为准确高效探究结构参数不确定性对无筋砌体结构抗震性能的影响,文中提出基于元效应的全局敏感性分析方法结合地震易损性分析.以4幢3～6层典型无筋砌体结构为例,进行阻尼比对地震易损性的敏感性分析并与局部敏感性分析结果对比.结果表明结构参数相关性对敏感性结果影响较小,在合理取样下可用局部敏感性分析替代全局敏感性分析以提高计算效率.

In the construction of traditional public space in small cities and towns, because the "supply" of planning and design implementation does not match the "demand" of local residents, high growth and inefficiency of construction land are common, which seriously restricts the high-quality development of new-type urbanization. Based on this, combined with the design practice of Wukangqiao Community Center, we discuss how to optimize the public space of small towns and stimulate the space vitality of small towns through the construction of public facilities under the background of emphasizing stock development and based on the comprehensive analysis of the overall urban form. We also point out that the construction of public facilities in small cities and towns should be properly centralized and functionally mixed under the premise of balancing the interests and needs of all parties, so as to create conditions for the win-win situation of providing high-quality public services and market operation, and to promote the sustainable and healthy development of small towns. This project provides some reference for the construction of public facilities in small towns in rapid transformation and development, and also provides scientific basis for the organic renewal of small towns.

Stochastic subspace identification (SSI) and Bayesian methods are now the main representative approaches for high-performance system identification and uncertainty quantification. Both methods have been extensively used in engineering applications for the last 10 years despite them having quite different views in principle. However, no comparison has been carried out to inform which method is actually better for engineering applications in terms of modal identification accuracy, uncertainty quantification, and so forth. This paper therefore investigates the differences between these two quantifications of parameter uncertainty in system identification. Synthetic data for a six-degree-of-freedom spring-mass numerical system are first studied to compare their identification accuracy and applicable conditions. The investigation is then extended to field data from actual structures, which involves ambient modal testing of Heritage Court Tower in Canada, Canton Tower in China, and Ting Kau Bridge in Hong Kong. Results from the synthetic data show that, under white-noise excitation, identified modal parameters and quantified uncertainty for both methods are highly consistent with the values of frequentist statistics. However, when the excitation is not white noise, there may be some spurious modes identified via SSI, and the uncertainty quantified under colored-noise excitation via SSI is almost always larger than those under white noise, while the Bayesian method is not disturbed. Results for the structural field data indicate that in general applications under environmental excitation the identified modal parameters from both methods are almost identical; the quantified uncertainty of the SSI method is slightly larger than that of the Bayesian method, but they are of the same order of magnitude and can meet engineering requirements.

An automatic Bayesian modal identification method is proposed using the blind source separation (BSS) technique. The determination of resonant frequency bands, which is the initial step of the fast Bayesian FFT (fast Fourier transform) method, requires human intervention and hence, is labour-intensive and subjective. To automate the determination of resonant frequency bands, the BSS technique is introduced here for band selection process. After estimating the modal responses from measured data, the hump criterion curves are drawn to sharpen the border of the resonant humps. And the frequency bands can thus be determined automatically by locating the resonant humps with a peak picking algorithm. The proposed method was validated with a simulated 6- degree-of-freedom spring-mass model, a simulated 4-story benchmark model, the Heritage Court Tower in Vancouver, Canada. The robust identification results indicate that the proposed method can identify automatically and accurately the physical modes together with their uncertainty.

贝叶斯方法因其既能识别结构的模态参数,又能量化结果的不确定性,已经在工程界得到广泛应用.不同的动态加速度采样时长会对其识别结果及不确定性量化结果产生一定影响.文中以六自由度弹簧质量数值模型和广州新电视塔为研究对象,分别设置1、5、10、30、60min的动态加速度采样时长,利用快速贝叶斯FFT模态参数识别算法识别结构的模态参数及其不确定性,探究采样时长对结果的影响.分析结果表明,贝叶斯方法可以准确识别出结构的模态参数,并有效量化其不确定性;随着动态加速度采样时长的增加,模态参数的识别结果会愈加可靠,不确定性越来越小;综合考虑时间成本及识别结果的可靠度,推荐将动态加速度采样时长设置为10～30min.

结构在强震作用下进入弹塑性阶段,此时考虑将结构屈服周期代替结构一阶自振周期作为控制指标形成新双频段选波方法.新方法下,文中针对最大值与平均值两种方法分别选择6400组与12800组地震波计算了3、6、9、12层四个钢筋混凝土框架结构的地震响应,以结构最大层间位移角、顶点位移和基底剪力为分析指标进行大量系统性分析.结果 表明结构在强震作用下,新双频段选波方法比原双频段选波方法更加准确可靠,最后给出了新方法的工程实际选波建议.

