【作者】 孙强；

【导师】 钟万勰；

【作者基本信息】 大连理工大学 ， 动力学与控制， 2024， 博士

【摘要】 随着全球能源需求的持续增长,海洋油气资源的开发显得尤为重要。悬链式单点系泊系统(Catenary Anchor Leg Mooring,CALM)作为连接海上油气运输油船与海底管道的核心设施,承担着海上油气的运输中转、油船的系泊定位等多重关键功能。因此,在复杂海洋环境作用下,悬链式单点系泊系统的水动力特性对海上油气作业的安全与效率具有重要的影响。鉴于以上工程背景,本文针对悬链式单点系泊系统的水动力特性分析及其优化设计方法开展了研究。(1)基于经典动力学分析方法,针对悬链式单点系泊系统建立了力学简化分析模型,提出了悬链式单点系泊分析方法,深入分析了完整作业、完整自持、破损作业以及破损自持四种典型工况下锚链布置方案对浮筒偏移量和锚链安全系数的影响,确定了最优锚链布置方案;针对以上最优锚链布置方案,开展了单点浮筒的主尺度选型研究,给出了单点浮筒直径的可行域,为单点浮筒的前期预研设计提供科学依据;进一步基于水动力分析软件AQWA,建立了悬链式单点系泊系统-油船浮体耦合动力学时域分析模型(以下简称:系泊-船耦合时域分析模型),结合动力分析时程响应最大值统计方法,提出了系泊极限强度分析方法。与传统线性频域分析模型相比,本文所建立的系泊-船耦合时域分析模型可充分考虑非线性因素和时域动态特性。针对悬链式单点系泊系统开展了水池物理模型试验,验证了上述系泊-船耦合时域分析模型的准确性,为悬链式单点系泊系统的水动力特性分析提供了有效的分析工具。(2)基于所建立的系泊-船耦合时域分析模型,开展了悬链式单点系泊系统的水动力特性研究,重点分析了风、浪、流共线和非共线两种工况下油船鱼尾效应对悬链式单点系泊系统响应的影响。研究表明油船的鱼尾效应会导致悬链式单点系泊系统发生较大的极值响应,对系泊系统的稳定性和作业安全带来潜在危险,因此需要在工程设计中予以重点考虑;为了有效减弱鱼尾效应的不利影响,系统地分析了面向鱼尾效应的悬链式单点系泊的设计环境条件,从静力和动力时程角度研究了导缆孔孔距对悬链式单点系泊油船的艏摇非稳定性运动的联合影响规律。研究表明,不同风速、流速、缆绳长度以及拖轮力等条件下,均可通过调整船艏导缆孔的孔距,有效控制悬链式单点系泊油船的鱼尾效应,并深入阐明了鱼尾效应的控制机理。基于此研究,提出了通过调整导缆孔孔距以控制由鱼尾效应引起的油船非稳定性运动的控制方法。(3)采用所建立的系泊-船耦合时域分析模型,结合二次函数的自适应取样方法,构建了锚链成本与锚链线型设计参数之间的高精度代理分析模型;以悬链式单点系泊系统锚链成本最小化为优化目标,设置锚链直径与预张力角为设计变量,将浮筒偏移量、锚链张力、疲劳损伤以及锚链垂向力等影响系泊系统安全性的四个关键参数作为约束条件,采用遗传算法与序列二次规划法相结合的优化算法,建立了悬链式单点系泊系统的多约束多设计变量的优化设计分析方法。实际工程案例应用结果表明,基于本文所提出的优化设计分析方法,既确保了悬链式单点系泊的性能,又实现了锚链成本的优化设计。相较于传统的参数敏感性优化分析,本文所提出的优化设计分析方法可大幅缩短前期的设计周期,实现高效、成本优化的系泊设计。(4)基于系泊-船耦合时域分析模型,系统地考虑了多种海洋环境条件和系泊系统设计参数,进行了大规模数值计算,构建了包含11.8万组数据的悬链式单点系泊系统响应数据库。基于上述数据库,全面分析了决策树、随机森林、极端随机森林、梯度提升决策树以及K近邻五种经典的机器学习方法对悬链式单点系泊系统响应预测的准确度。研究表明,基于随机森林方法的悬链式单点系泊响应预测具有良好的精度。因此,本文采用随机森林机器学习方法,构建了悬链式单点系泊响应预测的机器学习模型,实现了对不同海洋环境参数和锚链线型设计参数条件下悬链式单点系泊系统响应的快速预测,为面向多种海况的悬链式单点系泊快速设计提供了有益探索。更多还原

