Chemical hydrogen storage in organic materials is a promising method thanks to its high storage density,reversibility,and safety.However,the dehydrogenation process of organic materials requires high temperatures due ...Chemical hydrogen storage in organic materials is a promising method thanks to its high storage density,reversibility,and safety.However,the dehydrogenation process of organic materials requires high temperatures due to their unfavorable thermodynamic properties.This study proposes a strategy to design a new type of hydrogen storage materials,i.e.,alkali metal pyridinolate/piperidinolate pairs,by combining the effects of a heteroatom and an alkali metal in one molecule to achieve suitable dehydrogenation thermodynamics along with high hydrogen storage capacities.These air-stable compounds can be synthesized using low-cost reactants and water as a green solvent.Thermodynamic predictions indicate that enthalpy changes of dehydrogenation(ΔH_(d))can be significantly reduced to the optimal range for efficient hydrogen release,exemplified by lithium 2-piperidinolate with a 5.6 wt%hydrogen capacity and a suitableΔH_(d)of 32.2 kJ/mol-H_(2).Experimental results obtained using sodium systems validate the computational predictions,demonstrating reversible hydrogen storage even below 100℃.The superior hydrogen desorption performance of alkali metal piperidinolates could be attributed to their suitableΔH_(d)induced by the combined effect of ring nitrogen and metal substitution on their structures.This study not only reports new low-cost hydrogen storage materials but also provides a rational design strategy for developing metalorganic compounds possessing high hydrogen capacities and suitable thermodynamics for efficient hydrogen storage.展开更多
Transmission line faults pose a significant threat to power system resilience,underscoring the need for accurate and rapid fault identification to facilitate proper resource monitoring,economic loss prevention,and bla...Transmission line faults pose a significant threat to power system resilience,underscoring the need for accurate and rapid fault identification to facilitate proper resource monitoring,economic loss prevention,and blackout avoidance.Extreme learning machine(ELM)offers a compelling solution for rapid classification,achieving network training in a single epoch.Leveraging the Internet of Things(IoT)and the virtual instrumentation capabilities of LabVIEW,ELM can enable the swift and precise identification of transmission line faults.This paper presents a regularized radial basis function(RBF)ELM-based fault detection and classification system for transmission lines,utilizing a LabVIEW based virtual phasor measurement unit(PMU)and IoT sensors.The transmission line fault is identified using the phaselet algorithm applied to the phase current acquired from the virtual PMU.Classification is then performed using the ELM algorithm.The proposed methodology is validated in real-time on a practical transmission line,achieving an accuracy of 99.46%.This has the potential to significantly influence future fault detection strategies incorporating virtual PMUs and machine learning.展开更多
The gap between the projected urban areas in the current trend(UAC)and those in the sustainable scenario(UAS)is a critical factor in understanding whether cities can fulfill the requirements of sustainable development...The gap between the projected urban areas in the current trend(UAC)and those in the sustainable scenario(UAS)is a critical factor in understanding whether cities can fulfill the requirements of sustainable development.However,there is a paucity of knowledge on this cutting-edge topic.Given the extensive and rapid urbanization in the United States(U.S.)over the past two centuries,accurately measuring this gap between UAS and UAC is of critical importance for advancing future sustainable