Applications of continuum fatigue risk monitoring in riser connectors system integrity management. (1st February 2022)

• Record Type: Journal Article
• Authors: Yan, Yifei; Zhang, Shengyue; Jin, Xin; Xu, Liangbin; Yan, Xiangzhen
• Is Part Of: Ocean engineering. Volume 245(2022)
• Publication Date: 2022-02-01
• Subjects: Marine riser connectors -- Multi-coupling models -- Fatigue failure risk -- Risk-based inspection -- Risk early warning; Ocean engineering -- Periodicals; Ocean engineering; Periodicals; 620.4162
• DOI: 10.1016/j.oceaneng.2022.110540
• Languages: English
• ISSNs: 0029-8018

Abstract

Abstract: This paper discusses the effects of marine riser connectors (MRCs) on fatigue failure risk using multi-coupling models. When MRCs are subjected to vortex-induced fatigue, a risk early warning strategy is used for effective risk monitoring as a preventive measure for the operation of MRCs. To ensure safe marine oil and gas production and to improve fatigue failure prevention and control, research on offshore structural fatigue risk failure theory criteria and evaluation and the vibration fatigue characteristics of MRCs over time and frequency is characterized. By coupling data-driven technological means, including rain-flow counting (RFC), Hilbert–Huang transform (HHT), power spectral density (PSD), artificial neural network (ANN), risk-based inspection (RBI), and risk management measures (RMM) to assess the fatigue risk characteristics of in-service members in special offshore environments. The fluid-structure interaction characteristics of MRCs and a load spectrum are analysed by empirical mode decomposition (EMD) to determine the changes in the components of an intrinsic mode function (IMF) and investigate the degree of fatigue damage. The fatigue damage of MRCs is predicted using the PSD of the IMF and the RFC of the IMF-coupled ANN time series prediction model. The results show that using any IMF component or combined component data as input to the ANN and training set facilitates cross-validation of the analysis. When the input layer and a hidden layer of the ANN are optimized and properly trained, the predicted performance is suitable. Through early warnings and coordination, the spread of fatigue failure risk trends can be predicted, and warning information is issued accordingly. Highlights: Proposed a multi-coupling early warning process framework. Proposed a fatigue failure risk analysis method for marine riser connectors. Developed a neural network time-series-based fatigue risk analysis method. Developed a preventive fatigue risk early warning process approach.