【Abstract】 As the global demand for energy continues to grow,the development of marine oil and gas resources becomes particularly important.The Catenary Anchor Leg Mooring(CALM)system,as a core facility connecting marine oil and gas transportation vessels with subsea pipelines,undertakes multiple key functions such as the transportation and transshipment of marine oil and gas,and the mooring positioning of vessels.Therefore,under the influence of complex marine environments,the hydrodynamic characteristics of the CALM system have a significant impact on the safety and efficiency of marine oil and gas operations.In view of the above engineering background,this thesis has conducted research on the hydrodynamic characteristics analysis and optimization design methods of the CALM system.(1)Based on the classical dynamic analysis method,a simplified mechanical analysis model was established for the CALM system,and a CALM analysis method was proposed.An in-depth analysis was conducted on the impact of anchor chain arrangement schemes on the displacement of the buoy and the safety factor of the anchor chain under four typical working conditions: operation,stand-alone,operation with one mooring line damaged,and stand-alone with one mooring line damaged.The optimal anchor chain arrangement scheme was determined.For the above optimal anchor chain arrangement scheme,a study on the main size selection of the single-point buoy was carried out,and a feasible domain for the diameter of the single-point buoy was given,providing a scientific basis for the preliminary research and design of the single-point buoy.Further,based on the hydrodynamic analysis software AQWA,a coupled dynamic time-domain analysis model of the CALM system-oil tanker floating body(hereinafter referred to as the mooring-ship coupled time-domain analysis model,MSTAM)was established.Combined with the maximum value statistical method of dynamic analysis time history response,a mooring limit strength analysis method was proposed.Compared with the traditional linear frequency-domain analysis model,the proposed MSTAM can fully consider nonlinear factors and time-domain dynamic characteristics.A physical model test of the CALM system was carried out in the water tank,verifying the accuracy of the proposed MSTAM,providing an effective analysis tool for the hydrodynamic characteristics analysis of the CALM system.(2)Based on the proposed MSTAM,a study on the hydrodynamic characteristics of the CALM system was carried out,focusing on the impact of the fishtail effect of the oil tanker on the response of the CALM system under two conditions: collinear and non-collinear wind,wave,and current.The study shows that the fishtail effect of the oil tanker can cause a large extreme response of the CALM system,posing potential dangers to the stability and safety of the mooring system,which needs to be considered in engineering design.To effectively reduce the adverse effects of the fishtail effect,the design environmental conditions of the CALM system facing the fishtail effect were systematically analyzed.The combined influence law of the fairlead hole spacing on the instability motion of the bow sway of the CALM oil tanker was studied from the static and dynamic time history perspectives.The study shows that under different wind speeds,current speeds,cable lengths,and tugboat forces,the fishtail effect of the CALM oil tanker can be effectively controlled by adjusting the spacing of the bow fairlead holes,and the control mechanism of the fishtail effect was deeply explained.Based on this research,a control method for the instability motion of the oil tanker caused by the fishtail effect by adjusting the fairlead hole spacing was proposed.(3)Using the established MSTAM,combined with the adaptive sampling method of the quadratic function,a high-precision surrogate analysis model between the anchor chain cost and the anchor chain line type design parameters was constructed.With the minimization of the anchor chain cost of the CALM system as the optimization goal,the anchor chain diameter and the pretension angle were set as design variables,and four key parameters affecting the safety of the mooring system,such as the displacement of the buoy,the tension of the anchor chain,fatigue damage,and the vertical force of the anchor chain,were used as constraints.An optimization algorithm combining genetic algorithm with sequential quadratic programming was employed to establish a multi-constraint multi-design variable optimization design analysis method for the CALM system.The application results of the actual engineering case show that the proposed optimization design analysis method not only ensures the performance of the CALM system but also achieves the optimized design of the anchor chain cost.Compared with the traditional parameter sensitivity optimization analysis,the proposed optimization design analysis method can greatly shorten the preliminary design cycle and achieve efficient and costoptimized mooring design.(4)Based on the proposed MSTAM,a large number of numerical calculations were carried out by systematically considering various marine environmental conditions and mooring system design parameters,and a response database of the CALM system containing 118,000 sets of data was constructed.Based on the above database,a comprehensive analysis was conducted on the accuracy of five classic machine learning methods,i.e.,decision tree,random forest,extreme random forest,gradient boosting decision tree,and K-nearest neighbor,for predicting the response of the CALM system.The study shows that the CALM response surrogate model based on the random forest method has good accuracy.Therefore,this paper uses the random forest machine learning method to construct a machine learning model for predicting the response of the CALM system,achieving rapid prediction of the response of the CALM system under different marine environmental parameters and anchor chain line type design parameters,providing a beneficial exploration for the rapid design of the CALM system facing various sea conditions.更多还原