urban development,as well as having significant global implications.This study finds that although the 740 U.S.cities have a large UAC in 2100,these cities will encom pass a significant gap from UAC to UAS(approximately 165,000 km2),accounting for 30%UAC at that time.The study also reveals the spatio-temporal heterogeneity of the gap.The gap initially increases before reaching a inflection point in 2090,and it disparates greatly from−100%to 240%at city level.While cities in the Northwestern U.S.maintain UAC that exceeds UAS from 2020 to 2100,cities in other regions shift from UAC that exceeds UAS to UAC that falls short of UAS.Filling the gap without additional urban growth planning could lead to a reduction of crop production ranging from 0.3%to 3%and a 0.68%loss of biomass.Hence,dynamic and forward-looking urban planning is essential for addressing the challenges of sustainable development posed by urbanization,both within the U.S.and globally.展开更多
目的 在高分辨率遥感影像语义分割任务中,仅利用可见光图像很难区分光谱特征相似的区域(如草坪和树、道路和建筑物),高程信息的引入可以显著改善分类结果。然而,可见光图像与高程数据的特征分布差异较大,简单的级联或相加的融合方式不...目的 在高分辨率遥感影像语义分割任务中,仅利用可见光图像很难区分光谱特征相似的区域(如草坪和树、道路和建筑物),高程信息的引入可以显著改善分类结果。然而,可见光图像与高程数据的特征分布差异较大,简单的级联或相加的融合方式不能有效处理两种模态融合时的噪声,使得融合效果不佳。因此如何有效地融合多模态特征成为遥感语义分割的关键问题。针对这一问题,本文提出了一个多源特征自适应融合模型。方法 通过像素的目标类别以及上下文信息动态融合模态特征,减弱融合噪声影响,有效利用多模态数据的互补信息。该模型主要包含3个部分:双编码器负责提取光谱和高程模态的特征;模态自适应融合模块协同处理多模态特征,依据像素的目标类别以及上下文信息动态地利用高程信息强化光谱特征,使得网络可以针对特定的对象类别或者特定的空间位置来选择特定模态网络的特征信息;全局上下文聚合模块,从空间和通道角度进行全局上下文建模以获得更丰富的特征表示。结果 对实验结果进行定性、定量相结合的评价。定性结果中,本文算法获取的分割结果更加精细化。定量结果中,在ISPRS(International Society for Photogrammetry and Remote Sensing)Vaihingen和GID(Gaofen Image Dataset)数据集上对本文模型进行评估,分别达到了90.77%、82.1%的总体精度。与DeepLab V3+、PSPNet(pyramid scene parsing network)等算法相比,本文算法明显更优。结论 实验结果表明,本文提出的多源特征自适应融合网络可以有效地进行模态特征融合,更加高效地建模全局上下文关系,可以广泛应用于遥感领域。展开更多
基金partially supported by the National Key R&D Program of China(2023YFE0198900)support provided by the National Natural Science Foundation of China(52171226,22309174)。
文摘Chemical hydrogen storage in organic materials is a promising method thanks to its high storage density,reversibility,and safety.However,the dehydrogenation process of organic materials requires high temperatures due to their unfavorable thermodynamic properties.This study proposes a strategy to design a new type of hydrogen storage materials,i.e.,alkali metal pyridinolate/piperidinolate pairs,by combining the effects of a heteroatom and an alkali metal in one molecule to achieve suitable dehydrogenation thermodynamics along with high hydrogen storage capacities.These air-stable compounds can be synthesized using low-cost reactants and water as a green solvent.Thermodynamic predictions indicate that enthalpy changes of dehydrogenation(ΔH_(d))can be significantly reduced to the optimal range for efficient hydrogen release,exemplified by lithium 2-piperidinolate with a 5.6 wt%hydrogen capacity and a suitableΔH_(d)of 32.2 kJ/mol-H_(2).Experimental results obtained using sodium systems validate the computational predictions,demonstrating reversible hydrogen storage even below 100℃.The superior hydrogen desorption performance of alkali metal piperidinolates could be attributed to their suitableΔH_(d)induced by the combined effect of ring nitrogen and metal substitution on their structures.This study not only reports new low-cost hydrogen storage materials but also provides a rational design strategy for developing metalorganic compounds possessing high hydrogen capacities and suitable thermodynamics for efficient hydrogen storage.
基金supported by the Experimental-Demonstration project PN-IV-P7-7.1-PED-2024-0567(Improving the Fuel Cell Hybrid Electric Vehicle Drivetrain by Implementing a Novel Optimal Real-Time Power Management Strategy),contract no.58PED(N.B.)the APC was funded by S.C.ECAI CONFERENCE S.R.Lsupported by Science and Engineering Research Board,India with grant number ECR/2017/000812.
