There is instability in the distributed energy storage cloud group end region on the power grid side.In order to avoid large-scale fluctuating charging and discharging in the power grid environment and make the capaci...There is instability in the distributed energy storage cloud group end region on the power grid side.In order to avoid large-scale fluctuating charging and discharging in the power grid environment and make the capacitor components showa continuous and stable charging and discharging state,a hierarchical time-sharing configuration algorithm of distributed energy storage cloud group end region on the power grid side based on multi-scale and multi feature convolution neural network is proposed.Firstly,a voltage stability analysis model based onmulti-scale and multi feature convolution neural network is constructed,and the multi-scale and multi feature convolution neural network is optimized based on Self-OrganizingMaps(SOM)algorithm to analyze the voltage stability of the cloud group end region of distributed energy storage on the grid side under the framework of credibility.According to the optimal scheduling objectives and network size,the distributed robust optimal configuration control model is solved under the framework of coordinated optimal scheduling at multiple time scales;Finally,the time series characteristics of regional power grid load and distributed generation are analyzed.According to the regional hierarchical time-sharing configuration model of“cloud”,“group”and“end”layer,the grid side distributed energy storage cloud group end regional hierarchical time-sharing configuration algorithm is realized.The experimental results show that after applying this algorithm,the best grid side distributed energy storage configuration scheme can be determined,and the stability of grid side distributed energy storage cloud group end region layered timesharing configuration can be improved.展开更多
Accurate measurements of the associated vegetation phenological dynamics are crucial for understanding the relationship between climate change and terrestrial ecosystems. However, at present, most vegetation phenologi...Accurate measurements of the associated vegetation phenological dynamics are crucial for understanding the relationship between climate change and terrestrial ecosystems. However, at present, most vegetation phenological calculations are based on a single algorithm or method. Because of the spatial, temporal, and ecological complexity of the vegetation growth processes, a single algorithm or method for monitoring all these processes has been indicated to be elusive. Therefore, in this study, from the perspective of plant growth characteristics, we established a method to remotely determine the start of the growth season(SOG) and the end of the growth season(EOG), in which the maximum relative change rate of the normalized difference vegetation index(NDVI) corresponds to the SOG, and the next minimum absolute change rate of the NDVI corresponds to the EOG. Taking the Three-River Headwaters Region in 2000–2013 as an example, we ascertained the spatiotemporal and vertical characteristics of its vegetation phenological changes. Then, in contrast to the actual air temperature data, observed data and other related studies, we found that the SOG and EOG calculated by the proposed method is closer to the time corresponding to the air temperature, and the trends of the SOG and EOG calculated by the proposed method are in good agreement with other relevant studies. Meantime, the error of the SOG between the calculated and observed in this study is smaller than that in other studies.展开更多
近几年,基于Transformer的预训练模型展现了强大的模态表征能力,促使了多模态的下游任务(如图像描述生成任务)正朝着完全端到端范式的趋势所转变,并且能够使得模型获得更好的性能以及更快的推理速度.然而,该技术所提取的网格型视觉特征...近几年,基于Transformer的预训练模型展现了强大的模态表征能力,促使了多模态的下游任务(如图像描述生成任务)正朝着完全端到端范式的趋势所转变,并且能够使得模型获得更好的性能以及更快的推理速度.然而,该技术所提取的网格型视觉特征中缺乏区域型的视觉信息,从而导致模型对对象内容的描述不精确.因此,预训练模型在图像描述生成任务上的适用性在很大程度上仍有待探索.针对这一问题,提出一种基于视觉区域聚合与双向协作学习的端到端图像描述生成方法(visual region aggregation and dual-level collaboration,VRADC).为了学习到区域型的视觉信息,设计了一种视觉区域聚合模块,将有相似语义的网格特征聚合在一起形成紧凑的视觉区域表征.接着,双向协作模块利用交叉注意力机制从两种视觉特征中学习到更加有代表性的语义信息,进而指导模型生成更加细粒度的图像描述文本.基于MSCOCO和Flickr30k两个数据集的实验结果表明,所提的VRADC方法能够大幅度地提升图像描述生成的质量,实现了最先进的性能.展开更多
基金supported by State Grid Corporation Limited Science and Technology Project Funding(Contract No.SGCQSQ00YJJS2200380).
