共查询到19条相似文献，搜索用时 656 毫秒
1.
为了有效利用结构健康诊断中的多源不确定数据,提高损伤识别的准确率,通过改进D-S证据理论(Dempster提出由Shafer完善形成的一种推理理论)中的基本概率赋值函数和组合规则,提出了一种基于改进D-S证据理论的结构损伤识别新方法.该方法利用灰色关联理论和熵权理论处理信息源的基本概率赋值问题,利用改进的D-S组合规则处理信息源间的相关、冲突问题.通过钢管焊接结构的试验研究验证所提方法的有效性,结果表明所提出的方法优于传统的D-S损伤识别方法;相对于单一损伤信息的损伤识别而言,基于多源损伤信息能够有效降低损伤误判的可能性,获得更准确的损伤识别结果. 相似文献
2.
在位场数据处理及解释中,断裂的提取至关重要,因此,如何更有效提取边界信息,压制数据中的噪声成为解释人员重点研究目标.本文通过讨论用于识别重力断裂构造模型的导数方法与Daubechies小波分析方法在信号分析过程中的异同,证明了这两种方法在断裂识别中的有效性.提出将这两种断裂识别方法相结合进行断裂识别的新方法.通过对漠河盆地高精度重力剖面测线数据使用经典断裂反演方法和Daubechies多尺度小波分析相结合的方法划分出10条断裂构造,并在漠河盆地中依据平面重力异常利用上述方法识别出四组断裂构造,详细分析了由剖面数据及平面数据获得的各个断裂的构造特征.通过在漠河盆地高精度实测数据中的应用,多尺度小波分析与经典方法相结合的断裂构造识别方法可有效的避免单用经典方法时产生的多个离散点的影响及对断裂位置划分的干扰.该方法可明显去除一些不收敛的反演点位,使断裂图的清晰度得到大幅度提高. 相似文献
3.
《地震工程与工程振动》2021,41(4)
传统结构损伤识别需对采集数据进行分析,提取相应特征进行损伤诊断。特征提取过程需消耗大量的计算成本,无法满足结构健康监测在线损伤识别的需求。为提高损伤识别的计算效率和自动化程度,提出基于一维卷积神经网络的结构损伤识别方法,其特点是可以直接从原始振动信号中自主学习损伤特征,并准确快速地识别结构的损伤位置和损伤程度。采用简支梁数值模型和IABMAS BHM Benchmark数值模型验证所提方法的有效性。数值结果表明:所建立的一维卷积神经网络模型能够准确识别结构的损伤位置和损伤程度,具备一定的抗噪性能,整体模型收敛快,对单条样本测试延迟低。设计了钢框架结构损伤识别试验,采用所提方法对框架结构的损伤情况进行了识别。分析结果表明:所提方法可准确识别结构损伤程度及损伤类别,测试集准确率为100%,验证了方法在实际结构损伤识别的应用可行性。 相似文献
4.
基于加速度时域测试数据进行结构损伤识别计算时,所用测试数据的点数必须足够多才能够使识别有效,但往往又容易出现收敛到局部极小解的情况。为解决这一问题,本文提出了基于多重测点数目标函数族的结构损伤识别方法;所谓多重测点数目标函数族,即由不同点数的测试数据出发构造一族目标函数,以取代传统的基于单一点数的目标函数;迭代计算时采用了：Tikhonov正则化技术以抑制解答的病态性。以Benchmark结构为算例,分析了两种基于单一点数的损伤识别计算难以收敛到正确解答的原因;并考证了文中提出的方法。计算结果表明：基于多重测点数目标函数族的结构损伤识别方法,使得识别计算易于收敛到正确解答;从而证明了本文提出的方法是有效的。 相似文献
5.
特高压变压器套管具有较高的地震易损性,为研究其在地震过程中出现结构损伤的识别问题,基于改进的希尔伯特黄变换算法提出一种利用设备加速度响应信号进行实时损伤识别的方法. 采用高通滤波以及集合经验模态分解提取信号的异常高频成分;然后将其作为损伤特征,定义高频能量比用于损伤定位;最后通过数值算例模拟不同损伤工况下结构的地震响应,验证所提损伤识别方法的准确性.研究表明,地震过程中结构突发损伤会使加速度响应信号中产生瞬时高频成分;信号中瞬时高频成分的能量大小与采集点到损伤位置的距离有关,距离越近瞬时高频成分的能量量级越大.所提方法仅需结构的加速度响应作为算法输入即可实现损伤判定和损伤定位,数据需求简单. 相似文献
6.
