Session: 02-11-02 Fatigue and Fracture Reliability 2

Paper Number: 104540

104540 - Intelligent Assessment of Fatigue Damage Based on the Monitoring Data of Offshore Structures 

Structural fatigue damage is one of the primary damage forms to offshore structures. The fatigue damage assessment of typical offshore structures is of great significance to ensure the safety and reliability of the overall and local structures during the whole lifetime. The core of fatigue damage assessment based on the monitoring data is to obtain the hot-spot stress information in the structure through the sensors. However, the monitor blind spots on the hot-spot stress maybe existed since the sensors cannot be deployed to cover all the key positions in some actural scenarios, which brings difficulty on the fatigue life assessment. To address this issue, a novel hot-spot stress inversion model based on the multilayer BP neural network is proposed in this paper, which can effectively invert the hot-spot stress through the nearby limited sensors. With the inverted hot-spot stress, the stress spectrum is obtained by rain flow counting method and the fatigue life corresponding to the stress level is obtained according to S-N curve. Experiments have been condcuted based on the simulation data and actural measured data by using model test for connector structure of offshore engineering. The results showed that the proposed hot-spot stress inversion model can predict the hot-spot stress values with high accuracy and can be used to assess the fatigue damage. 

Presenting Author: Xueliang Wang China Ship Scientific Research Center, Wuxi, China

Presenting Author Biography: Xueliang Wang graduated from China Ship Scientific Research Center in 2006 and 2012, respectively, with a master's degree and a PhD degree in Naval Architecture and Ocean Engineering. In 2016, he served as deputy director of ship structure research department of China Ship Scientific Research Center. His research interests include ship wave loads, structural safety monitoring, structural digital twin, etc.

Authors:

Xueliang Wang China Ship Scientific Research Center, Wuxi, China
Libin Zhou School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, China
Guoqing Wu China Ship Scientific Research Center, Wuxi, China
Yucheng Wang School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, China
Hongtao Mei Jiangnan University
Hengyang Lu Jiangnan University
Zhe Liu School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, China
Wei Fang School of Artificial Intelligence and Computer Science, Jiangnan University, Wuxi, China