering 3D Effects

Abstract

In developing a relief well contingency, current industry standard practice is to use a one-dimensional (1D) multiphase flow model to determine the requirements such as pump rate and mud weight to kill the blowout well. However, this does not consider certain variables. For instance, this analysis returns the same results regardless of the intercept angle of the relief well (i.e. results are identical if the relief well intercepts at a steep angle into the direction of flow, or if it intercepts at a steep angle in the same direction of flow). A novel approach has been developed that considers the three-dimensional vector effects when a blowing out well is killed by means of a relief well intercept. This analysis offers safety and environmental benefits for well kill design and operations because it provides results which more accurately reflect the physical principles. This allows for optimization of a relief well design. For instance, this more comprehensive analysis allows the possibility to design a relief well with shallower intercept depth and lower pump requirements with the potential for an earlier kill. This new method considers the complex interaction of the countercurrent flow that occurs at the relief well intercept and can utilize computational fluid dynamics (CFD). This enables optimization of parameters previously not considered such as specific spray design from the relief well. For instance, industry currently assumes that there is an advantage to pumping the kill fluid down the annulus of the relief well. The frictional losses are lowered (compared to pumping down the drillstring), thereby minimizng the pump requirements and maximizing the achievable rate for given pump capability. Whilst this may indeed be the case, the research covered in this paper shows that, depending on the circumstances, there can be a benefit to designing the spray pattern from the flow from the relief well at the intercept. Various relief well kill cases were analyzed by both the standard method and the new method and the results compared. The new method predicted that relief well kills could be made with lower rates and pressures than predicted by the standard model. Using the new method offers an opportunity to increase the accuracy of relief well planning, enabling a more precise understanding of what may be achievable. A test apparatus was designed and created to verify which of the two methods was more accurate. Physical experiments indicate that results from the proposed method match the test results closer than the standard approach.