Offshore Wind Foundation Diagnostic Monitoring System (OFDiMoS)

海上风力涡轮机（OWTs）是一种科学成熟的可再生能源技术，具有显著降低社会碳排放的潜力。近年来，该领域发展迅速，制造了更高效、更大的涡轮机，需要更高的支撑塔和更大的基础。这些新结构与以前的模型越来越不同，用于估计其基础性能的设计方法变得越来越不适用和可靠。新的owt也被安装在海上更不确定和更具挑战性的地面条件下。这意味着这些发展的行为可能与预期不同，由于对强风和海浪与这些结构之间的相互作用知之甚少，从而增加了发生破坏的风险。基础建设和维护占整个风电场成本的20-40%，因此，通过降低能源生产成本，任何提高其可靠性的努力都可以为英国消费者带来直接的经济效益。该项目将开发一种计算机工具，通过分析安装在这些机器上的传感器的数据来识别OWT行为的损害和实时变化。该工具创建了结构-基础系统的数字孪生体，可以进行实时损伤诊断。数字孪生是一个跨越其生命周期的对象的虚拟表示，从实时数据更新，并使用模拟，机器学习和推理来辅助决策。对于owt，这表示物理结构的精确对应（或孪生）。该工具将使用先进的岩土模型来估计土壤-结构相互作用的行为，实时更新，这是非常新颖的。该系统将告知实际结构的执行情况，并将使未来的性能预测能够进行，并估计剩余的使用寿命。鉴于该行业最近的快速发展以及围绕新OWT装置行为的不确定性日益增加，这是及时的。数字孪生在其他行业也有应用，但尚未成功应用于OWT资产管理。一项具体的创新是在目标结构上安装传感器，该传感器将反馈实时动态数据，从而获得近乎瞬时的状态评估，便于在瞬态事件（风暴）期间检测到损坏时进行快速维护干预。可以分析数天/数周的长期趋势，以识别较慢的损害积累，如腐蚀。长期性能可以根据设计案例进行基准测试，用于确定基础的大小，从而在最小化风险的情况下，促进围绕延长使用寿命的明智决策，最终降低能源成本。

Offshore Wind Turbines (OWTs) are a scientifically-mature renewable energy technology with potential to significantly lower society's carbon emissions. There have been rapid recent developments in this sector, with more efficient and larger turbines being fabricated, requiring higher support towers and larger foundations. These new structures are increasingly unlike previous models and the design methods used to estimate their foundation behaviour are becoming less applicable and reliable. New OWTs are also being installed further offshore in less certain and more challenging ground conditions. This means that these developments might behave differently than expected, increasing the risk of damage occurrence due to poorly understood interactions between harsh wind and waves with these structures. Foundation construction and maintenance constitute 20-40% of overall windfarm cost and therefore any effort made to increase their reliability has direct financial benefits for UK consumers, by lowering the cost of energy production.This project will develop a computer tool to identify damage and real-time changes in OWT behaviour by analysing data from sensors installed on these machines. The tool creates a digital twin of the structure-foundation system, which can carry out real-time damage diagnosis. A digital twin is a virtual representation of an object that spans its lifecycle, is updated from real-time data, and uses simulation, machine learning and reasoning to assist decision-making. For OWTs, this represents an exact counterpart (or twin) of the physical structure. The tool will use advanced geotechnical models to estimate the soil-structure interaction behaviour, updated in real-time, which is highly novel. This system will inform how the real structure is performing, and will also enable future performance predictions be made, with estimation of remaining useful life. This is timely given the recent rapid development of the sector and the increasing uncertainty surrounding the behaviour of new OWT installations. Digital twins have been used in other industries but have not yet been successfully applied to OWT asset management. A specific innovation is the use of installed sensors on the target structure that will feedback real-time dynamic data, enabling near-instantaneous condition assessments be obtained, facilitating rapid maintenance interventions should damage be detected during transient events (storms). Longer term trends over days/weeks can be analysed to identify slower damage accumulation such as corrosion. Long-term performance can be benchmarked against design cases used to size the foundations, facilitating informed decision-making surrounding service life extension with minimised risk, which ultimately lowers energy costs.