Automated Trend-Based Alerting Enhances Real-Time Hazard Avoidance

Abstract

A method is presented to efficiently analyze drilling data using trend analysis to predict downhole hazards and allow proactive decision making. The workflow is a data-driven process that provides analysis on both real-time (RT) and historical drilling events. For RT application, the method provides the user with an alert functionality prior to the hazardous events occurring. This method is designed to augment the Real-Time Operation Center (RTOC) to provide additional layers of real-time predictive capabilities, historical hindsight, situational awareness, and hazard avoidance. A software program has been developed to incorporate the methodology into a real-time workflow.

The core tenet of drilling hazard management (DHM) is process awareness, which is more relevant than ever in the industry. Routine activities such as daily drilling reports (DDR) analysis and manual RT data analysis such as planned versus actual (PVA) data are time consuming and not very efficient because they are highly dependent on the skill sets and experience of the personnel. By using a standardized automated trend analysis, time- or depth-based data can be analyzed to determine the risks and downhole hazards involved in the drilling and completion operations.

This paper presents on-bottom analysis on historical wells to show how trend deviation is used to identify potential hazards before they occur. The trend analysis uses percent deviation for specific parameters, such as standpipe pressure, surface torque, flow rate and hookload, to alert the user when thresholds for normal operations have been exceeded. This logic was developed and validated using more than 50 wells throughout the world, and the logic's thresholds were calibrated from these test cases. Alerts are triggered only when criteria for specific hazards are met.

This method champions hazards avoidance through prevention and ultimately reduces noproductive time (NPT) in well construction. This system enables efficient and standardized DHM for both real-time and historical analysis. The proactive approach in hazard avoidance helps users to be aware of trend deviation in real-time and increases the user's situational awareness through alerts. The software provides audit capabilities through documenting time-stamped acknowledgment or notes from users. This system also enables faster offset well analysis and supplements real-time data engineers to improve overall efficiency. In this paper you will see real results of how the program performed in a blind testing environment. Results from the test case demonstrate that the program can generate useable alerts well before a potentially hazardous event. The alerts prior to hazardous events could improve situational awareness for rig staff and on-site engineers.