An unidirectional and bidirectional hybrid connective star network model with coupling time-delay is constructed in this paper. According to synchronization error systems, adaptive controllers for each node are struct...An unidirectional and bidirectional hybrid connective star network model with coupling time-delay is constructed in this paper. According to synchronization error systems, adaptive controllers for each node are structured by using the linear system stability method and the Lyapunov stability method. These adaptive controllers can realize the modified functional projective synchronization between each node of star network and an isolated node by argument and analysis. Finally, the corrective and effective of the adaptive controllers are illustrated by some numerical examples.展开更多
The macrostructure and properties of the thin walled copper tube prepared by the downward continuous unidirectional solidification （DCUS） method were studied. The result shows that the macrostructure is closely rela...The macrostructure and properties of the thin walled copper tube prepared by the downward continuous unidirectional solidification （DCUS） method were studied. The result shows that the macrostructure is closely related to the solid-liquid interface profile, which is influenced by the distance between the cooling water location and the solidification front. The mechanical properties of the thin walled copper tube prepared by the DCUS method are near those of the normal cast copper, and it has good relative density, electrical conductivity, and elongation, which are not greatly affected by casting speed. The thin walled copper tube prepared by the DCUS method also has good processing properties that can be taken to further drawing procedures directly without an intermediate process, and obtains good mechanical properties with the total processing rate of 89.8%.展开更多
In this study we performed a classical spectrum analysis of seismic waveforms recorded at far field stations of the great MW7.9 Wenchuan earthquake to observe the shifts of the corner frequency with azimuth due to the...In this study we performed a classical spectrum analysis of seismic waveforms recorded at far field stations of the great MW7.9 Wenchuan earthquake to observe the shifts of the corner frequency with azimuth due to the Doppler effect.Our results show that this damaging great earthquake had a dominating rupture propagation direction of 64.0°.The equivalent radius of the fault rupture surface was estimated to be 33 km,yielding the rupture area of about 3 500 km2.Thus the length of the rupture fault surface is about 230 km if the depth（or width） extent is 15 km.The computer program developed in this study can quickly provide the information about the source of a future large（damaging） earthquake,which could be very useful for predicting aftershocks and planning the rescue operations.展开更多
At present, the volume pipe cleaning methods mainly include diesel method and water-based method. The water-based method is widely used in the volume pipe cleaning of each handover station because of its low cost and ...At present, the volume pipe cleaning methods mainly include diesel method and water-based method. The water-based method is widely used in the volume pipe cleaning of each handover station because of its low cost and high safety. Due to the certain difference in structure between the two-way volume pipe and the one-way volume pipe, the consideration of cleaning flow is not perfect, which is easy to lead to high labor intensity, low cleaning efficiency and high cost. The purpose of this paper is to make use of the structural characteristics of two-way volume pipe, and reasonably plan the cleaning process, install small devices and optimize the stages of hot water circulation and reagent circulation in order to enhance the cleaning effect, improve the cleaning efficiency and reduce the cost.展开更多
目前在高校C语言编程课程中,使用客观评价的题目难度考验学生的学习情况是非常重要的手段。目前大部分难度评估方法都针对特有科目和特有题型,而对中文编程题目的难度评估存在不足。因此,提出一种融合题目文本和知识点标签的基于BERT(Bi...目前在高校C语言编程课程中,使用客观评价的题目难度考验学生的学习情况是非常重要的手段。目前大部分难度评估方法都针对特有科目和特有题型,而对中文编程题目的难度评估存在不足。因此,提出一种融合题目文本和知识点标签的基于BERT(Bidirectional Encoder Representations from Transformers)和双向长短时记忆(Bi-LSTM)模型的C语言题目难度预测模型FTKB-BiLSTM(Fusion of Title and Knowledge based on BERT and Bi-LSTM)。首先,利用BERT的中文预训练模型获得题目文本和知识点的词向量;其次,融合模块将融合后的信息通过BERT处理得到文本的信息表示,并输入Bi-LSTM模型中学习其中的序列信息,提取更丰富的特征;最后,把经Bi-LSTM模型得到的特征表示通过全连接层并经过Softmax函数处理得到题目难度分类结果。在Leetcode中文数据集和ZjgsuOJ平台数据集上的实验结果表明,相较于XLNet等主流的深度学习模型,所提模型的准确率更优,具有较强的分类能力。展开更多
基金Supported by the National Natural Science Foundation of China(11161027)Natural Science Foundation of Gansu Province(1610RJZA080)the Foundation of Gansu Education Bureau(2017A-155)
文摘An unidirectional and bidirectional hybrid connective star network model with coupling time-delay is constructed in this paper. According to synchronization error systems, adaptive controllers for each node are structured by using the linear system stability method and the Lyapunov stability method. These adaptive controllers can realize the modified functional projective synchronization between each node of star network and an isolated node by argument and analysis. Finally, the corrective and effective of the adaptive controllers are illustrated by some numerical examples.
