共查询到20条相似文献，搜索用时 15 毫秒
1.
当前震后建筑经济损失评估模型得到的震后建筑经济损失评估精确度、效率低,针对单一神经网络易产生局部极值等问题,对神经网络方法进行了改进,提出LM-BP神经网络在震后建筑损失评估模型中的应用。输入样本要素为影响震后建筑经济损失的5项因素,输出样本是震后建筑经济损失评估结果,在此基础上采用LM-BP神经网络将训练转化成最小二乘问题,结合LM算法重新定义隐含层节点数量,构建基于LM-BP的神经网络震后经济损失评估模型,采用该模型获取最优震后建筑经济损失评估结果。仿真实验结果表明,所设计的评估模型最小评估误差为0.1%,相比同类模型具有高精确度的优势,是一种可靠的震后建筑经济损失评估模型。 相似文献
2.
The paper introduces the horizontal crustal movement obtained from GPS observations in the regional networks (including the
basic network and the fiducial network) of the Crustal Movement Observation Network of China (CMONOC) carried out in 1999
and 2001. This paper is characterized by the acquisition of the horizontal displacement velocities during the period from
1999 to 2001 at the observation stations in the regional networks with datum definition of a group of stable stations with
small mutual displacements in east China. Based on the most detailed map of horizontal crustal movement in Chinese mainland,
the division of blocks, their displacements and deformations are studied. An approach to analysis of the intensity of the
horizontal crustal deformation is proposed. The general characteristics of the recent horizontal crustal movement in Chinese
mainland and that before the Kunlunshan earthquake of M=8.1 on November 14, 2001 are analyzed.
Foundation item: The National Development and Programming Project for Key Basic Research (95-13-03-07). 相似文献
3.
Empirical relationships between natural vibration period and height of buildings in Singapore 
下载免费PDF全文
下载免费PDF全文 Singapore is a classic case of a modern metropolis with low hazard but high exposure to the seismicity in Sumatra. Because of land shortage, more than 80% of the population lives in high‐rise residential buildings. As part of the efforts to assess the seismic performance of buildings in Singapore subjected to long‐distance Sumatran earthquakes, relationships between the natural vibration period and height of high‐rise public residential buildings in Singapore are derived empirically by conducting ambient vibration tests on 116 buildings. The measured buildings have a height ranging from 4 to 30 stories. The aspect ratio of buildings in plan is found to be insignificant in affecting the natural vibration period of the first mode of the buildings. The period‐height relationships are derived using regression analysis considering the site properties of a building. It is concluded that the vibration periods estimated from the proposed period‐height relationship for buildings located at soft‐soil site are about 40% longer than the vibration periods estimated for buildings located at firm‐soil site. Measurements are also conducted to study the influence of buildings on the measured frequency of the surrounding soil. For this purpose, two buildings with 25 and 30 stories located at firm‐soil site and soft‐soil site, respectively, are selected. It is found that the distance of building influence on the measured frequency of the surrounding soil may reach up to one building height for a firm‐soil site and two building heights for a soft‐soil site. Additional data of natural vibration periods of 19 instrumented residential buildings, which have height ranging from 9 to 30 stories, were obtained from the building response recorded during the September 30, 2009 Sumatran earthquake event. The natural vibration periods of these buildings are compared with those estimated using the proposed period‐height relationships, and the absolute differences are found to be less than 12%. Copyright © 2013 John Wiley & Sons, Ltd. 相似文献
4.
