(54bc) Developing Probabilistic Barrier Failure Models to Predict Kicks While Drilling

Predicting a kick timely and efficiently is often a challenging task due to the complexities of drilling and other well intervention activities. Leading indicators based probabilistic barrier failure models for different stages of drilling are developed in this work to assess performance of primary well control barrier – hydrostatic head. These models would help to determine the probabilities of kick initiating events and identify key leading indicators for predicting kicks and preventing blowouts. The key causal factors for primary well control barrier (hydrostatic head) failure have been identified by conducting fault-tree analysis and analyzing historical incident data. Abnormal pore pressure and swabbing are found to be the major contributors for low hydrostatic head or mud column failure. To predict abnormal pore pressure and swabbing events effectively, leading indicators model are developed combining organizational, operational and real-time indicators. A Bayesian network tool is used to construct probabilistic well barrier failure models for these causal elements. The probability distribution for observing changes in real-time parameters when a kick is developing due to abnormal pore pressure or swabbing are also determined. These parameters are function of both kick initiating events and some influencing organizational factors. Proposed models have two major functions. First, with available leading indicators data probabilities of kick initiating events can be estimated and key parameters which need further improvement for successful kick prevention can be identified. And next, efficiencies of different kick detection parameters can be determined for a confirmed kick scenario and performance influencing factors for these parameters can be identified and improved for better kick detection efficiency. This study would allow both predictive (causes to effect) and diagnostic (effect to causes) reasoning of kicks and blowouts for better understanding of well control system while drilling. Developed risk models enable informed decision making with a relatively clear picture of the risk of barrier failure and provide useful information on actions required to prevent escalation of well control events.