The oceanic mixed layer in the Southern Ocean is characterized by numerous fronts due to the stirring of freshwater influxes arising from ice melting.The interaction of these fronts with winds modulates the evolution ...The oceanic mixed layer in the Southern Ocean is characterized by numerous fronts due to the stirring of freshwater influxes arising from ice melting.The interaction of these fronts with winds modulates the evolution of the mixed layer and affects atmosphere−ocean energy exchanges.However,the underlying mechanism behind the wind-front interaction remains obscure due to a lack of three-dimensional observations of the ocean,particularly in terms of velocities.To address this issue,this study investigates the dynamics of fronts within the mixed layer during a storm by employing a subset of the global submesoscale-permitting simulation,Northeast Weddell Sea Pre-SWOT Level-4 Hourly MITgcm LLC4320 Native Grid 2km Oceanographic Dataset(ROAM_MIZ).We first compare the ROAM_MIZ data to glider data to assess the performance of the model simulation and find that the ROAM_MIZ can,to a large degree,capture sub-mesoscale features within a mixed layer.Subsequent analyses based on a subset of ROAM_MIZ show that lateral density gradients within the mixed layer rapidly decrease during high winds associated with the storm.Down-front winds accelerate this process as the Ekman buoyancy transport responsible for enhancing the instability of the fronts is primarily dominated by horizontal baroclinic components.After the storm,the fronts strengthen again in the presence of weaker winds due to the frontogenesis by the larger-scale strain.Moreover,the non-geostrophic turbulence induces a modification of the relative vorticity,affecting the instability within the mixed layer.These findings offer valuable guidance for the deployment of observational instruments and subsequent analysis,as well as deepen the understanding of air−sea interactions in the Southern Ocean.展开更多
针对双馈式风电机组发电机前轴承劣化趋势问题,提出了一种新的组合建模方法对发电机前轴承健康度进行趋势预测。采用高斯混合模型(Gaussian Mixture Model, GMM)对机组运行工况进行辨识,并在各个子工况内分别建立基于极限学习机(Extreme...针对双馈式风电机组发电机前轴承劣化趋势问题,提出了一种新的组合建模方法对发电机前轴承健康度进行趋势预测。采用高斯混合模型(Gaussian Mixture Model, GMM)对机组运行工况进行辨识,并在各个子工况内分别建立基于极限学习机(Extreme Learning Machine, ELM)的发电机前轴承温度模型,将温度残差特征与前轴承振动信号时频域特征相融合,并计算前轴承健康度,提出基于注意力机制的双向长短期记忆(Bi-directional Long Short Term Memory, Bi-LSTM)神经网络对前轴承健康度进行建模并预测其趋势。实验结果表明:该组合建模方法具有较高的准确度和泛化能力。展开更多
基金supported by the National Natural Science Foundation of China (Grant Nos. 42406241,42325604,42227901)the Ministry of Science and Technology of China (No. 2021YFC2803304)+2 种基金the Program of Shanghai Academic/Technology Research Leader (22XD1403600)supported by the Swedish Research Council (Nos. 2020–03190 and 2024-04209)the Swedish Research Council for the Environment, Agricultural Sciences and Spatial Planning (No. 202400375)
文摘The oceanic mixed layer in the Southern Ocean is characterized by numerous fronts due to the stirring of freshwater influxes arising from ice melting.The interaction of these fronts with winds modulates the evolution of the mixed layer and affects atmosphere−ocean energy exchanges.However,the underlying mechanism behind the wind-front interaction remains obscure due to a lack of three-dimensional observations of the ocean,particularly in terms of velocities.To address this issue,this study investigates the dynamics of fronts within the mixed layer during a storm by employing a subset of the global submesoscale-permitting simulation,Northeast Weddell Sea Pre-SWOT Level-4 Hourly MITgcm LLC4320 Native Grid 2km Oceanographic Dataset(ROAM_MIZ).We first compare the ROAM_MIZ data to glider data to assess the performance of the model simulation and find that the ROAM_MIZ can,to a large degree,capture sub-mesoscale features within a mixed layer.Subsequent analyses based on a subset of ROAM_MIZ show that lateral density gradients within the mixed layer rapidly decrease during high winds associated with the storm.Down-front winds accelerate this process as the Ekman buoyancy transport responsible for enhancing the instability of the fronts is primarily dominated by horizontal baroclinic components.After the storm,the fronts strengthen again in the presence of weaker winds due to the frontogenesis by the larger-scale strain.Moreover,the non-geostrophic turbulence induces a modification of the relative vorticity,affecting the instability within the mixed layer.These findings offer valuable guidance for the deployment of observational instruments and subsequent analysis,as well as deepen the understanding of air−sea interactions in the Southern Ocean.
文摘针对双馈式风电机组发电机前轴承劣化趋势问题,提出了一种新的组合建模方法对发电机前轴承健康度进行趋势预测。采用高斯混合模型(Gaussian Mixture Model, GMM)对机组运行工况进行辨识,并在各个子工况内分别建立基于极限学习机(Extreme Learning Machine, ELM)的发电机前轴承温度模型,将温度残差特征与前轴承振动信号时频域特征相融合,并计算前轴承健康度,提出基于注意力机制的双向长短期记忆(Bi-directional Long Short Term Memory, Bi-LSTM)神经网络对前轴承健康度进行建模并预测其趋势。实验结果表明:该组合建模方法具有较高的准确度和泛化能力。