文摘The macrostructure and properties of the thin walled copper tube prepared by the downward continuous unidirectional solidification （DCUS） method were studied. The result shows that the macrostructure is closely related to the solid-liquid interface profile, which is influenced by the distance between the cooling water location and the solidification front. The mechanical properties of the thin walled copper tube prepared by the DCUS method are near those of the normal cast copper, and it has good relative density, electrical conductivity, and elongation, which are not greatly affected by casting speed. The thin walled copper tube prepared by the DCUS method also has good processing properties that can be taken to further drawing procedures directly without an intermediate process, and obtains good mechanical properties with the total processing rate of 89.8%.
文摘In this study we performed a classical spectrum analysis of seismic waveforms recorded at far field stations of the great MW7.9 Wenchuan earthquake to observe the shifts of the corner frequency with azimuth due to the Doppler effect.Our results show that this damaging great earthquake had a dominating rupture propagation direction of 64.0°.The equivalent radius of the fault rupture surface was estimated to be 33 km,yielding the rupture area of about 3 500 km2.Thus the length of the rupture fault surface is about 230 km if the depth（or width） extent is 15 km.The computer program developed in this study can quickly provide the information about the source of a future large（damaging） earthquake,which could be very useful for predicting aftershocks and planning the rescue operations.
文摘At present, the volume pipe cleaning methods mainly include diesel method and water-based method. The water-based method is widely used in the volume pipe cleaning of each handover station because of its low cost and high safety. Due to the certain difference in structure between the two-way volume pipe and the one-way volume pipe, the consideration of cleaning flow is not perfect, which is easy to lead to high labor intensity, low cleaning efficiency and high cost. The purpose of this paper is to make use of the structural characteristics of two-way volume pipe, and reasonably plan the cleaning process, install small devices and optimize the stages of hot water circulation and reagent circulation in order to enhance the cleaning effect, improve the cleaning efficiency and reduce the cost.
文摘目前在高校C语言编程课程中,使用客观评价的题目难度考验学生的学习情况是非常重要的手段。目前大部分难度评估方法都针对特有科目和特有题型,而对中文编程题目的难度评估存在不足。因此,提出一种融合题目文本和知识点标签的基于BERT(Bidirectional Encoder Representations from Transformers)和双向长短时记忆(Bi-LSTM)模型的C语言题目难度预测模型FTKB-BiLSTM(Fusion of Title and Knowledge based on BERT and Bi-LSTM)。首先,利用BERT的中文预训练模型获得题目文本和知识点的词向量;其次,融合模块将融合后的信息通过BERT处理得到文本的信息表示,并输入Bi-LSTM模型中学习其中的序列信息,提取更丰富的特征;最后,把经Bi-LSTM模型得到的特征表示通过全连接层并经过Softmax函数处理得到题目难度分类结果。在Leetcode中文数据集和ZjgsuOJ平台数据集上的实验结果表明,相较于XLNet等主流的深度学习模型,所提模型的准确率更优,具有较强的分类能力。
