Addressing the Pressing Needs of Offshore Ultra-Deepwater Floating Facilities and Risers: Near Real-time Management System for Deepwater Drilling Risers

摘要

Subsea drilling operations are moving into harsher, deeper and less familiar offshore environments. Tornavoid disruptive events due to uncertainties associated with these environments, a Riser LifecyclernManagement System (RLMS) has been developed for near real-time condition monitoring and fatiguernestimation of drilling risers. In the digital age of the Industrial Internet, decisioning platforms that monitor,rnassess and make recommendations, like RLMS for UDW equipment, are critical to manage risk. Unlikernconventional techniques such as strain gauges for direct strain/stress measurement, the RLMS measuresrnthe vibrations of the riser string using accelerometers at selected joints along the drilling riser. The systemrnthen transmits the vibration data via acoustic telemetry in near real-time to a topside data acquisitionrnsystem on the drilling vessel where fatigue life estimates for all riser joints are made using machinernlearning techniques. Preliminary laboratory and sub-scale rig testing have been conducted to test keyrnsubsystems and validate the system functionality and sensor signal fidelity under simulated environmentsrnand have demonstrated good results.rnThe RLMS is an integrated platform of both hardware and software tools. The RLMS hardwarernincludes subsea sensing modules each with acoustic telemetry, which enables wireless data communicationrnwith a topside data acquisition system on the drilling vessel. A modular approach was used forrndesigning the subsea platform. The platform consists of an acoustic modem and transducer, rechargeablernbatteries, tri-axial accelerometers and gyroscopes, and a micro-processor for data acquisition and processing.rnThe topside system includes software algorithms for data processing, riser fatigue analysis, andrnvisualization and alerts for enhanced operational decision-making by drilling contractors and operators.rnThe fatigue estimation algorithm takes as inputs the riser configuration (geometry, material properties,rnand modal data), the associated digital Radio Frequency Identification (RFID) data from each riser joint,rnand the measured accelerometer data, and in turn generates transfer functions that calculate the oceanrncurrent profile. Specifically, an artificial neural network (ANN) model, combined with an optimizationrnalgorithm, is used to develop the transfer functions. The inputs to the neural network model are currentrnintensities and the outputs are acceleration features at locations along the riser string where the motion sensors are attached. An optimization algorithm is used to match predicted acceleration from the neuralrnnetwork model with measured acceleration features in order to back-calculate the current intensities. Therncurrent intensities are then input into SHEAR7 for estimation of fatigue damage rates.rnThe next phase of the research program will include a field trial to test the integrated RLMSrnfunctionality on a drilling rig for near real-time visibility into drilling risers and riser life assessment.