Pulse-like ground motions are capable of inflicting significant damage to structures. Efficient classification of pulse-like ground motion is of great importance when performing the seismic assessment in near-fault re...Pulse-like ground motions are capable of inflicting significant damage to structures. Efficient classification of pulse-like ground motion is of great importance when performing the seismic assessment in near-fault regions. In this study, a new method for identifying the velocity pulses is proposed, based on different trends of two parameters: the short-time energy and the short-time zero crossing rate of a ground motion record. A new pulse indicator, the relative energy zero ratio(REZR), is defined to qualitatively identify pulse-like features. The threshold for pulse-like ground motions is derived and compared with two other identification methods through statistical analysis. The proposed procedure not only shows good accuracy and efficiency when identifying pulse-like ground motions but also exhibits good performance for classifying records with high-frequency noise and discontinuous pulses. The REZR method does not require a waveform formula to express and fit the potential velocity pulses;it is a purely signal-based classification method. Finally, the proposed procedure is used to evaluate the contribution of pulse-like motions to the total input energy of a seismic record, which dramatically increases the seismic damage potential.展开更多
The medium-short term forecast for a certain kinds of main earthquake events might be possible with the time-to-failure method presented by Varnes (1989), Bufe and Varnes (1993), which is to simulate an accelerative r...The medium-short term forecast for a certain kinds of main earthquake events might be possible with the time-to-failure method presented by Varnes (1989), Bufe and Varnes (1993), which is to simulate an accelerative releasing model of precursory earthquake energy. By fitting the observed data with the theoretical formula, a medium-short term forecast technique for the main shock events could be established, by which the location, time and magnitude of the main shock could be determined. The data used in the paper are obtained from the earthquake catalogue recorded by Yunnan Regional Seismological Network with a time coverage of 1965~2002. The statistical analyses for the past 37 years show that the data of M2.5 earthquakes were fairly complete. In the present paper, 30 main shocks occurred in Yunnan region were simulated. For 25 of them, the forecasting time and magnitude from the simulation of precursory sequence are very close to the actual values with the precision of about 0.57 (magnitude unit). Suppose that the last event of the precursory sequence is known, then the time error for the forecasting main shock is about 0.64 year. For the other 5 main shocks, the simulation cannot be made due to the insufficient precursory events for the full determination of energy accelerating curve or disturbance to the energy-release curve. The results in the paper indicate that there is no obviously linear relation in the optimal searching radius for the main shock and the precursory events because Yunnan is an active region with damage earthquakes and moderate and small earthquakes. However, there is a strong correlation between the main shock moment and the coefficient k/m. The optimal fitting range for the forecasting time and magnitude can be further reduced using the relation between the main shock moment lgM0 and the coefficient lgk/m and the value range of the restricting index m, by which the forecast precision of the simulated main shock can be improved. The time-to-failure method is used to fit 30 main shocks in the paper and more than 80% of them have acquired better results, indicating that the method is prospective for its ability to forecast the known main shock sequence. Therefore, the prospect is cheerful to make medium-short term forecast for the forthcoming main shocks by the precursory events.展开更多
针对气象因素对多元负荷变化的灵敏度差异及多元负荷间耦合强度的差异导致多任务学习(multi-tasklearning,MTL)预测模型精度受限的问题,该文提出一种MTL和单任务学习(single-tasklearning,STL)组合的多元负荷预测方法。首先使用基于长...针对气象因素对多元负荷变化的灵敏度差异及多元负荷间耦合强度的差异导致多任务学习(multi-tasklearning,MTL)预测模型精度受限的问题,该文提出一种MTL和单任务学习(single-tasklearning,STL)组合的多元负荷预测方法。首先使用基于长短期记忆(long and short-term memory,LSTM)网络的MTL模型提取多元负荷间的耦合信息进行初步预测;然后采用基于前置双重注意力长短期记忆(dual attention before LSTM,DABLSTM)网络的STL模型减少输入噪声进行二次预测;同时将初步的预测值输入STL模型,使得STL模型可以考虑未来的时序信息;最后,通过全连接层对两个模型的预测结果进行融合得到最终的预测结果。实验结果表明,所提组合模型相比单一的MTL和STL模型具有更高的预测精度。展开更多
全球人口的快速增长和技术进步极大地提高了世界的总发电量,电能消耗预测对于电力系统调度和发电量管理发挥着重要的作用,为了提高电能消耗的预测精度,针对能耗数据的复杂时序特性,文中提出了一种将注意力机制(Attention)放置于双层长...全球人口的快速增长和技术进步极大地提高了世界的总发电量,电能消耗预测对于电力系统调度和发电量管理发挥着重要的作用,为了提高电能消耗的预测精度,针对能耗数据的复杂时序特性,文中提出了一种将注意力机制(Attention)放置于双层长短期记忆人工神经网络(Double layer Long Short-Term Memory,DLSTM)中的新颖夹层结构,即A-DLSTM。该网络结构利用夹层中的注意力机制自适应地关注单个时间单元中不同的特征量,通过双层LSTM网络对序列中的时间信息进行抓取,以对序列数据进行预测。文中的实验数据为UCI机器学习数据集上某家庭近5年的用电量,采用网格搜索法进行调参,实验对比了A-DLSTM与现有的模型在能耗数据上的预测性能,文中的网络结构在均方误差、均方根误差、平均绝对误差、平均绝对百分比误差上均达到了最优,且通过热力图对注意力层进行了分析,确定了对用电量预测影响最大的因素。展开更多
基金Supported by:National Natural Science Foundation of China under Grant Nos.51378341,51427901 and 51678407National Key Research and Development Program under Grant No.2016YFC0701108
文摘Pulse-like ground motions are capable of inflicting significant damage to structures. Efficient classification of pulse-like ground motion is of great importance when performing the seismic assessment in near-fault regions. In this study, a new method for identifying the velocity pulses is proposed, based on different trends of two parameters: the short-time energy and the short-time zero crossing rate of a ground motion record. A new pulse indicator, the relative energy zero ratio(REZR), is defined to qualitatively identify pulse-like features. The threshold for pulse-like ground motions is derived and compared with two other identification methods through statistical analysis. The proposed procedure not only shows good accuracy and efficiency when identifying pulse-like ground motions but also exhibits good performance for classifying records with high-frequency noise and discontinuous pulses. The REZR method does not require a waveform formula to express and fit the potential velocity pulses;it is a purely signal-based classification method. Finally, the proposed procedure is used to evaluate the contribution of pulse-like motions to the total input energy of a seismic record, which dramatically increases the seismic damage potential.
