Seismic isolation devices are usually designed to protect structures from the strong horizontal component of earthquake ground shaking. However, the effect of near-fault (NF) vertical ground motions on seismic responses of buildings has become an important consideration due to the observed building damage caused by vertical excitation. As the structure needs to maintain its load bearing capacity, using the horizontal isolation strategy in vertical seismic isolation will lead to the problem of larger static displacement. In particular, the bearings may generate large deformation responses of isolators for NF vertical ground motions. A seismic isolation system including quasi-zero stiffness (QZS) and vertical damper (VD) is used to control NF vertical earthquakes. The characteristics of vertical seismic isolated structures incorporating QZS and VD are presented. The formula for the maximum bearing capacity of QZS isolation considering the stiffness of vertical spring components is obtained by theoretical derivation. From the static analysis, it is found that the static capacity of the QZS isolation system with vertical seismic isolation components increases when the configurative parameter reduces. Seismic response analyses of the seismic isolated structure model with QZS and VD subjected to NF vertical earthquakes are conducted. The results show that seismic responses of the structure can be controlled by setting the appropriate static equilibrium position, vertical isolation period, and vertical damping ratio. Adding a damping ratio is effective in controlling the vertical large deformation of the isolator.

No abstract

With the rapid development of sensor technology, machine learning, and the Internet of Things, wireless sensor networks have gradually become a research hotspot. In order to improve the data fusion performance of wireless sensor networks and ensure network security in the event of external attacks, this paper proposes a wireless sensor optimization algorithm model, involving wireless sensor networks, the Internet of Things, and other related fields. This paper first analyzes the role of the Internet of Things in wireless sensor networks, studies the localization mechanism and hierarchy of the Internet of Things based on wireless sensor networks, and improves the LE-RLPCCA (Position Estimation Robust Local Retention Criteria Correlation Analysis) localization algorithm model based on sensor grids. This paper discusses the problems of machine learning in wireless sensor networks, constructs a sensor-based machine learning model, and designs a data fusion algorithm for a wireless sensor networks’ machine learning model. The application of wireless sensors in engineering mechanics experiments is summarized, and the optimization algorithm model of the wireless sensor in engineering mechanics experiments is proposed. The analysis results show that the average accuracy of the DKFCM-FSVM (Density aware Kernel-based Fuzzy C-means Clustering algorithm Fuzzy Support Vector Machine) algorithm in detecting five behaviors is 0.997, 0.992, 0.904, 0.996, and 0.946, respectively, and the accuracy in detecting different behaviors is the best, 0.005, 0.01, 0.003, and 0.006 respectively. It achieves the lowest false positive rate in the detection of different behaviors, and the average false positive rate is 0.004, 0.003, 0.003, 0.008, and 0.005, respectively, which shows that the DKFCM-FSVM algorithm model of wireless sensor networks in engineering mechanics experiments is the optimal solution. The work of this paper has good reference value for the application of wireless sensor networks and the optimization of engineering mechanics experimental methods and is helpful for further research of sensor technology.

The traditional identification methods have limited ability to identify damage location of bridge structures. Therefore, a bridge structural damage location identification method based on deep learning is proposed. In addition, the sigmoid function is the activation function, and the cross entropy is the cost function. Meanwhile, take the Gaussian noise as the addition method and take the softmax as the classifier. So the constructed SDAE deep learning model can realize damage location identification of the simply supported the continuous beam bridges. Compared with the traditional identification methods of bridge structures, namely BP network and SVM, the proposed method shows higher identification accuracy and antinoise performance. Here, the average identification accuracy of the method for continuous beam bridge is 99.8%. As can be seen that the proposed method is more suitable for practical bridge structure damage location identification.