In wireless sensor network, the primary design is to save the energy consumption as much as possible while achieving the given task. Most of recent researches works have only focused on the individual layer issues and...In wireless sensor network, the primary design is to save the energy consumption as much as possible while achieving the given task. Most of recent researches works have only focused on the individual layer issues and ignore the importance of inter working between different layers in a sensor network. In this paper, we use a cross-layer approach to propose an energy-efficient and extending the life time of the sensor network. This protocol which uses routing in the network layer, and the data scheduling in MAC layer. The main ob-jective of this paper is to provide a possible and flexible approach to solve the conflicts between the require-ments of large scale, long life-time, and multi-purpose wireless sensor networks. This OEEXLM module gives better performance compared to all other existing protocols. The performance of OEEXLM module compared with S-MAC and directed diffusion protocol.展开更多
滚动轴承是机械设备中的常见关键部件,准确预测其剩余使用寿命对机械设备的安全稳定运行至关重要。针对目前轴承寿命预测存在的轴承退化特征不明显、模型泛化能力差以及数据长期依赖关系难以捕捉的问题,提出基于时频域信号优化器(Time-F...滚动轴承是机械设备中的常见关键部件,准确预测其剩余使用寿命对机械设备的安全稳定运行至关重要。针对目前轴承寿命预测存在的轴承退化特征不明显、模型泛化能力差以及数据长期依赖关系难以捕捉的问题,提出基于时频域信号优化器(Time-Frequency domain signal Ratio Optimizer,TFRO)的多重膨胀多核时间卷积网络(Multi inflated Multi kernel Time Convolutional Network,Mi-MkTCN)模型。TFRO优化器为了精准记忆重要信息,在每一个时间节点上,将过去信息和当前信息重组,其中过去信息中的重要的时频域特征经过了有比例的分配。Mi-MkTCN利用多重膨胀确保重要特征不丢失,再利用多核时间卷积网络实现对不同尺度特征的提取。最终的消融对比实验验证了改进方法的有效性,模型的平均绝对误差、均方误差及均方根误差指标分别为0.00145、0.05069和0.12045。实验结果表明,所提方法显著提升了轴承剩余使用寿命的预测精度,为轴承剩余使用寿命预测提供了高精度、高鲁棒性的解决方案。展开更多
Wireless sensor networks are provided with a limited source of power. The lifetime of such networks is an overwhelming matter in most network applications. This lifetime depends strongly on how efficiently such energy...Wireless sensor networks are provided with a limited source of power. The lifetime of such networks is an overwhelming matter in most network applications. This lifetime depends strongly on how efficiently such energy is distributed over the nodes especially during transmitting and receiving data. Each node may route messages to destination nodes either through short hops or long hops. Optimizing the length of these hops may save energy, and therefore extend the lifetime of WSNs. In this paper, we propose a theorem to optimize the hop’s length so to make WSN power consumption minimal. The theorem establishes a simple condition on hop’s length range. Computer simulation when performing such condition on Mica2 sensors and Mica2dot sensors reveals good performance regarding WSNs energy consumption.展开更多
针对航空发动机剩余使用寿命(RUL)预测方法空间特征提取不充分、时间特征利用不充分,导致RUL预测准确性较低的问题,提出一种融合注意力机制的时空图卷积网络模型GCNBL-A3T(Graph Convolutional Network combined with Bidirectional Lon...针对航空发动机剩余使用寿命(RUL)预测方法空间特征提取不充分、时间特征利用不充分,导致RUL预测准确性较低的问题,提出一种融合注意力机制的时空图卷积网络模型GCNBL-A3T(Graph Convolutional Network combined with Bidirectional Long short-term memory and ATTenTion mechanism)。首先,使用一维卷积神经网络(1D-CNN)提取初始特征;其次,依次使用图卷积网络(GCN)和双向长短期记忆(Bi-LSTM)网络分别提取空间特征和时间特征;再次,利用自注意力机制处理特征并重新分配权重;最后,输入全连接网络获得RUL预测结果。使用商用模块化航空推进系统仿真(C-MAPSS)数据集验证所提模型的有效性。实验结果显示,与先进模型相比,所提模型的Score分数在3个数据子集上取得最小值,在1个数据子集上取得次小值;均方根误差(RMSE)在1个数据子集上取得最小值,在3个数据子集上取得次小值。消融实验结果也验证了所提模型的各模块能有效提升预测精度。展开更多
文摘In wireless sensor network, the primary design is to save the energy consumption as much as possible while achieving the given task. Most of recent researches works have only focused on the individual layer issues and ignore the importance of inter working between different layers in a sensor network. In this paper, we use a cross-layer approach to propose an energy-efficient and extending the life time of the sensor network. This protocol which uses routing in the network layer, and the data scheduling in MAC layer. The main ob-jective of this paper is to provide a possible and flexible approach to solve the conflicts between the require-ments of large scale, long life-time, and multi-purpose wireless sensor networks. This OEEXLM module gives better performance compared to all other existing protocols. The performance of OEEXLM module compared with S-MAC and directed diffusion protocol.
