Structural Integrity Management of Offshore Structures via RB-FEA and Fast Full Load Mapping Based Digital Twins
- David J. Knezevic (Akselos, Inc.) | HeonYong Kang (Texas A&M University) | Partha Sharma (DNV-GL) | Grzegorz Malinowski (DNV-GL) | Trong Thuc Nguyen (Akselos S.A.)
- Document ID
- International Society of Offshore and Polar Engineers
- The 28th International Ocean and Polar Engineering Conference, 10-15 June, Sapporo, Japan
- Publication Date
- Document Type
- Conference Paper
- 2018. International Society of Offshore and Polar Engineers
- Digital twins, full load mapping, FEA, structural analysis
- 4 in the last 30 days
- 67 since 2007
- Show more detail
In this work we discuss new capabilities for structural integrity analysis of critical offshore assets such as semi-submersibles and FPSOs that are enabled by Reduced Basis FEA (RB-FEA) coupled with Fast Full Load Mapping to enable high-fidelity Digital Twins. RB-FEA is a new structural simulation framework based on component-based model order reduction, in which (i) a large model is decomposed into components, (ii) each component is endowed with parameters which enable geometric, material, and/or load data to be modified on the fly, (iii) a reduced order model of each component (valid for the entire parameter range) is prepared based on an automated pre-analysis of individual components and groups of components, and (iv) component-based models are assembled and solved very efficiently due to re-use of the reduced order model data. For large models this framework enables orders-of-magnitude increase in model detail and solver speed compared to conventional FEA, without sacrificing accuracy. Fast Full Load Mapping obtains full distributive loads acting on a floating body in random waves at costs of conventional global performance analyses, without any costly CFD. In time domain, these full distributive loads are fed into RB-FEA, and the combination of these two powerful approaches enables new capabilities for detailed hydrodynamic and structural analysis of critical offshore assets. We demonstrate these capabilities with some illustrative examples.
Structural integrity management of critical floating assets such as FPSOs and semisubmersibles is a daunting challenge. The assets are extremely large and complex, subjected to harsh offshore environments for decades, and are often life-extended beyond their original design life. Operators are increasingly appreciating the need to leverage next-generation digital technology in order to address these extreme challenges.
A crucial digital technology for providing insight into the current state of floating assets is a so-called digital twin, i.e. a detailed digital model of the entire asset, which represents all relevant aspects of the asset, e.g. strength, fatigue, topside and ballast weight, metocean conditions, etc. Conventional attempts to develop digital twins have relied on finite element analysis (FEA) (Zienkiewicz and Taylor, 2014) for structural modeling, and frequency domain hydrodynamic analysis for wave loading. However, these approaches impose severe limitations on the size and detail of the digital twin that can be developed. For example, it is well-known that FEA is highly computationally expensive, and hence FEA models have strong limits on the detail level that can be incorporated, and hence generally FEA models of large assets are not detailed enough to incorporate the current state of the asset, including corrosion, cracks, damage. As a result, the coarse modeling that is necessitated by FEA is inherently incompatible with the concept of a digital twin, which is intended to capture all relevant data for the current state of the asset in one global model. Similarly, frequency-domain hydrodynamic loading also imposes severe restrictions on the wave loading analysis, e.g. it is restricted to regular waves, and nonlinear modeling of moorings and risers cannot be incorporated.
|File Size||869 KB||Number of Pages||7|