文摘There is instability in the distributed energy storage cloud group end region on the power grid side.In order to avoid large-scale fluctuating charging and discharging in the power grid environment and make the capacitor components showa continuous and stable charging and discharging state,a hierarchical time-sharing configuration algorithm of distributed energy storage cloud group end region on the power grid side based on multi-scale and multi feature convolution neural network is proposed.Firstly,a voltage stability analysis model based onmulti-scale and multi feature convolution neural network is constructed,and the multi-scale and multi feature convolution neural network is optimized based on Self-OrganizingMaps(SOM)algorithm to analyze the voltage stability of the cloud group end region of distributed energy storage on the grid side under the framework of credibility.According to the optimal scheduling objectives and network size,the distributed robust optimal configuration control model is solved under the framework of coordinated optimal scheduling at multiple time scales;Finally,the time series characteristics of regional power grid load and distributed generation are analyzed.According to the regional hierarchical time-sharing configuration model of“cloud”,“group”and“end”layer,the grid side distributed energy storage cloud group end regional hierarchical time-sharing configuration algorithm is realized.The experimental results show that after applying this algorithm,the best grid side distributed energy storage configuration scheme can be determined,and the stability of grid side distributed energy storage cloud group end region layered timesharing configuration can be improved.
基金supported by National Natural Science Foundation of China (Grant No. 41801099)
文摘Accurate measurements of the associated vegetation phenological dynamics are crucial for understanding the relationship between climate change and terrestrial ecosystems. However, at present, most vegetation phenological calculations are based on a single algorithm or method. Because of the spatial, temporal, and ecological complexity of the vegetation growth processes, a single algorithm or method for monitoring all these processes has been indicated to be elusive. Therefore, in this study, from the perspective of plant growth characteristics, we established a method to remotely determine the start of the growth season(SOG) and the end of the growth season(EOG), in which the maximum relative change rate of the normalized difference vegetation index(NDVI) corresponds to the SOG, and the next minimum absolute change rate of the NDVI corresponds to the EOG. Taking the Three-River Headwaters Region in 2000–2013 as an example, we ascertained the spatiotemporal and vertical characteristics of its vegetation phenological changes. Then, in contrast to the actual air temperature data, observed data and other related studies, we found that the SOG and EOG calculated by the proposed method is closer to the time corresponding to the air temperature, and the trends of the SOG and EOG calculated by the proposed method are in good agreement with other relevant studies. Meantime, the error of the SOG between the calculated and observed in this study is smaller than that in other studies.
文摘近几年,基于Transformer的预训练模型展现了强大的模态表征能力,促使了多模态的下游任务(如图像描述生成任务)正朝着完全端到端范式的趋势所转变,并且能够使得模型获得更好的性能以及更快的推理速度.然而,该技术所提取的网格型视觉特征中缺乏区域型的视觉信息,从而导致模型对对象内容的描述不精确.因此,预训练模型在图像描述生成任务上的适用性在很大程度上仍有待探索.针对这一问题,提出一种基于视觉区域聚合与双向协作学习的端到端图像描述生成方法(visual region aggregation and dual-level collaboration,VRADC).为了学习到区域型的视觉信息,设计了一种视觉区域聚合模块,将有相似语义的网格特征聚合在一起形成紧凑的视觉区域表征.接着,双向协作模块利用交叉注意力机制从两种视觉特征中学习到更加有代表性的语义信息,进而指导模型生成更加细粒度的图像描述文本.基于MSCOCO和Flickr30k两个数据集的实验结果表明,所提的VRADC方法能够大幅度地提升图像描述生成的质量,实现了最先进的性能.
