Session: 06-17-03 AI Technology for Ocean Engineering III

Paper Number: 124829

124829 - Mooring Line Tensions Prediction Based on Mlp for Soft Yoke Mooring System 

The soft-yoke single point mooring system is a widely used for FPSOs in shallow water. However, there were accidents of mooring system being pulled down by FPSO in field. It is significant for conducting real-time prediction of mooring line tensions. Under complex environmental loads, there is a nonlinear relationship between mooring line tensions and environmental loads due to the coupling effect between a single point system and an FPSO. In recent years, given the strong nonlinear fitting ability of neural networks, scholars explore the application of neural networks to fit the nonlinear relationship between mooring line tensions and environmental loads so that mooring line tensions can be predicted using environmental conditions as inputs.

A multilayer perceptron (MLP) neural network model is established to predict the maximum mooring line tensions of a soft-yoke mooring system for a FPSO in Bohai in this work. A coupled time-domain model of FPSO soft yoke single-point mooring system is established and sensitivity analyses are carried out for the effect of hull draft, and environmental parameters on the maximum mooring line tensions. Accordingly, a series of numerical analyses are conducted to obtain the datasets for machine learning. The MLP model to predict the maximum mooring line tensions is established. The results show that the model has a good fitting effect in predicting the maximum mooring line tensions for short term sea states. This study provides a theoretical basis for predicting the maximum mooring line tensions based on environmental parameters and draft in actual engineering.

Presenting Author: Ying Li Tianin University

Presenting Author Biography: I obtained PhD degree in 2005 from University of Plymouth, UK，joined Tianjin University in 2011. I have work experience both in industry and academic. My field of study is concentrated on design and optimization of offshore structures.

Authors:

Ying Li Tianin University
Qiyuan Zhong Tianjin University
Qiang Gan Tianjin University