The accurate analysis of the seismic response of isolated structures requires incorporation of the flexibility of supporting soil.However,it is often customary to idealize the soil as rigid during the analysis of such structures.In this paper,seismic response time history analyses of base-isolated buildings modelled as linear single degree-of-freedom(SDOF) and multi degree-of-freedom(MDOF) systems with linear and nonlinear base models considering and ignoring the flexibility of supporting soil are conducted.The flexibility of supporting soil is modelled through a lumped parameter model consisting of swaying and rocking spring-dashpots.In the analysis,a large number of parametric studies for different earthquake excitations with three different peak ground acceleration(PGA) levels,different natural periods of the building models,and different shear wave velocities in the soil are considered.For the isolation system,laminated rubber bearings(LRBs) as well as high damping rubber bearings(HDRBs) are used.Responses of the isolated buildings with and without SSI are compared under different ground motions leading to the following conclusions:(1) soil flexibility may considerably influence the stiff superstructure response and may only slightly influence the response of the flexible structures;(2) the use of HDRBs for the isolation system induces higher structural peak responses with SSI compared to the system with LRBs;(3) although the peak response is affected by the incorporation of soil flexibility,it appears insensitive to the variation of shear wave velocity in the soil;(4) the response amplifications of the SDOF system become closer to unit with the increase in the natural period of the building,indicating an inverse relationship between SSI effects and natural periods for all the considered ground motions,base isolations and shear wave velocities;(5) the incorporation of SSI increases the number of significant cycles of large amplitude accelerations for all the stories,especially for earthquakes with low and moderate PGA levels;and(6) buildings with a linear LRB base-isolation system exhibit larger differences in displacement and acceleration amplifications,especially at the level of the lower stories. 相似文献
5.
High-rise buildings are usually considered as flexible structures with low inherent damping. Therefore, these kinds of buildings are susceptible to wind-induced vibration. Tuned Mass Damper (TMD) can be used as an effective device to mitigate excessive vibrations. In this study, Artificial Neural Networks is used to find optimal mechanical properties of TMD for high-rise buildings subjected to wind load. The patterns obtained from structural analysis of different multi degree of freedom (MDF) systems are used for training neural networks. In order to obtain these patterns, structural models of some systems with 10 to 80 degrees-of-freedoms are built in MATLAB/SIMULINK program. Finally, the optimal properties of TMD are determined based on the objective of maximum displacement response reduction. The Auto-Regressive model is used to simulate the wind load. In this way, the uncertainties related to wind loading can be taken into account in neural network’s outputs. After training the neural network, it becomes possible to set the frequency and TMD mass ratio as inputs and get the optimal TMD frequency and damping ratio as outputs. As a case study, a benchmark 76-story office building is considered and the presented procedure is used to obtain optimal characteristics of the TMD for the building. 相似文献
6.
《Advances in water resources》2005,28(9):899-909
Functional networks were recently introduced as an extension of artificial neural networks (ANNs). Unlike ANNs, they estimate unknown neuron functions from given functional families during the training process. Here, we applied two types of functional network models, separable and associativity functional networks, to forecast river flows for different lead-times. We compared them with a conventional artificial neural network model, an ARMA model and a simple baseline model in three catchments. Results show that functional networks are flexible and comparable in performance to artificial neural networks. In addition, they are easier and quicker to train and so are useful tools as an alternative to artificial neural networks. These results were obtained with only the simplest structures of functional networks and it is possible that a more detailed study with more complex forms of the model will improve even further on these results. Thus we recommend that the use of functional networks in discharge time series modelling and forecasting should be further investigated. 相似文献
7.
Ｔｈｅｍｅｄｉｕｍ┐ａｎｄｓｈｏｒｔ┐ｔｅｒｍｐｒｅｄｉｃｔｉｏｎｍｅｔｈｏｄｓｏｆｓｔｒｏｎｇｅａｒｔｈｑｕａｋｅｓｂａｓｅｄｏｎｎｅｕ┐ｒａｌｎｅｔｗｏｒｋＺＨＩ－ＱＩＡＮＧＨＡＮ（韩志强）ＢＩ－ＱＵＡＮＷＡＮＧ（王碧泉）Ｉｎｓｔｉｔｕｔｅｏｆ... 相似文献
8.
基于BP神经网络模型的多层砖房震害预测方法 总被引：8，自引：2，他引：8
针对传统的基于地震烈度的建筑物震害预测方法的不足,本文以地震动峰值加速度作为建筑物震害预测的地震动指标,结合几次大地震中多层砖房的震害实例,提出了一种基于BP神经网络模型的建筑物震害预测方法,模型的输入为反映结构抗震性能的各类物理参数,输出为给定地震动峰值加速度下建筑物破坏状态的概率。研究表明:基于BP网络模型的多层砖房的震害预测结果与震害实例的实际情况比较吻合,本文的思路和方法可推广于其他不同类型的建筑结构的震害预测。 相似文献
9.
