Big Data Analysis for Monitoring of Kick Formation in Complex Underwater Drilling Projects

Abstract

Over the years, 33 incidents of offshore blowout during drilling have caused significant damage to life and property. This research suggests that field engineers could prevent offshore blowout using timely detection of the impending danger, a.k.a. a kick, using big data analysis tools. In this study, the objective is to develop an algorithm that can detect a kick and display the result without any human intervention. The research purpose is to achieve immediate response time in the detection of kicks to deploy control measures. The contribution of the study is an inexpensive and generalizable automated mud-logging system for the analysis and detection of kicks in underwater drilling projects. The research innovatively implemented the genetic wavelet neural network method to monitor and predict kicks in real time. The developed kick-detection system is able to capture and represent complex input and output relationships of the primary and secondary indicators of the kick to detect any anomalies and display warning messages. Field engineers then use these messages to confirm the possible occurrence of a kick and inform the driller to perform the appropriate control. This algorithm has the potential to be a lifesaving solution for engineers and managers working on offshore projects pile drilling for bridges or waterworks.