A novel method for automatic identification of structural modal parameters is proposed, based on new developments in both uncertainty quantification for stochastic subspace identification and deep learning. An uncertainty diagram is first constructed to visualize uncertainty estimates, for clearly distinguishing spurious modes. Because the uncertainty of spurious modes is significantly larger than that of the real ones, a convolutional neural network (CNN) is adopted to automatically analyse the uncertainty diagram and efficiently determine the physical structural modes. The method is then applied to identify modal parameters for a six-degree-of-freedom spring–mass model, the Heritage Court Tower building in Canada, and the Ting Kau Bridge in Hong Kong. Results indicate for all three structures that the constructed CNN is effective for analysing the uncertainty diagram and can automatically and accurately obtain the real modes.

针对开洞约束砌体结构的抗震分析,目前国内外尚未提出简单有效的宏观单元模型.文中将等效框架模型引入约束砌体的建模中,详细介绍该模型在有限元软件OpenSEES中的实现方法,并结合一组实验原型验证其有效性.结果表明,模型pushover曲线与实验结果相符,该宏观单元模型可有效模拟约束砌体结构的整体地震响应.

Due to the high dispersion of structural parameters and high nonlinearity of seismic responses, it is particularly necessary to consider the structural parameter uncertainties in seismic fragility analysis of unreinforced masonry structures. Taking four unreinforced masonry structures with different storeys as the research object and the equivalent frame finite element models are established in OpenSees. The incremental dynamic analysis and the FOSM method are employed to investigate the impact of the uncertainties of ground motion and structural parameters on seismic fragility analysis. The analysis results indicate that the uncertainties of ground motion and structural parameters should be considered in the seismic fragility analysis of unreinforced masonry structures and the impact is greater when the structural seismic damage is severer. The impact of structural parameter uncertainty is almost the same as that of ground motion uncertainty. Compared with ground motion uncertainty, the impact of structural parameter uncertainty becomes even more obvious with the smaller number of structure storeys. The changing of median value of seismic fragility was investigated through sensitivity analysis, and it was found that the influence of structural damping ratio is 4 times that of other single parameter, demonstrating the highest sensitivity of seismic response to structural damping ratio.

A novel automatic operational modal analysis method is proposed based on the image recognition of stabilization diagrams with uncertainty quantification. The method not only enriches the contents of the stabilization diagrams to make them much clearer—it can also avoid heavy manual analysis of the stabilization diagrams by automatically obtaining operational modal parameters. In order to increase the efficiency in identifying modal parameters of structures, a traditional stabilization diagram is re-constructed to convey the uncertainty estimates. These stabilization diagrams are then resolved into single mode stabilization diagrams (SMSDs) with a specified frequency interval, for image recognition. Subsequently, a convolutional neural network (CNN) is adopted to automatically analyze the SMSDs. In this study, the CNN is trained by the SMSDs derived from the stabilization diagrams of two numerical examples and three engineering structures. The trained CNN is then validated with a 6 degree-of-freedom model, the Heritage Court Tower building, and the Ting Kau Bridge. The robust learning and prediction results establish that the constructed CNN is effective for analyzing the stabilization diagrams of different structures. It can automatically and accurately identify the physical modes on the stabilization diagrams, without extracting any characteristic parameters.

The first-order second-moment (FOSM), Latin hypercube sampling (LHS), and Monte Carlo simulation (MC) methods were applied to perform seismic fragility analyses of unreinforced masonry (URM) buildings, incorporating incremental dynamic analysis (IDA). The epistemic uncertainty caused by parameter uncertainty was also considered by all three methods, which therefore obtained more accurate results compared with the conventional IDA analysis that only considers the aleatory uncertainty caused by the record-to-record variability. Nine structural parameters were selected as the epistemic uncertainty inputs, and the URM buildings were modeled by the equivalent frame models in OpenSees. The analysis results indicated that the epistemic uncertainty was non-negligible because of the high dispersion of the structural parameters and high nonlinearity of the seismic responses of the URM buildings. At the same time, the epistemic uncertainty values derived by LHS-IDA and MC-IDA were slightly larger than that derived by FOSM-IDA as a result of the consideration of the nonlinear response. In the final comparison, the LHS-IDA was proven to be the most efficient and accurate way to consider the influence of epistemic uncertainty.

等效框架模型兼具了高计算效率与高计算精度的优点,是目前世界上抗震性能评估研究与工程领域使用最广泛的砌体结构建模工具.文中介绍了无筋砌体结构等效框架模型在有限元软件OpenSees中的实现,并以帕维亚大学1、2层、足尺无筋砌体结构原型地震试验为研究对象,对其有效性进行验证.结果 表明,计算pushover曲线与试验数据符合较好,文中的等效框架模型能够有效模拟无筋砌体结构的整体地震响应.