Introduction In the Crustal Movement Observation Network of China (CMONOC) there are 25 fiducialstations, 56 basic stations and 1 000 regional stations. They are scattered on 10 major blocks inChinese mainland with high density of observation stations on the blocks of high seismic activityin the regional networks. 10 major blocks or regions (they will be referred to as blocks in the paper,a letter is used as a symbol for each block) were divided during the design of the regionalnetwo… 相似文献
10.
基于BP神经网络的空间索杆结构节点损伤识别研究 总被引：1，自引：0，他引：1
针对某实际空间索杆结构的节点损伤现象,采用BP神经网络与基于振动的损伤识别两步法对其进行了识别研究,即首先确定可能发生节点损伤的子区域,在此基础上利用对应子区域的子网络识别出具体的损伤位置和程度。识别过程中采用两个杆单元模拟发生节点损伤的杆件,用抗弯刚度降低的端部短杆单元模拟节点损伤。研究表明,虽然空间索杆结构的动力性能较为复杂,但基于结构固有频率和模态位移的组合指标对节点损伤仍较为敏感,利用它们进行节点损伤识别是有效的。 相似文献
11.
《水文科学杂志》2013,58(2)
Abstract Artificial neural network (ANN) models provide huge potential for simulating nonlinear behaviour of hydrological systems. However, the potential of ANN is yet to be fully exploited due to the problems associated with improving the model generalization performance. Generalization refers to the ability of a neural network to correctly process input data that have not been used for calibrating the neural network model. In the hydrological context, better generalization performance implies higher precision of forecasting. The primary objectives of this study are to explore new measures for improving the generalization performance of an ANN-based rainfall–runoff model, and to evaluate the applicability of the new measures. A modified neural network model (entitled goal programming (GP) neural network) for modelling the rainfall–runoff process has been developed, in which three enhancements are made as compared to the widely-used backpropagation (BP) network. The three enhancements are (a) explicit integration of hydrological prior knowledge into the neural network learning; (b) incorporation of a modified training objective function; and (c) reduction of network sensitivity to input errors. Seven watersheds across a range of climatic conditions and watershed areas in China were selected for examining the alternative networks. The results demonstrate that the GP consistently outperformed the BP both in the calibration and verification periods and three proposed measures yielded improvement of performance. 相似文献
12.
A temporal artificial neural network‐based model is developed and applied for long‐lead rainfall forecasting. Tapped delay lines and recurrent connections are two different components that are used along with a static multilayer perceptron network to design a time‐delay recurrent neural network. The proposed model is, in fact, a combination of time‐delay and recurrent neural networks. The model is applied in three case studies of the Northwest, West, and Southwest basins of Iran. In addition, an autoregressive moving average with exogenous inputs (ARMAX) model is used as a baseline in order to be compared with the time‐delay recurrent neural networks developed in this study. Large‐scale climate signals, such as sea‐level pressure, that affect the rainfall of the study area are used as the predictors in the models, as well as the persistence between rainfall data. The results of winter‐spring rainfall forecasts are discussed thoroughly. It is demonstrated that in all cases the proposed neural network results in better forecasts in comparison with the statistical ARMAX model. Moreover, it is found that in two of three case studies the time‐delay recurrent neural networks perform better than either recurrent or time‐delay neural networks. The results demonstrate that the proposed method can significantly improve the long‐lead forecast by utilizing a non‐linear relationship between climatic predictors and rainfall in a region. Copyright © 2007 John Wiley & Sons, Ltd. 相似文献
13.
