Session: 04-01-02 Flexible Pipes and Umbilicals II

Paper Number: 104905

104905 - Frequency Domain Analysis and Machine Learning Technique in Field Fatigue Monitoring of Flexible Risers 

Field fatigue monitoring of flexible riser has been gaining more interests in offshore industry. The traditional monitoring approach is based on the time domain simulation, where the real time environmental and functional loading data are collected and converted into the time history stress and strain results of the strength components of the pipe and the subsequent accumulative damage. The time domain approach with real-time data does produce more realistic fatigue results, but at the cost of time-consuming simulation. It sometimes will cause a lag between real-time input data and computed riser fatigue results.

The paper presents an application of frequency domain approach in the online fatigue monitoring of flexible risers to reduce computational effort. The study is focused on the tensile wires, as these layers usually exhibit larger fatigue damage than other layers. The frequency domain method was implemented in fatigue analysis with artificial Intelligence technique, where guided machine learning is used to establish the responses pattern of the tensile armour wire along its configuration. With enough data learnt, the field analysis could reduce the calculation to certain locations to speed up the computation and reduce the lag. Fatigue damage accumulation at other locations can be predicted by the trained model.

Presenting Author: Jiabei yuan Baker Hughes

Presenting Author Biography: Jiabei Yuan graduated from the University of Texas at Austin in 2011 with a PhD in civil engineering. He has been working for Baker Hughes as a riser engineer since 2011.

Authors:

Jiabei yuan Baker Hughes
Linfa Zhu Baker Hughes
Yucheng Hou Baker Hughes
Zhimin Tan Baker Hughes
Eric Wilson Baker Hughes