Flexible Riser Remnant Life Assessment Using Artificial Neural Networks and Advanced Inspection Methods

Abstract

Abstract This paper presents a methodology that is currently being utilized to assess the remnant life of flexible risers. The objective is to accurately determine whether flexible risers can safely continue to operate by determining the current condition of the annulus and armor wires, and determining the remnant life using advanced analysis methods in combination with inspection data. The inspection part of this method consists of ultrasonic scanning of the flexible riser in key locations to confirm whether the annulus is dry or flooded and, if flooded, determine the remaining tensile wire thickness. The results of the ultrasonic scanning are then used in combination with global and local riser analysis to determine the remnant fatigue life of the riser. In order to have a more refined and realistic numerical model, time-domain irregular wave approach is used. However, the combination of the need for detailed nonlinear models and the demand for long time series simulations makes this type of analysis extremely time consuming. The method proposed to reduce computational costs in this type of numerical model is a hybrid methodology that combines Finite Element Analyses (FEA) and Artificial Neural Networks (ANN). The proposed hybrid methodology aims to predict the loads at a defined section of the flexible riser. Firstly, using short FEA simulations to train the ANN, and then using only the ANN and the prescribed global riser response to get the remainder of the stress time histories. Results to date have shown that the general assumptions used on corrosion rates and the regular wave dynamic analysis approach are each very conservative, and significant life extensions have been justified based on the results of the combined inspection and analysis method. The use of these methods has the potential to confirm the acceptability of flexible riser life extension beyond the original design life, or to confirm the acceptability to continue to operate with annulus conditions outside the original design specification. The use of ANN for flexible riser fatigue analysis and the use of ultrasonic scanning to determine the present conditions of the flexible riser annulus and armor wires are each advanced techniques in flexible riser integrity management. When combined together as this paper proposes, these methods provide both a current status of the flexible pipe integrity as well as a forecast of the remaining allowable life, allowing key field decisions to be made quickly and accurately.