Seismic response analysis of an irregular base isolated building 总被引：3，自引：0，他引：3
This paper assesses the reliability of code-compliant linear and nonlinear dynamic analyses for irregular buildings with base
isolation system (BIS). Comprehensive analyses are carried out for a case study comprising a large reinforced concrete multi-storey
framed hospital with 327 high-damping rubber bearings. Spectral and time history (linear and nonlinear) analyses were performed
on the three-dimensional (3D) finite element model (FEM) of the structure; simplified analyses were also conducted on single-degree-of-freedom
(SDOF) systems. It is found that, at damageability limit state, the values of maximum interstorey drifts (d/h) computed with
spectral analyses on the three-dimensional FEM range between 1/6 and 1/10 of the code limit (d/h = 0.33%); thus more stringent
code limits should be required for buildings with BISs. The maximum floor acceleration is reduced by about 70% with respect
to the ground acceleration (free field site); the acceleration profile is uniform along the height of the multi-storey frame.
Threshold values of floor accelerations to assess the seismic performance of equipments in buildings with BIS are lacking.
At ultimate limit state (ULS), spectral analyses provide values of actions and deformations that are less conservative than
those derived through time history analyses. To perform reliable dynamic analyses of base isolated buildings it is crucial
to select natural earthquake ground motions compliant with the fundamental period of vibration of the structural system. Nevertheless,
it is not straightforward to select adequate natural strong motions in the catalogues available world-wide; buildings incorporating
BISs possess periods of vibration which are generally higher than 2.0 s. As a result, distant and high-magnitude earthquakes
are effective for base isolated buildings; nevertheless, such earthquakes are scarce in the seismic databases. The outcomes
of the present study also demonstrate that simplified linear analyses tend to provide estimates of the response quantities,
displacements of base isolators and base shear of the superstructure, which can be reliably employed at preliminary design
stage. Spectral analysis results of the 3D model tend to match those of the SDOF systems, even for irregular superstructure,
provided that modal mass participating ratios are greater than 85–90%. The results of spectral analyses on both SDOF and three-dimensional
FEM envelope the outcomes of linear time histories. 相似文献
14.
四川汶川8．0级地震山西数字强震动记录特征与分析 总被引：1，自引：1，他引：0
对2008年5月12日四川汶川8．0级地震,山西数字强震动台网27个台的地震加速度记录,进行了处理和分析,包括对加速度波形数据的基线校正,滤波,加速度傅立叶和反应谱计算,以及速度和位移计算。结果表明：局部场地的介质特性对地震动特征有相当大的影响,较厚的覆盖土层对较长周期的地震波有明显的放大作用。厚覆盖土层场地的强震动观测对自振周期较长建筑物的抗震设计具有重要的意义。 相似文献
15.
《水文科学杂志》2013,58(1):114-118
Abstract A reliable flood warning system depends on efficient and accurate forecasting technology. A systematic investigation of three common types of artificial neural networks (ANNs) for multi-step-ahead (MSA) flood forecasting is presented. The operating mechanisms and principles of the three types of MSA neural networks are explored: multi-input multi-output (MIMO), multi-input single-output (MISO) and serial-propagated structure. The most commonly used multi-layer feed-forward networks with conjugate gradient algorithm are adopted for application. Rainfall—runoff data sets from two watersheds in Taiwan are used separately to investigate the effectiveness and stability of the neural networks for MSA flood forecasting. The results indicate consistently that, even though the MIMO is the most common architecture presented in ANNs, it is less accurate because its multi-objectives (predicted many time steps) must be optimized simultaneously. Both MISO and serial-propagated neural networks are capable of performing accurate short-term (one- or two-step-ahead) forecasting. For long-term (more than two steps) forecasts, only the serial-propagated neural network could provide satisfactory results in both watersheds. The results suggest that the serial-propagated structure can help in improving the accuracy of MSA flood forecasts. 相似文献
16.
