Digital twins have been developed in the oil and gas industry to support a more precise risk assessment that enables performance improvement of offshore structures throughout their life span. Usually, a floating, production, storage and offloading (FPSO) unit is required to exploit oil fields in deep and ultra-deepwater with the postponement of the decommissioning stage, leading to increasing maintenance costs due to aging effects. This paper proposes an adaptive methodology for the development of a digital twin that performs automated numerical analysis via finite element model updating (FEMU) based on coupled systems of a high-fidelity FPSO hull model. The methodology employs a three-cargo tank length finite element (FE) model to receive data and automatically solve multiple numerical analyses. If during results checking any alert level is reached by any structural component, a more complex structural reliability method is applied to provide failure probability considering material strength statistical distribution. The application herein investigates the effects of corrosion, but other phenomena can be considered within the developed framework. A corrosion prediction model is used to create different hypotheses of degradation for bottom plates, while data provided from coupled systems are considered to investigate the effect of deterioration. The results demonstrate a consistent probability of failure when compared to the evolution of the predicted corrosion during service life.

石油和天然气行业开发了数字孪生，以支持更精确的风险评估，从而提高海上结构在其整个生命周期内的性能。通常，深水和超深水油田开采需要浮式生产储卸油装置（FPSO），且退役阶段推迟，由于老化影响，维护成本增加。本文提出了一种用于开发数字孪生的自适应方法，该数字孪生通过基于高保真 FPSO 船体模型耦合系统的有限元模型更新 (FEMU) 执行自动数值分析。该方法采用三货舱长度有限元 (FE) 模型来接收数据并自动求解多个数值分析。如果在结果检查期间任何结构部件达到任何警报级别，则应用更复杂的结构可靠性方法来提供考虑材料强度统计分布的失效概率。本文的应用研究了腐蚀的影响，但在开发的框架内也可以考虑其他现象。腐蚀预测模型用于创建底板退化的不同假设，同时考虑耦合系统提供的数据来研究退化的影响。结果表明，与使用寿命期间预测的腐蚀演变相比，故障概率是一致的。