为提高基于模态参数的损伤识别方法的损伤敏感性和噪声鲁棒性,将多源数据融合技术引入到苏通大桥主梁损伤定位方法中。基于D-S证据理论对模态柔度和模态应变能指标进行数据融合,并以苏通大桥扁平钢箱梁为分析对象,对融合后损伤定位指标的应用效果进行了讨论。结果表明：基于数据融合的损伤定位方法具有较强的损伤敏感性,只需要较少的低阶模态信息就能识别主梁的早期损伤;数据融合后,损伤定位指标可以在较强的噪声环境下准确地识别斜拉桥钢箱梁的损伤,具有较好的工程实用性。 相似文献
7.
地震相干属性是地震解释中识别大中尺度构造和地层异常的关键技术，目前常规属性技术基于叠后三维数据进行.宽方位地震采集和解释技术的发展为提高地下构造的地震识别精度提供了新思路，但如何利用丰富的宽方位地震数据开展地震几何属性研究仍具有挑战.因此，本文充分利用地质不连续面造成的地震异常在不同观测方位上的差异，以及垂直断裂构造走向的部分方位叠加数据对识别断裂有更好的效果，发展了一种基于方位地震数据提取多方位相干属性的方法.首先，在不同方位地震数据的三维分析窗中对地震道插值，沿不同方向构造方位分析窗，其次，在计算方位相干时，对方位分析窗内不同位置的地震道使用反距离加权算法计算权重，最后通过主成分分析技术实现多方位信息融合，并将归一化后的结果定义为最终多方位相干.实际应用验证了所提方法的有效性和稳定性，与常规方法相比，本文所提方法包含了地质构造不同方位的特征，在所有方向上突出了地质边缘，同时提供了局部和整体的不连续性，并且有效地提高了相干图像的信噪比. 相似文献
8.
应用人工神经网络技术的大型斜拉桥子结构损伤识别研究 总被引：12，自引：0，他引：12
本文应用人工神经网络技术对大型斜拉桥结构进行了子结构损伤识别研究。文中首先介绍了子结构损伤识别的基本方法，然后应用自组织竞争神经网络建立了对于大型桥梁结构识别子结构损伤情况的子结构损伤识别方法，并且应用BP网络进一步建立了大型桥梁结构各子结构内部的损伤位置和损伤程度的识别方法，数值模拟了一大跨度斜拉桥子结构损伤以及子结构内部损伤的识别过程，最后得出结论：(1)基于自组织竞争网络的子结构损伤识别方法能迅速准确地识别大型结构的损伤情况；(2)基于BP网络所建立的结构损伤识别方法，能对子结构中结构损伤的位置和程度进行进一步的识别；(3)基于人工神经网络技术的结构损伤识别方法是大型土木工程结构损伤识别的有效方法，可在工程结构损伤识别中广泛应用。 相似文献
9.
边界识别技术是位场数据解释中一项基本的工作,现有的边界识别方法多存在边界识别结果发散和不能均衡深浅地质体异常的缺点.目前一些均衡边界识别方法会因正负异常同时存在而引起额外的错误边界或者存在人为主观因素去除错误边界信息的缺点.本文充分利用重力位场张量梯度的多信息成分,提出利用水平方向解析信号及其垂向导数与传统的均衡边界识别方法做结合的方式定义新的探测边界的方法.通过理论模型试验证明新方法同传统方法相比,能够更加清晰、准确的圈定出深浅地质体的边界.最后将新的边界识别方法应用到实测重力异常数据解释中,取得了良好的边界识别结果并能够发现更多的构造细节. 相似文献
10.
空间归一化边界识别方法用于判断地质体的水平位置及深度(英文) 总被引：1，自引：1，他引：0
边界识别是重磁数据解释中的常用方法之一,依据其结果可划分出地质体的水平范围。边界识别结果受地质体埋深及导数计算误差的影响所识别边界与真实边界之间存在一定的差距,且边界识别法无法直观地给出地质体的深度信息。为了获得异常体的水平位置和深度信息,本文提出空间归一化边界识别方法,其对不同深度的边界识别函数进行归一化计算,空间归一化边界识别法的最大值对应于异常体的水平位置和深度。常规边界识别结果的误差随理深的减小而减小,而空间归一化边界识别法是通过最大值来判断地质体的位置,最大值是在地质体处获得,因此归一化边界识别方法所获得的结果是准确的。通过理论模型试验证明归一化边界识别方法能有效地完成异常体的水平位置和深度的计算,所获得的水平位置和深度信息与理论值相一致,为下一步的勘探计划提供了更加可靠的依据。将其应用于实际航磁数据的解释,获得了断裂的具体分布形式。 相似文献
11.