文摘滚动轴承是机械设备中的常见关键部件,准确预测其剩余使用寿命对机械设备的安全稳定运行至关重要。针对目前轴承寿命预测存在的轴承退化特征不明显、模型泛化能力差以及数据长期依赖关系难以捕捉的问题,提出基于时频域信号优化器(Time-Frequency domain signal Ratio Optimizer,TFRO)的多重膨胀多核时间卷积网络(Multi inflated Multi kernel Time Convolutional Network,Mi-MkTCN)模型。TFRO优化器为了精准记忆重要信息,在每一个时间节点上,将过去信息和当前信息重组,其中过去信息中的重要的时频域特征经过了有比例的分配。Mi-MkTCN利用多重膨胀确保重要特征不丢失,再利用多核时间卷积网络实现对不同尺度特征的提取。最终的消融对比实验验证了改进方法的有效性,模型的平均绝对误差、均方误差及均方根误差指标分别为0.00145、0.05069和0.12045。实验结果表明,所提方法显著提升了轴承剩余使用寿命的预测精度,为轴承剩余使用寿命预测提供了高精度、高鲁棒性的解决方案。
文摘Wireless sensor networks are provided with a limited source of power. The lifetime of such networks is an overwhelming matter in most network applications. This lifetime depends strongly on how efficiently such energy is distributed over the nodes especially during transmitting and receiving data. Each node may route messages to destination nodes either through short hops or long hops. Optimizing the length of these hops may save energy, and therefore extend the lifetime of WSNs. In this paper, we propose a theorem to optimize the hop’s length so to make WSN power consumption minimal. The theorem establishes a simple condition on hop’s length range. Computer simulation when performing such condition on Mica2 sensors and Mica2dot sensors reveals good performance regarding WSNs energy consumption.
文摘针对航空发动机剩余使用寿命(RUL)预测方法空间特征提取不充分、时间特征利用不充分,导致RUL预测准确性较低的问题,提出一种融合注意力机制的时空图卷积网络模型GCNBL-A3T(Graph Convolutional Network combined with Bidirectional Long short-term memory and ATTenTion mechanism)。首先,使用一维卷积神经网络(1D-CNN)提取初始特征;其次,依次使用图卷积网络(GCN)和双向长短期记忆(Bi-LSTM)网络分别提取空间特征和时间特征;再次,利用自注意力机制处理特征并重新分配权重;最后,输入全连接网络获得RUL预测结果。使用商用模块化航空推进系统仿真(C-MAPSS)数据集验证所提模型的有效性。实验结果显示,与先进模型相比,所提模型的Score分数在3个数据子集上取得最小值,在1个数据子集上取得次小值;均方根误差(RMSE)在1个数据子集上取得最小值,在3个数据子集上取得次小值。消融实验结果也验证了所提模型的各模块能有效提升预测精度。