文摘Transmission line faults pose a significant threat to power system resilience,underscoring the need for accurate and rapid fault identification to facilitate proper resource monitoring,economic loss prevention,and blackout avoidance.Extreme learning machine(ELM)offers a compelling solution for rapid classification,achieving network training in a single epoch.Leveraging the Internet of Things(IoT)and the virtual instrumentation capabilities of LabVIEW,ELM can enable the swift and precise identification of transmission line faults.This paper presents a regularized radial basis function(RBF)ELM-based fault detection and classification system for transmission lines,utilizing a LabVIEW based virtual phasor measurement unit(PMU)and IoT sensors.The transmission line fault is identified using the phaselet algorithm applied to the phase current acquired from the virtual PMU.Classification is then performed using the ELM algorithm.The proposed methodology is validated in real-time on a practical transmission line,achieving an accuracy of 99.46%.This has the potential to significantly influence future fault detection strategies incorporating virtual PMUs and machine learning.
基金supported by the National Natural Science Foun-dation of China(Grants No.42330103,42271469)the Ningbo Science and Technology Bureau(Grant No.2022Z081).
文摘The gap between the projected urban areas in the current trend(UAC)and those in the sustainable scenario(UAS)is a critical factor in understanding whether cities can fulfill the requirements of sustainable development.However,there is a paucity of knowledge on this cutting-edge topic.Given the extensive and rapid urbanization in the United States(U.S.)over the past two centuries,accurately measuring this gap between UAS and UAC is of critical importance for advancing future sustainable urban development,as well as having significant global implications.This study finds that although the 740 U.S.cities have a large UAC in 2100,these cities will encom pass a significant gap from UAC to UAS(approximately 165,000 km2),accounting for 30%UAC at that time.The study also reveals the spatio-temporal heterogeneity of the gap.The gap initially increases before reaching a inflection point in 2090,and it disparates greatly from−100%to 240%at city level.While cities in the Northwestern U.S.maintain UAC that exceeds UAS from 2020 to 2100,cities in other regions shift from UAC that exceeds UAS to UAC that falls short of UAS.Filling the gap without additional urban growth planning could lead to a reduction of crop production ranging from 0.3%to 3%and a 0.68%loss of biomass.Hence,dynamic and forward-looking urban planning is essential for addressing the challenges of sustainable development posed by urbanization,both within the U.S.and globally.
文摘目的 在高分辨率遥感影像语义分割任务中,仅利用可见光图像很难区分光谱特征相似的区域(如草坪和树、道路和建筑物),高程信息的引入可以显著改善分类结果。然而,可见光图像与高程数据的特征分布差异较大,简单的级联或相加的融合方式不能有效处理两种模态融合时的噪声,使得融合效果不佳。因此如何有效地融合多模态特征成为遥感语义分割的关键问题。针对这一问题,本文提出了一个多源特征自适应融合模型。方法 通过像素的目标类别以及上下文信息动态融合模态特征,减弱融合噪声影响,有效利用多模态数据的互补信息。该模型主要包含3个部分:双编码器负责提取光谱和高程模态的特征;模态自适应融合模块协同处理多模态特征,依据像素的目标类别以及上下文信息动态地利用高程信息强化光谱特征,使得网络可以针对特定的对象类别或者特定的空间位置来选择特定模态网络的特征信息;全局上下文聚合模块,从空间和通道角度进行全局上下文建模以获得更丰富的特征表示。结果 对实验结果进行定性、定量相结合的评价。定性结果中,本文算法获取的分割结果更加精细化。定量结果中,在ISPRS(International Society for Photogrammetry and Remote Sensing)Vaihingen和GID(Gaofen Image Dataset)数据集上对本文模型进行评估,分别达到了90.77%、82.1%的总体精度。与DeepLab V3+、PSPNet(pyramid scene parsing network)等算法相比,本文算法明显更优。结论 实验结果表明,本文提出的多源特征自适应融合网络可以有效地进行模态特征融合,更加高效地建模全局上下文关系,可以广泛应用于遥感领域。