This paper presents two methods to perform system identification at the substructural level, taking advantage of reduction in the number of unknowns and degrees of freedom (DOFs) involved, for damage assessment of fairly large structures. The first method is based on first‐order state space formulation of the substructure where the eigensystem realization algorithm (ERA) and the observer/Kalman filter identification (OKID) are used. Identification at the global level is then performed to obtain the second‐order model parameters. In the second method, identification is performed at the substructural level in both the first‐ and second‐order model identification. Both methods are illustrated using numerical simulation studies where results indicate their significantly better performance than identification using the global structure, in terms of efficiency and accuracy. A 12‐DOF system and a fairly large structural system with 50 DOFs are used where the effects of noisy data are considered. In addition to numerical simulation studies, laboratory experiments involving an eight‐storey frame model are carried out to illustrate the performance of the proposed method. The identification results presented in terms of the stiffness integrity index show that the proposed methodology is able to locate and quantify damage fairly accurately. Copyright © 2005 John Wiley & Sons, Ltd. 相似文献
12.
利用小波变换时-频局部化性能,提出了基于小波能量分布向量的结构损伤识别方法。首先建立无损结构响应信号小波能量分布的总体向量;其次,将实测动力响应信号分解为小波包组分,计算其小波能量分布向量(样本向量);通过样本向量和总体向量之间的马氏距离识别损伤。该方法仅利用单测点结构响应数据进行损伤识别,实验方便,计算简单,并通过钢梁试验对损伤识别方法进行了试验验证,识别结果表明小波能量分布向量是一个比较好的结构损伤指标。 相似文献
13.
通过叠前反演获得的单参数或组合参数都有一定的流体识别能力,但如何将多种流体识别因子有效融合是目前进行流体识别的一个难题.利用人工参与进行流体性质的综合解释是目前流体识别因子融合的主要途径,但这种方法人为干扰较大,不确定性强.鉴于此,本文提出了一种基于近似支持向量机的流体识别方法.该方法首先以实际工区测井资料为依据,优选出对工区内储层所含流体特征敏感的流体识别因子作为输入参数,然后通过近似支持向量机进行流体性质的判别,实例证明该方法的识别结果客观准确,是一种可靠的流体识别方法. 相似文献
14.
Two‐stage damage detection algorithms of structure using modal parameters identified from recursive subspace identification 
下载免费PDF全文
下载免费PDF全文 Structural damage assessment under external loading, such as earthquake excitation, is an important issue in structural safety evaluation. In this regard, appropriate data analysis and feature extraction techniques are required to interpret the measured data and to identify the state of the structure and, if possible, to detect the damage. In this study, the recursive subspace identification with Bona‐fide LQ renewing algorithm (RSI‐BonaFide‐Oblique) incorporated with moving window technique is utilized to identify modal parameters such as natural frequencies, damping ratios, and mode shapes at each instant of time during the strong earthquake excitation. From which the least square stiffness method (LSSM) combined with the model updating technique, called efficient model correction method (EMCM), is used to estimate the first‐stage system stiffness matrix using the simplified model from the previously identified modal parameters (nominal model). In the second stage, 2 different damage assessment algorithms related to the nominal system stiffness matrix were derived. First, the model updating technique, called EMCM, is applied to correct the nominal model by the newly identified modal parameters during the strong motion. Second, the element damage index can be calculated using element damage index method (EDIM) to quantify the damage extent in each element. Verification of the proposed methods through the shaking table test data of 2 different types of structures and a building earthquake response data is demonstrated to specify its corresponding damage location, the time of occurrence during the excitation, and the percentage of stiffness reduction. 相似文献
15.
Daily reservoir inflow forecasting using artificial neural networks with stopped training approach 总被引：28，自引：0，他引：28
In this paper, an early stopped training approach (STA) is introduced to train multi-layer feed-forward neural networks (FNN) for real-time reservoir inflow forecasting. The proposed method takes advantage of both Levenberg–Marquardt Backpropagation (LMBP) and cross-validation technique to avoid underfitting or overfitting on FNN training and enhances generalization performance. The methodology is assessed using multivariate hydrological time series from Chute-du-Diable hydrosystem in northern Quebec (Canada). The performance of the model is compared to benchmarks from a statistical model and an operational conceptual model. Since the ultimate goal concerns the real-time forecast accuracy, overall the results show that the proposed method is effective for improving prediction accuracy. Moreover it offers an alternative when dynamic adaptive forecasting is desired. 相似文献
16.
