Relative Motion Prediction for Integrated Floating Offshore Wind Turbine Installation Scheme Based on Data-Driven
Published in International Conference on Ocean, Offshore and Arctic Engineering (OMAE), 2024
本文面向海上整体化对接作业中“预组装浮式风机部件与基础”之间相对运动不可预测的难题，提出一种数据驱动的相对运动极值预测方法。通过在多种海况下对安装过程进行数值仿真，并将相对运动的统计特征（均值与中心矩等）作为输入，构建反向传播神经网络模型，用于预测关键对接点的最大/最小相对位移与速度。结果显示该模型具有较高预测精度（(R^2) 最高可达 0.95，且 MSE 较低），可为潜在风险提供早期预警，并支撑新型双体船安装方案的作业安全决策。该数据驱动框架有助于提升深水区域浮式风机安装的可行性与可靠性。
This research addresses the challenge of unpredictable relative motions between preassembled floating wind turbine components and foundations during offshore mating operations. By simulating installation scenarios across diverse sea states and utilizing statistical motion data (mean values and central moments) as inputs, a backpropagation neural network model is developed to forecast the maximum and minimum relative displacements and velocities at critical mating points. The model demonstrates high prediction accuracy (with R² values up to 0.95 and low MSE), providing crucial early warnings of potential hazards and supporting operational safety decisions for the novel catamaran-based installation method. This data-driven approach enhances the feasibility of installing floating wind turbines in deeper waters.
Recommended citation: Can Ma, Taiyu Zhang, Zongyuan Yang, Yang An, Xiang Yuan Zheng, and Zhengru Ren*. (2024). "Relative Motion Prediction for Integrated Floating Offshore Wind Turbine Installation Scheme Based on Data-Driven." ASME 2024 OMAE.
Download Paper