An efficient 2.5D finite element numerical modeling approach was developed to simulate wave motions generated in ground by high-speed train passages. Fourier transform with respect to the coordinate in the track direc...An efficient 2.5D finite element numerical modeling approach was developed to simulate wave motions generated in ground by high-speed train passages. Fourier transform with respect to the coordinate in the track direction was applied to re-ducing the three-dimensional dynamic problem to a plane strain problem which has been solved in a section perpendicular to the track direction. In this study, the track structure and supporting ballast layer were simplified as a composite Euler beam resting on the ground surface, while the ground with complicated geometry and physical properties was modeled by 2.5D quadrilateral elements. Wave dissipation into the far field was dealt with the transmitting boundary constructed with fre-quency-dependent dashpots. Three-dimensional responses of track structure and ground were obtained from the wavenumber expansion in the track direction. The simulated wave motions in ground were interpreted for train moving loads traveling at speeds below or above the critical velocity of a specific track-ground system. It is found that, in the soft ground area, the high-speed train operations can enter the transonic range, which can lead to resonances of the track structure and the sup-porting ground. The strong vibration will endanger the safe operations of high-speed train and accelerate the deterioration of railway structure.展开更多
文摘短期电力负荷预测在电网安全运行和制定合理调度计划方面发挥着重要作用。为了提高电力负荷时间序列预测的准确度,提出了一种由完整自适应噪声集成经验模态分解(complete ensemble empirical mode decomposition with adaptive noise,CEEMDAN)和基于注意力机制的长短期记忆神经网络(long short-term memory network based on attention mechanism,LSTM-Attention)相结合的短期电力负荷预测模型。完整自适应噪声集成经验模态分解有效地将负荷时间序列分解成多个层次规律平稳的本征模态分量,并通过神经网络模型预测极大值,结合镜像延拓方法抑制边界效应,提高分解精度,同时基于注意力机制的长短期记忆神经网络自适应地提取电力负荷数据输入特征并分配权重进行预测,最后各预测模态分量叠加重构后获得最终预测结果。通过不同实际电力负荷季节数据分别进行实验,并与其他电力负荷预测模型结果分析进行比较,验证了该预测方法在电力负荷预测精度方面具有更好的性能。
基金Supported by the National Natural Science Foundation of China (Grant No. 10702063) the Research Fund for the Doctoral Program of Higher Education of China (Grant No. 20070335086)
文摘An efficient 2.5D finite element numerical modeling approach was developed to simulate wave motions generated in ground by high-speed train passages. Fourier transform with respect to the coordinate in the track direction was applied to re-ducing the three-dimensional dynamic problem to a plane strain problem which has been solved in a section perpendicular to the track direction. In this study, the track structure and supporting ballast layer were simplified as a composite Euler beam resting on the ground surface, while the ground with complicated geometry and physical properties was modeled by 2.5D quadrilateral elements. Wave dissipation into the far field was dealt with the transmitting boundary constructed with fre-quency-dependent dashpots. Three-dimensional responses of track structure and ground were obtained from the wavenumber expansion in the track direction. The simulated wave motions in ground were interpreted for train moving loads traveling at speeds below or above the critical velocity of a specific track-ground system. It is found that, in the soft ground area, the high-speed train operations can enter the transonic range, which can lead to resonances of the track structure and the sup-porting ground. The strong vibration will endanger the safe operations of high-speed train and accelerate the deterioration of railway structure.