文摘The medium-short term forecast for a certain kinds of main earthquake events might be possible with the time-to-failure method presented by Varnes (1989), Bufe and Varnes (1993), which is to simulate an accelerative releasing model of precursory earthquake energy. By fitting the observed data with the theoretical formula, a medium-short term forecast technique for the main shock events could be established, by which the location, time and magnitude of the main shock could be determined. The data used in the paper are obtained from the earthquake catalogue recorded by Yunnan Regional Seismological Network with a time coverage of 1965~2002. The statistical analyses for the past 37 years show that the data of M2.5 earthquakes were fairly complete. In the present paper, 30 main shocks occurred in Yunnan region were simulated. For 25 of them, the forecasting time and magnitude from the simulation of precursory sequence are very close to the actual values with the precision of about 0.57 (magnitude unit). Suppose that the last event of the precursory sequence is known, then the time error for the forecasting main shock is about 0.64 year. For the other 5 main shocks, the simulation cannot be made due to the insufficient precursory events for the full determination of energy accelerating curve or disturbance to the energy-release curve. The results in the paper indicate that there is no obviously linear relation in the optimal searching radius for the main shock and the precursory events because Yunnan is an active region with damage earthquakes and moderate and small earthquakes. However, there is a strong correlation between the main shock moment and the coefficient k/m. The optimal fitting range for the forecasting time and magnitude can be further reduced using the relation between the main shock moment lgM0 and the coefficient lgk/m and the value range of the restricting index m, by which the forecast precision of the simulated main shock can be improved. The time-to-failure method is used to fit 30 main shocks in the paper and more than 80% of them have acquired better results, indicating that the method is prospective for its ability to forecast the known main shock sequence. Therefore, the prospect is cheerful to make medium-short term forecast for the forthcoming main shocks by the precursory events.
文摘针对气象因素对多元负荷变化的灵敏度差异及多元负荷间耦合强度的差异导致多任务学习(multi-tasklearning,MTL)预测模型精度受限的问题,该文提出一种MTL和单任务学习(single-tasklearning,STL)组合的多元负荷预测方法。首先使用基于长短期记忆(long and short-term memory,LSTM)网络的MTL模型提取多元负荷间的耦合信息进行初步预测;然后采用基于前置双重注意力长短期记忆(dual attention before LSTM,DABLSTM)网络的STL模型减少输入噪声进行二次预测;同时将初步的预测值输入STL模型,使得STL模型可以考虑未来的时序信息;最后,通过全连接层对两个模型的预测结果进行融合得到最终的预测结果。实验结果表明,所提组合模型相比单一的MTL和STL模型具有更高的预测精度。
文摘全球人口的快速增长和技术进步极大地提高了世界的总发电量,电能消耗预测对于电力系统调度和发电量管理发挥着重要的作用,为了提高电能消耗的预测精度,针对能耗数据的复杂时序特性,文中提出了一种将注意力机制(Attention)放置于双层长短期记忆人工神经网络(Double layer Long Short-Term Memory,DLSTM)中的新颖夹层结构,即A-DLSTM。该网络结构利用夹层中的注意力机制自适应地关注单个时间单元中不同的特征量,通过双层LSTM网络对序列中的时间信息进行抓取,以对序列数据进行预测。文中的实验数据为UCI机器学习数据集上某家庭近5年的用电量,采用网格搜索法进行调参,实验对比了A-DLSTM与现有的模型在能耗数据上的预测性能,文中的网络结构在均方误差、均方根误差、平均绝对误差、平均绝对百分比误差上均达到了最优,且通过热力图对注意力层进行了分析,确定了对用电量预测影响最大的因素。
文摘快速、准确定位振荡源是抑制电力系统强迫功率振荡的关键。为提高电力系统强迫振荡源定位精度和效率,该文提出一种基于耗散能量谱的电力系统强迫振荡源频域定位方法。该方法首先将电力系统广域测量信息进行短时傅里叶变换(short-time Fourier transform,STFT);然后,根据信号的时—频域特性,推导出时域耗散能量与时频域耗散能量间的关系,在此基础上,构建出频域耗散能量谱,论证了时域耗散能量与频域耗散能量谱的等价性;进而根据频域耗散能量谱辨识系统强迫振荡频率、定位强迫振荡源;最后,将所提方法应用到WECC179节点测试系统和ISO New England中进行仿真、验证,结果验证了所提方法的准确性和有效性。
