【作者】 李娜；

【导师】 滕伟； 吴仕明；

【作者基本信息】 华北电力大学(北京) ， 工程硕士（专业学位）， 2024， 硕士

【摘要】 未来海上风电发展方向逐渐趋于深远海作业,其所处环境恶劣,实时工况多变,监测数据庞杂,运行机理不明,而全面细致可实施的运维调度是海上风电快速稳定发展的关键,针对海上风电机组的定制化运维决策是大幅降低维修成本、保证海上作业安全,提升风力发电效益的重要手段。本文从更为精细化的部件级运维决策和窗口期内更为合理化的机组级运维排程角度展开。在海上复杂多变的运维工作环境下,计及动态变化的运维任务及多种海上资源的约束,以最小运维成本为目标,优化迭代一种海上风电机组动态运维决策。论文的主要研究内容如下:(1)建立了一种融合寿命预测的精细化部件级机会成组维修策略。以可靠度模型分析为基础,针对无法以可靠度量化的部件,利用寿命预测模型延拓部件范围,建立最小化维修成本为目标的机会成组迭代模型。通过案例验证所提方法可以有效降低成本,提升部件级运维决策的精细化程度。(2)研究了一种考虑均衡分配的海上风电机组群坐标区域划分的方法。基于文森特公式建立了海上风电机群坐标,针对实际风场中坐标系统不统一的问题,提出一种基于二维七参数的转化模型实现不同坐标系之间的转化。研究了基于预知任务的运维船路径排程算法,比较不同排程算法结果,分析最优排程方法。研究了基于坐标先验下的聚类算法,实现风场的均衡区域划分。案例分析表明,本文所建立的海上风电机组坐标与实际情况的偏差较小。根据路径排程算法的结果给出符合实际的最短运维路径。基于坐标先验的聚类算法的区域划分结果更为稳定与均衡。(3)提出了一种基于区域划分的海上风电机组级机会维修策略。在满足维修窗口期的前提下,研究了在区域划分建议下各运维船的维修任务的二次分配,根据现场运维计划与人员情况计算预计维修时间,研究了通过机组任务排序做满额分配的机组群维修任务筛选,并给出机组级的运维调度计划。利用多因素影响作为待维修机组的重要度排序,再根据时间窗的限制对维修任务实现满额分配。通过案例验证,基于区域划分的机组级机会成组维修策略使得在时间窗口中机组维修任务执行更多,维修计划更灵活,运维成本更低。更多还原

【Abstract】 The future development direction of offshore wind turbine has gradually tended to deep and distant ocean operations,the environment it operates in is harsh,real-time working conditions are variable,monitoring data is complex,and the operating mechanism is unclear.Comprehensive,detailed and achievable operation and maintenance scheduling is the key to the rapid development of offshore wind turbine.Customized operation and maintenance decisions for offshore wind turbines behave prominent meaning to significantly reduce maintenance costs,ensure offshore operation safety,and improve wind power generation efficiency.The manuscript explores more refined component-level operation and maintenance decisions and more rational turbine-level operation and maintenance scheduling in the window duration.In the ever-changing operation and maintenance environment,taking into account the dynamic changes of operation and maintenance tasks and the constraints of various offshore resources,an optimized and iterative dynamic operation and maintenance decision for offshore wind turbines is proposed with the target of minimizing operation and maintenance costs.The main contributions are as follows.(1)The manuscript is utilized to establish a detailed component-level opportunity group maintenance strategy that integrates life prediction.Based on reliability model analysis,a life prediction model is used to extend the range of components for components that cannot be quantified by reliability.Through case verification,the proposed method can be used to reduce costs and improve the refinement of component-level operation and maintenance decisions.(2)The manuscript studies a method for dividing the coordinate regions of offshore wind turbine groups considering balanced allocation.Based on the Vincent formula,the coordinates of offshore wind turbine groups were established.To address the issue of inconsistent coordinates in actual wind farms,a two-dimensional seven-parameters transformation model is proposed to achieve the transformation between different coordinate systems.Studing the path scheduling algorithm for operation and maintenance ships is based on predictive tasks,which is used to compare the results of different scheduling algorithms for analysing the optimal scheduling method.Studing clustering algorithms based on coordinate priors is utilized to achieve balanced region division of wind fields.Through case verification,the deviation between the coordinates of offshore wind turbines established in this paper and the actual situation is relatively small.Based on the results of the path scheduling algorithm,provide the shortest practical operation and maintenance path.The clustering algorithm based on coordinate priors results in more stable and balanced region partitioning.(3)The manuscript proposes a regional division based opportunity maintenance strategy for offshore wind turbines.On the premise of meeting the maintenance window duration,the secondary allocation of maintenance tasks is studied under the regional division suggestion.The estimated duration is calculated based on the on-site maintenance plan and personnel situation.The selection of maintenance tasks for turbine-groups with full allocation was studied through turbine-task sorting,and a turbine-level operation and maintenance scheduling plan is provided.Using multiple factors to prioritize the importance of the turbines to be repaired,and then achieving full allocation of maintenance tasks based on window duration constraints.Through case studies,it has been verified that the strategy of maintenance tasks for turbinegroups based on regional division enables more execution of maintenance tasks,more flexible maintenance plans,and lower operation and maintenance costs within the window duration.更多还原