Modified three‐dimensional seismic assessment method for buildings based on ambient vibration tests: extrapolation to higher shaking levels and measuring the dynamic amplification portion of natural torsion 下载免费PDF全文
下载免费PDF全文 This paper presents applications of the modified 3D‐SAM approach, a three‐dimensional seismic assessment methodology for buildings directly based on in situ experimental modal tests to calculate global seismic demands and the dynamic amplification portion of natural torsion. Considering that the building modal properties change from weak to strong motion levels, appropriate modification factors are proposed to extend the application of the method to stronger earthquakes. The proposed approach is consistent with the performance‐based seismic assessment approach, which entails the prediction of seismic displacements and drift ratios that are related to the damage condition and therefore the functionality of the building. The modified 3D‐SAM is especially practical for structures that are expected to experience slight to moderate damage levels and in particular for post‐disaster buildings that are expected to remain functional after an earthquake. In the last section of this paper, 16 low to mid‐rise irregular buildings located in Montreal, Canada, and that have been tested under ambient vibrations are analyzed with the method, and the dynamic amplification portion of natural torsion of the dataset is reported and discussed. The proposed methodology is appropriate for large‐scale assessments of existing buildings and is applicable to any seismic region of the world. Copyright © 2016 John Wiley & Sons, Ltd. 相似文献
17.
The cause-and-effect relations of the dynamics of high-latitude geomagnetic activity (in terms of the AL index) and the type of the magnetic cloud of the solar wind are studied with the use of artificial neural networks. A recurrent neural network model has been created based on the search for the optimal physically coupled input and output parameters characterizing the action of a plasma flux belonging to a certain magnetic cloud type on the magnetosphere. It has been shown that, with IMF components as input parameters of neural networks with allowance for a 90-min prehistory, it is possible to retrieve the AL sequence with an accuracy to ~80%. The successful retrieval of the AL dynamics by the used data indicates the presence of a close nonlinear connection of the AL index with cloud parameters. The created neural network models can be applied with high efficiency to retrieve the AL index, both in periods of isolated magnetospheric substorms and in periods of the interaction between the Earth’s magnetosphere and magnetic clouds of different types. The developed model of AL index retrieval can be used to detect magnetic clouds. 相似文献
18.
钢结构房屋动力特性脉动法测试研究 总被引：3，自引：0，他引：3
对上海地区的10幢钢结构建筑进行脉动法测试并采集数据,得到广义钢结构房屋的动力特性。选取其中1栋典型建筑通过多次测试和数值模拟分别验证测试的稳定性和准确性。通过分析处理测试数据建立钢结构建筑一阶周期与结构层数或高度的线性关系式,并归纳总结了等效阻尼比的测试结果,为验证结构动力特性理论计算结果、钢结构建筑减震隔震设计以及鉴定、加固改造、损伤识别提供依据。 相似文献
19.
Hopfield neural networks are massive parallel automata that support specific models and are adept in solving optimization problems. They suffer from a ‘rough’ solution space and convergence properties that are highly dependent on the starting model or prior. These detractions may be overcome by introducing regularization into the network in the form of local feedback smoothing. Application of regularized Hopfield networks to over 50 optimization test cases has yielded successful results, even with uniform (minimal information) priors. In particular, the non-linear, one- and two-dimensional magnetotelluric inverse problems have been solved by means of these regularized networks. The solutions compare favourably with those produced by other methods. Such regularized networks, with either hardware or programmed, parallel-computer implementation, can be extended to the problem of three-dimensional magnetotelluric inversion. Because neural networks are natural analog-to-digital converters, it is predicted that they will be the basic building blocks of future magnetotelluric instrumentation. 相似文献
20.
基于灰关联与人工神经网络综合评价模型的多层砖房震害预测 总被引：1，自引：0，他引：1
结合几次大地震中多层砖房的实际震害资料,基于灰关联识别方法,解析了各影响因子对多层砖房抗震性能的影响程度。以反映结构抗震性能的各类物理参数作为输入数据,以给定地震动峰值加速度下建筑物破坏状态的概率作为输出数据,采用8-6-5层结构,建立了基于BP人工神经网络的非线性模型,并对震害样本进行了训练。结果表明:利用灰关联分析,可得出各因子对多层砖房抗震性能影响程度的大小排序,有利于实际的工程抗震设计;基于BP人工神经网络模型的多层砖房的震害预测结果与震害实例的实际情况比较吻合,其思路和方法可推广于其他不同类型的建筑结构的震害预测。 相似文献