The prevention of well-control incidents requires stringent well-design and casing standards, blowout prevention equipment, safe drilling practices, and multiple layers of controls and workflows. However, industry statistics suggest that most well-control incidents occur because of human factors such as misinterpretation of data, delayed response to abnormal well conditions, and drilling crew physical fatigue. Real-time data has long been used by major operators for drilling optimization and well-placement activities but not so much for well control.
Exception-based monitoring of real-time data by remote operators can increase operational awareness of drilling crews, which can significantly minimize well-control risks by taking proactive and precise actions. However, there are several challenges in the way that current technology is used. Real-time streaming data from a well can consist of 50 to 500 tags that make it challenging to simultaneously and continuously monitor such a large amount of data. The current technology requires determining the well condition from single-curve plots and generic alerts—a technique that is time consuming and allows for errors. Poor-quality data can also adversely affect interpretation, analysis, and decision making.
The authors have developed a new real-time alarm solution to overcome some of these challenges by combining signal processing and complex logic to manipulate raw and fragmented data and to emulate the thought process of a drilling operator. This real-time system enables drilling operators to set up different alarm thresholds for various activities of well construction such as drilling, tripping, and circulation. A user-friendly alarm console provides valuable situational awareness, enabling the operator to proactively respond to abnormal conditions using real-time alerts based on well condition.