We see significant progress in the preparation of nanoreactors that host incompatible catalysts for cascade reactions,yet their preparation typically requires specialized,multistep synthetic routes.Here,we utilize cor...We see significant progress in the preparation of nanoreactors that host incompatible catalysts for cascade reactions,yet their preparation typically requires specialized,multistep synthetic routes.Here,we utilize core-satellite microparticles as a versatile strategy for the site isolation of catalysts in core and satellites.The core-satellite clusters are prepared by mixing specific amounts of negatively charged core microparticles(e.g.,acid catalyst)with positively charged satellites(e.g.,base catalyst).Core and satellite are made from incompatible polymers allowing solvent-annealing of clusters into different morphologies(raspberry,patchy,and core-shell)while maintaining site isolation of the catalysts.The core-satellite microparticles show very good catalytic activity in a model one-pot acid/base cascade reaction with subtle differences regarding particle morphology.Raspberry microparticles thereby demonstrated the highest reaction rate and yield,likely due to an alleviated diffusion pathway for the reactants.Finally,we show that this colloidal engineering strategy can be extended to multifunctional microparticles suitable to perform multistep cascade reactions in one pot.展开更多
This paper proposes an engineering approach to examine the stability margin of cascade circuits. The proposed measurement injects an external perturbation current into the cascade interface and measures the load-side ...This paper proposes an engineering approach to examine the stability margin of cascade circuits. The proposed measurement injects an external perturbation current into the cascade interface and measures the load-side response current. The system stability margin can be figured out by comparing the magnitudes of the perturbation current and the response current. The proposed measurement is strictly derived in theory, implemented with details, and demonstrated by experiment.展开更多
Hydropower cascade development have brought considerable economic benefits to flood control, irrigation, power supply, but the positive or negative impact for the ecological environment at the same time. The scope of ...Hydropower cascade development have brought considerable economic benefits to flood control, irrigation, power supply, but the positive or negative impact for the ecological environment at the same time. The scope of the environmental impact of cascade hydropower development projects is larger, lasted longer, have influential groupment, systematicness, cumulativity, sweep property and the potential. This paper expounds the ecological status survey method, analyzing the influence of hydropowcr cascade development on water temperature, aquatic plants and so on. In order to realize the harmonious development of hydropower and ecology, taking the ＂ecological regulation＂the measures in the operation of cascade hydropower stations.展开更多
针对变电站建筑物屋面工程缺陷检测效率较低及检测精确度较差的问题,提出一种基于改进YOLOv5(you only look once version 5)的变电站屋面工程缺陷检测算法。首先,对图像进行预处理,减轻外界噪声给检测效果带来的影响。其次,在网络骨干...针对变电站建筑物屋面工程缺陷检测效率较低及检测精确度较差的问题,提出一种基于改进YOLOv5(you only look once version 5)的变电站屋面工程缺陷检测算法。首先,对图像进行预处理,减轻外界噪声给检测效果带来的影响。其次,在网络骨干中引入改进自注意力机制,提高计算效率,用多头自注意力层替换YOLOv5网络骨干末端的卷积层,使网络能够更好地捕捉全局关联信息。最后,在检测部分增加跨层加权级联结构,将浅层中缺陷的边缘信息、轮廓信息融入到深层特征中,提高网络对缺陷边界回归的精确度。实验结果表明,本文提出的改进YOLOv5变电站屋面工程缺陷检测算法对保温层、隔离层、隔汽层、防水层和找平层这5类工序的缺陷检测的平均精度均值达到了93.2%,每秒帧数达到163.5帧/s,解决了实际工程环境中出现的变电站屋面工程缺陷分布不均衡和目标多尺度变化的问题,对比其他同类算法拥有更好的精确度和实时性。展开更多
对航空发动机进行实时状态监测和健康管理可以有效降低发动机故障风险,确保飞机飞行安全。准确预测航空发动机的剩余寿命是有效监测发动机运行状态的一种重要手段,其中长短期记忆(long-short term me-mory,LSTM)网络常被使用。但由于航...对航空发动机进行实时状态监测和健康管理可以有效降低发动机故障风险,确保飞机飞行安全。准确预测航空发动机的剩余寿命是有效监测发动机运行状态的一种重要手段,其中长短期记忆(long-short term me-mory,LSTM)网络常被使用。但由于航空发动机复杂的机械结构与运行模式,使用传统的LSTM网络对航空发动机的剩余寿命进行单次预测后,所得预测结果的准确率不足以满足其寿命预测的精度要求。基于LSTM网络的广泛使用以及它对时间序列数据的有效预测能力,考虑到采用多级预测的方法能够有效降低预测误差,提出了一种新型的可自动扩展的长短期记忆(automatically expandable LSTM,AELSTM)预测模型。AELSTM模型依托多个子模块逐级连接的网络结构,不断地提取前一级模块的输出误差作为后一级模块的训练值,形成了误差的多级预测机制,有效降低了模型的预测误差,提升了预测结果的准确性。基于美国国家航空航天局发布的C-MAPSS数据集的四个子集对AELSTM模型的预测效果进行了测试,实验结果表明,与传统的LSTM网络相比,AELSTM模型在四个子集上的均方根误差平均减少了95.44%,同时它的预测效果也优于现有的一些先进算法。实验充分验证了AELSTM模型在提升航空发动机剩余寿命预测准确度方面的有效性及优势。展开更多
基金a result of a generous grant from the German Research Foundation(DFG)through the Emmy Noether Program(grant number no.GR 5075/2-1).
