Tower, Spar platform and mooring system are designed in the project based on a given 6-MW wind turbine. Under wind-induced only, wave-induced only and combined wind and wave induced loads, dynamic response is analyzed...Tower, Spar platform and mooring system are designed in the project based on a given 6-MW wind turbine. Under wind-induced only, wave-induced only and combined wind and wave induced loads, dynamic response is analyzed for a 6-MW Spar-type floating offshore wind turbine (FOWT) under operating conditions and parked conditions respectively. Comparison with a platform-fixed system (land-based system) ofa 6-MW wind turbine is carried out as well. Results demonstrate that the maximal out-of-plane deflection of the blade of a Spar-type system is 3.1% larger than that of a land-based system; the maximum response value of the nacelle acceleration is 215% larger for all the designed load cases being considered; the ultimate tower base fore-aft bending moment of the Spar-type system is 92% larger than that of the land-based system in all of the Design Load Cases (DLCs) being considered; the fluctuations of the mooring tension is mainly wave-induced, and the safety factor of the mooring tension is adequate for the 6-MW FOWT. The results can provide relevant modifications to the initial design for the Spar-type system, the detailed design and model basin test of the 6-MW Spar-type system.展开更多
大数据时代,流数据大量涌现.概念漂移作为流数据挖掘中最典型且困难的问题,受到了越来越广泛的关注.集成学习是处理流数据中概念漂移的常用方法,然而在漂移发生后,学习模型往往无法对流数据的分布变化做出及时响应,且不能有效处理不同...大数据时代,流数据大量涌现.概念漂移作为流数据挖掘中最典型且困难的问题,受到了越来越广泛的关注.集成学习是处理流数据中概念漂移的常用方法,然而在漂移发生后,学习模型往往无法对流数据的分布变化做出及时响应,且不能有效处理不同类型概念漂移,导致模型泛化性能下降.针对这个问题,提出一种面向不同类型概念漂移的两阶段自适应集成学习方法(two-stage adaptive ensemble learning method for different types of concept drift,TAEL).该方法首先通过检测漂移跨度来判断概念漂移类型,然后根据不同漂移类型,提出“过滤-扩充”两阶段样本处理机制动态选择合适的样本处理策略.具体地,在过滤阶段,针对不同漂移类型,创建不同的非关键样本过滤器,提取历史样本块中的关键样本,使历史数据分布更接近最新数据分布,提高基学习器有效性;在扩充阶段,提出一种分块优先抽样方法,针对不同漂移类型设置合适的抽取规模,并根据历史关键样本所属类别在当前样本块上的规模占比设置抽样优先级,再由抽样优先级确定抽样概率,依据抽样概率从历史关键样本块中抽取关键样本子集扩充当前样本块,缓解样本扩充后的类别不平衡现象,解决当前基学习器欠拟合问题的同时增强其稳定性.实验结果表明,所提方法能够对不同类型的概念漂移做出及时响应,加快漂移发生后在线集成模型的收敛速度,提高模型的整体泛化性能.展开更多
基金financially supported by the National Basic Research Program of China(973 Program,Grant No.2014CB046205)
文摘Tower, Spar platform and mooring system are designed in the project based on a given 6-MW wind turbine. Under wind-induced only, wave-induced only and combined wind and wave induced loads, dynamic response is analyzed for a 6-MW Spar-type floating offshore wind turbine (FOWT) under operating conditions and parked conditions respectively. Comparison with a platform-fixed system (land-based system) ofa 6-MW wind turbine is carried out as well. Results demonstrate that the maximal out-of-plane deflection of the blade of a Spar-type system is 3.1% larger than that of a land-based system; the maximum response value of the nacelle acceleration is 215% larger for all the designed load cases being considered; the ultimate tower base fore-aft bending moment of the Spar-type system is 92% larger than that of the land-based system in all of the Design Load Cases (DLCs) being considered; the fluctuations of the mooring tension is mainly wave-induced, and the safety factor of the mooring tension is adequate for the 6-MW FOWT. The results can provide relevant modifications to the initial design for the Spar-type system, the detailed design and model basin test of the 6-MW Spar-type system.
文摘大数据时代,流数据大量涌现.概念漂移作为流数据挖掘中最典型且困难的问题,受到了越来越广泛的关注.集成学习是处理流数据中概念漂移的常用方法,然而在漂移发生后,学习模型往往无法对流数据的分布变化做出及时响应,且不能有效处理不同类型概念漂移,导致模型泛化性能下降.针对这个问题,提出一种面向不同类型概念漂移的两阶段自适应集成学习方法(two-stage adaptive ensemble learning method for different types of concept drift,TAEL).该方法首先通过检测漂移跨度来判断概念漂移类型,然后根据不同漂移类型,提出“过滤-扩充”两阶段样本处理机制动态选择合适的样本处理策略.具体地,在过滤阶段,针对不同漂移类型,创建不同的非关键样本过滤器,提取历史样本块中的关键样本,使历史数据分布更接近最新数据分布,提高基学习器有效性;在扩充阶段,提出一种分块优先抽样方法,针对不同漂移类型设置合适的抽取规模,并根据历史关键样本所属类别在当前样本块上的规模占比设置抽样优先级,再由抽样优先级确定抽样概率,依据抽样概率从历史关键样本块中抽取关键样本子集扩充当前样本块,缓解样本扩充后的类别不平衡现象,解决当前基学习器欠拟合问题的同时增强其稳定性.实验结果表明,所提方法能够对不同类型的概念漂移做出及时响应,加快漂移发生后在线集成模型的收敛速度,提高模型的整体泛化性能.