Two-dimensional double-layer honeycomb(DLHC)materials are known for their diverse physical properties,but superconductivity has been a notably absent characteristic in this structure.We address this gap by investigati...Two-dimensional double-layer honeycomb(DLHC)materials are known for their diverse physical properties,but superconductivity has been a notably absent characteristic in this structure.We address this gap by investigating M_(2)N_(2)(M=Nb,Ta)with DLHC structure using first-principles calculations.Our results show that M_(2)N_(2)are stable and metallic,exhibiting superconducting behavior.Specifically,Nb_(2)N_(2)and Ta_(2)N_(2)display superconducting transition temperatures of 6.8 K and 8.8 K,respectively.Their electron-phonon coupling is predominantly driven by the coupling between metal d-orbitals and low-frequency metal-dominated vibration modes.Interestingly,two compounds also exhibit non-trivial band topology.Thus,M_(2)N_(2)are promising platforms for studying the interplay between topology and superconductivity and fill the gap in superconductivity research for DLHC materials.展开更多
液压系统压力传感器作为挖掘机自动控制系统的核心元件,其可靠性直接影响整机操控性能。针对复杂恶劣工况下压力传感器失效导致控制系统信号缺失的关键问题,提出一种基于深度学习的高精度压力数据实时预测方法。首先,基于37吨级挖掘机...液压系统压力传感器作为挖掘机自动控制系统的核心元件,其可靠性直接影响整机操控性能。针对复杂恶劣工况下压力传感器失效导致控制系统信号缺失的关键问题,提出一种基于深度学习的高精度压力数据实时预测方法。首先,基于37吨级挖掘机电液比例控制系统构建试验平台,采集实际挖装作业工况下多源传感器数据;其次,采用最大信息系数法进行特征相关性分析,将125维原始数据降维至10维有效特征,并通过卡尔曼滤波与标准化处理构建高质量数据集;进而设计基于注意力机制的特征权重分配模块,结合麻雀搜索算法(sparrow search algorithm,SSA)优化长短期记忆神经网络(long short term memory,LSTM)的超参数配置,构建SSA-LSTM-Attention融合预测模型。通过对比卷积神经网络(convolutional neural network,CNN)、循环神经网络(gate recurrent unit,GRU)、LSTM等典型预测模型的实验验证,该方法在关键压力数据预测中展现出显著优势。实验结果表明,相较于传统LSTM模型,SSA-LSTM-Attention模型的平均绝对误差和均方根误差分别降低54.45%和54.56%。研究证实所提方法能有效解决传感器失效工况下的数据补偿问题,为工程机械智能控制系统容错设计提供理论支撑。展开更多
基金Project supported by the National Natural Science Foundation of China(Grant Nos.12074213 and 11574108)the National Key R&D Program of China(Grant No.2022YFA1403103)+2 种基金the Major Basic Program of Natural Science Foundation of Shandong Province(Grant No.ZR2021ZD01)the Natural Science Foundation of Shandong Province(Grant No.ZR2023MA082)the Project of Introduction and Cultivation for Young Innovative Talents in Colleges and Universities of Shandong Province。
文摘Two-dimensional double-layer honeycomb(DLHC)materials are known for their diverse physical properties,but superconductivity has been a notably absent characteristic in this structure.We address this gap by investigating M_(2)N_(2)(M=Nb,Ta)with DLHC structure using first-principles calculations.Our results show that M_(2)N_(2)are stable and metallic,exhibiting superconducting behavior.Specifically,Nb_(2)N_(2)and Ta_(2)N_(2)display superconducting transition temperatures of 6.8 K and 8.8 K,respectively.Their electron-phonon coupling is predominantly driven by the coupling between metal d-orbitals and low-frequency metal-dominated vibration modes.Interestingly,two compounds also exhibit non-trivial band topology.Thus,M_(2)N_(2)are promising platforms for studying the interplay between topology and superconductivity and fill the gap in superconductivity research for DLHC materials.
文摘液压系统压力传感器作为挖掘机自动控制系统的核心元件,其可靠性直接影响整机操控性能。针对复杂恶劣工况下压力传感器失效导致控制系统信号缺失的关键问题,提出一种基于深度学习的高精度压力数据实时预测方法。首先,基于37吨级挖掘机电液比例控制系统构建试验平台,采集实际挖装作业工况下多源传感器数据;其次,采用最大信息系数法进行特征相关性分析,将125维原始数据降维至10维有效特征,并通过卡尔曼滤波与标准化处理构建高质量数据集;进而设计基于注意力机制的特征权重分配模块,结合麻雀搜索算法(sparrow search algorithm,SSA)优化长短期记忆神经网络(long short term memory,LSTM)的超参数配置,构建SSA-LSTM-Attention融合预测模型。通过对比卷积神经网络(convolutional neural network,CNN)、循环神经网络(gate recurrent unit,GRU)、LSTM等典型预测模型的实验验证,该方法在关键压力数据预测中展现出显著优势。实验结果表明,相较于传统LSTM模型,SSA-LSTM-Attention模型的平均绝对误差和均方根误差分别降低54.45%和54.56%。研究证实所提方法能有效解决传感器失效工况下的数据补偿问题,为工程机械智能控制系统容错设计提供理论支撑。