This paper deals with the identification of the parameters of a smoothed hysteretic model which was proposed by Bouc and Wen with emphasis on restoring force hysteresis. The problem of estimating the parameters of this system on the basis of input-output data, possibly noise corrupted, is considered. Through the application of various simulated time histories from the hysteretic model, a three-stage systematic method of system identification was proposed. Four different methods of identification are arranged and conducted in this three-stage system identification. The first stage, a sequential regressional analysis is used to identify the equivalent linear system from which elastic or inelastic response can be identified. The identified parameters can be used in the stage when the system is in elastic response. In the second stage, both time domain least-squares method and Gauss-Newton method are applied. The convergence of the Gauss-Newton method can be guaranteed if the identified results from least-squares method are adopted as the initial values for Gauss-Newton method. In the third stage, the extended Kalman filtering technique is needed to identify the noise-corrupt data. Application of this algorithm to a SDOF non-deteriorating system is verified. 相似文献
17.
Structural modal parameter identification and damage diagnosis based on Hilbert-Huang transform 总被引：1，自引：1，他引：0
Traditional modal parameter identifi cation methods have many disadvantages,especially when used for processing nonlinear and non-stationary signals.In addition,they are usually not able to accurately identify the damping ratio and damage.In this study,methods based on the Hilbert-Huang transform(HHT) are investigated for structural modal parameter identifi cation and damage diagnosis.First,mirror extension and prediction via a radial basis function(RBF) neural network are used to restrain the troublesome end-effect issue in empirical mode decomposition(EMD),which is a crucial part of HHT.Then,the approaches based on HHT combined with other techniques,such as the random decrement technique(RDT),natural excitation technique(NExT) and stochastic subspace identifi cation(SSI),are proposed to identify modal parameters of structures.Furthermore,a damage diagnosis method based on the HHT is also proposed.Time-varying instantaneous frequency and instantaneous energy are used to identify the damage evolution of the structure.The relative amplitude of the Hilbert marginal spectrum is used to identify the damage location of the structure.Finally,acceleration records at gauge points from shaking table testing of a 12-story reinforced concrete frame model are taken to validate the proposed approaches.The results show that the proposed approaches based on HHT for modal parameter identifi cation and damage diagnosis are reliable and practical. 相似文献
18.
In this study, signal processing approaches and nonlinear identification are used to measure seismic responses of reinforced concrete (RC) structures using the shaking table test. To analyze structural nonlinearity, an equivalent linear system with time‐varying model parameters, singular spectrum analysis to elucidate residual deformation, and wavelet packet transformation analysis to yield the energy distribution among components are adopted to detect the nonlinearity. Then, damage feature extraction is conducted using both the Holder exponent and the Level‐1 detail of the discrete wavelet component. Finally, the modified Bouc‐Wen hysteretic model and the system identification process are employed to the shaking table test data to evaluate the physical parameters, including the stiffness degradation, the strength deterioration and the pinching hysteresis. Finally, the identified stiffness and strength degradation functions from the test data of RC frames in relation to the degree of ground shaking, damage index and the identified nonlinear features are discussed. Based on the proposed method, both signal‐based and model‐based identifications, the relationship between the damage occurrence and severity of structural damage can be identified. Copyright © 2010 John Wiley & Sons, Ltd. 相似文献
19.
为高效准确识别桥梁结构损伤,将深度学习与结构动力特性相结合,提出基于双层深度置信网络的桥梁结构损伤识别方法。首先取结构前3阶竖向振动频率和跨中节点前3阶竖向振动模态位移为参数,将其共同作为首层深度置信网络(DBN)的输入数据对结构的损伤位置进行识别;然后以1阶竖向振动的模态位移差作为参数,基于二层DBN对结构损伤程度进行预测;最后以郑许市域铁路桥梁为例进行验证。计算结果显示,当不考虑误差时,基于双层深度置信网络的结构损伤方法进行识别且结果精确;当噪声程度不超过10%时,定位识别结果准确率达100%;当噪声程度不超过15%时,定量识别结果最大绝对误差限不超过1.15%,识别结果准确;与传统的BP神经网络方法相比,本方法识别精度更高,抗噪性更强。 相似文献