Probabilistic Models for Kick Tolerance

Abstract Kick tolerance is an evaluation of how large kick sizes i.e. kick volumes an open hole section can withstand without threatening the formation integrity at the shoe. If a certain kick size cannot be handled safely, the planned open hole section must be shortened, and the casing design must be altered. Three models for calculating the kick tolerances in a well will be compared for a long and short open hole section for various lengths of bottom hole assembly (BHA). The kick tolerance will be performed probabilistically by use of Monte Carlo simulations where important input parameters are considered as distributions. The paper will focus on where the models differ in their results and discuss various opportunities and challenges with using a probabilistic approach. The models will be integrated in a Monte Carlo simulation framework where the major input uncertainties will be pore pressure, fracture pressure and initial gas distribution in the well. The output will be a distribution of the casing pressure load that has to be compared to the fracture pressure distribution which results in a certain probability for fracturing for a given kick size. Only gas kick in water-based mud will be considered. First a transient model based on the single bubble concept was considered and integrated in the Monte Carlo simulation framework. This was first compared against an analytical model which calculates the maximum casing shoe pressure at static shut in conditions. The analytical model considers uncertainty in the initial gas distribution. A transient flow model based on the drift flux model was also considered. Both short and long open hole length were considered. BHA length and kick size were varied. The results show that the transient flow model provides the least conservative results but also the analytical model reduces the probability for fracturing compared to the single bubble model. In most cases, the maximum casing shoe pressure is achieved when kick is located at the BHA. This paper extends the application of methods for reliability-based casing design to also include probabilistic kick tolerances. This is a contribution related to how the well design process can become more risk based. Some challenges related to specification of tolerance requirements, required number of Monte Carlo simulations and computing time will be discussed. It also provides an overview of the differences between the models and which parameters that are most important for the results.