Automatic Gas Influxes Detection in Offshore Drilling Based on Machine Learning Technology

Abstract The gas influxes occur frequently during the exploration in the South China Sea, which can pose severe risks to well control strategies. The traditional detection method is to analyze the mud-log of mud tanks, which is slow and there is always a time-lag between the gas influxes occurrence time and the influxes detection time. In this work, a methodology is proposed that allows for real-time gas influxes detection using artificial intelligence and data analytics. To develop the methodology, mud-log (12 parameters) are collected for 208 influxes incidents for 62 drilled wells. Data analysis through artificial neural network is carried out to develop gas influxes warning system model. Then, a new comprehensive method using mud-log, is proposed for real-time gas influxes detection. This is one of the first attempts for real time gas influxes detection utilizing data analysis of mud logging and artificial intelligence. This methodology is successfully applied to gas field in South China Sea with accuracies up to 95% achieved.