文摘We see significant progress in the preparation of nanoreactors that host incompatible catalysts for cascade reactions,yet their preparation typically requires specialized,multistep synthetic routes.Here,we utilize core-satellite microparticles as a versatile strategy for the site isolation of catalysts in core and satellites.The core-satellite clusters are prepared by mixing specific amounts of negatively charged core microparticles(e.g.,acid catalyst)with positively charged satellites(e.g.,base catalyst).Core and satellite are made from incompatible polymers allowing solvent-annealing of clusters into different morphologies(raspberry,patchy,and core-shell)while maintaining site isolation of the catalysts.The core-satellite microparticles show very good catalytic activity in a model one-pot acid/base cascade reaction with subtle differences regarding particle morphology.Raspberry microparticles thereby demonstrated the highest reaction rate and yield,likely due to an alleviated diffusion pathway for the reactants.Finally,we show that this colloidal engineering strategy can be extended to multifunctional microparticles suitable to perform multistep cascade reactions in one pot.
文摘This paper proposes an engineering approach to examine the stability margin of cascade circuits. The proposed measurement injects an external perturbation current into the cascade interface and measures the load-side response current. The system stability margin can be figured out by comparing the magnitudes of the perturbation current and the response current. The proposed measurement is strictly derived in theory, implemented with details, and demonstrated by experiment.
文摘Hydropower cascade development have brought considerable economic benefits to flood control, irrigation, power supply, but the positive or negative impact for the ecological environment at the same time. The scope of the environmental impact of cascade hydropower development projects is larger, lasted longer, have influential groupment, systematicness, cumulativity, sweep property and the potential. This paper expounds the ecological status survey method, analyzing the influence of hydropowcr cascade development on water temperature, aquatic plants and so on. In order to realize the harmonious development of hydropower and ecology, taking the ＂ecological regulation＂the measures in the operation of cascade hydropower stations.
文摘针对变电站建筑物屋面工程缺陷检测效率较低及检测精确度较差的问题,提出一种基于改进YOLOv5(you only look once version 5)的变电站屋面工程缺陷检测算法。首先,对图像进行预处理,减轻外界噪声给检测效果带来的影响。其次,在网络骨干中引入改进自注意力机制,提高计算效率,用多头自注意力层替换YOLOv5网络骨干末端的卷积层,使网络能够更好地捕捉全局关联信息。最后,在检测部分增加跨层加权级联结构,将浅层中缺陷的边缘信息、轮廓信息融入到深层特征中,提高网络对缺陷边界回归的精确度。实验结果表明,本文提出的改进YOLOv5变电站屋面工程缺陷检测算法对保温层、隔离层、隔汽层、防水层和找平层这5类工序的缺陷检测的平均精度均值达到了93.2%,每秒帧数达到163.5帧/s,解决了实际工程环境中出现的变电站屋面工程缺陷分布不均衡和目标多尺度变化的问题,对比其他同类算法拥有更好的精确度和实时性。
文摘对航空发动机进行实时状态监测和健康管理可以有效降低发动机故障风险,确保飞机飞行安全。准确预测航空发动机的剩余寿命是有效监测发动机运行状态的一种重要手段,其中长短期记忆(long-short term me-mory,LSTM)网络常被使用。但由于航空发动机复杂的机械结构与运行模式,使用传统的LSTM网络对航空发动机的剩余寿命进行单次预测后,所得预测结果的准确率不足以满足其寿命预测的精度要求。基于LSTM网络的广泛使用以及它对时间序列数据的有效预测能力,考虑到采用多级预测的方法能够有效降低预测误差,提出了一种新型的可自动扩展的长短期记忆(automatically expandable LSTM,AELSTM)预测模型。AELSTM模型依托多个子模块逐级连接的网络结构,不断地提取前一级模块的输出误差作为后一级模块的训练值,形成了误差的多级预测机制,有效降低了模型的预测误差,提升了预测结果的准确性。基于美国国家航空航天局发布的C-MAPSS数据集的四个子集对AELSTM模型的预测效果进行了测试,实验结果表明,与传统的LSTM网络相比,AELSTM模型在四个子集上的均方根误差平均减少了95.44%,同时它的预测效果也优于现有的一些先进算法。实验充分验证了AELSTM模型在提升航空发动机剩余寿命预测准确度方面的有效性及优势。