A decision tree–based seismic vulnerability method for reinforcement concrete frames is proposed. Structures with stories equal to 3, 6, 9, and 12 were considered herein and 45,360 (4 × 11,340) reinforcement concrete frame damage samples with different micro-characteristic values were simulated using the capacity spectrum method. Afterward, with the adoption of CART algorithm, a decision tree was derived to visualize the relationship between the structural characteristics and damage states according to training samples. Damage prediction can then be made for unseen structures according to their characteristic values directly using the configured decision trees. Ten training and testing sets were established randomly from the sample library and their seismic vulnerabilities under three earthquake intensity levels were assessed to verify the proposed method. The results show that the decision tree predictor is efficient for seismic vulnerability assessment of reinforcement concrete frames, and the predictor shows high prediction accuracy and stability.

合适数量的地震动是保证结构弹塑性时程分析结果可靠性的关键问题之一.选取符合规范要求的足量天然波和人工波作为地震输入,计算3、6、9、12层4个钢筋砼框架结构的地震响应,以结构最大层间位移角、顶点位移和基底剪力为分析指标,将小样本结果与大样本基准值进行比较,分析不同数量地震动输入下计算结果的可靠性.结果表明,当计算结果取多条地震动输入下的最大值时,“2+1”是最合适的组合,与当前规范对于弹性时程分析的规定相一致;当计算结果取平均值时,采用算术平均值和几何平均值未表现出明显差异,“8+4”为最佳的组合,是结果可靠性和计算工作量之间的最佳平衡点.

提出一种基于支持向量机(SVM)的钢筋砼(RC)框架结构快速地震易损性分析方法.以6层框架结构为例,选取柱尺寸、柱配筋率、混凝土标号、层高、横向跨度和梁高跨度比6个结构微观特征,对不同特征取值的5.67×104个结构进行Pushover分析,并应用能力谱法评估抗震性能,建立震害样本库;从样本库中选取训练集,引入机器学习SVM算法,建立结构微观特征与抗震性能之间的映射关系,进而可对给定特征取值的目标结构进行易损性预测.用样本库中的非训练集结构验证方法的准确性,结果表明:该方法具有较好的易损性预测准确率和稳定性,预测性能与训练集的样本数量及其有效性相关.

提出一种基于卷积神经网络(CNN)的稳定图自动分析方法.在获得不同结构的稳定图之后,按照各自的频率识别精度要求,将稳定图均分成若干个频带,得到单一模态稳定图作为CNN训练样本;通过平移、改变稳定点标记等技术手段对样本进行扩充,再将预处理好的训练样本代入CNN,通过跟踪损失函数在训练过程中变化规律,对如学习率等CNN参数进行调优,最终得到可自动判别稳定图中虚假模态的CNN;以3自由度弹簧质量数值模型、7自由度弹簧质量数值模型、以及一座钢筋混凝土框架结构大楼、瑞士Z24桥加速度实测数据验证了所搭建CNN模型的有效性.训练和预测结果表明,搭建的CNN亦可用于其他一般结构的稳定图自动分析,具有一定的通用性.在无需人为提取任何特征参数,也无需设定任何阈值的情况下,即可自动且准确、快速地剔除稳定图上的虚假模态.

Combing the analytical hierarchy process of decision analysis,a new earthquake loss assessment method based on fuzzy similarity theory was proposed.Using analogy analysis,structural seismic performances could be evaluated with no need of real structural models calculations.Thus the numerical computational efficiency of structural seismic vulnerability could be increased enormously.Taking RC frame structures as examples,the main structural factors that affect the structural seismic performance were selected firstly.Secondly the weight coefficients of factors were calculated by analytical hierarchy process,and the slope values of attribute function of those factors were calculated statistically respectively.At last,the similarity degree of two buildings was computed by fuzzy similarity theory and the seismic vulnerability of target structures could be evaluated with the aforementioned processes.The proposed method and the finite element method are implemented to three groups of representative structures respectively,and the precision and applicability of the proposed method are validated.

An iterative genetic-fuzzy clustering (IG-FC) methodology was proposed to automatically interpret the stabilization diagram.With a randomly set initial clustering center number,the methodology attempted to search out the clustering centers that can represent the real modes by using genetic and fuzzy clustering algorithms iteratively.Finally,the spurious modes could be excluded automatically from the stabilization diagram.Once the parameters were set,the real modes of structure could thus be recognized without any human interaction.The proposed methodology was verified by a 7-degree of freedom (DOF) spring-mass model and the accelerometer data of a reinforced concrete frame structure in Canada and the Z24 bridge.The robust identification results of modal frequency prove that the proposed IG-FC methodology can automatically and accurately eliminate the spurious modes on the stabilization diagram.What's more,the methodology can automatically distinguish and recognize the adjacent modes and weak incentive modes.

预测结构在罕遇地震作用下的非线性行为是结构基于抗震性能设计和评估中的关键问题,而其首要问题在于解决结构构件的非线性分析模型问题.本文介绍了结构非线性分析中三种常用的宏观单元模型,即基于刚度法的梁柱单元(DB)、基于柔度法的梁柱单元(FB)以及基于柔度法的塑性铰单元(PH)模型.基于PEER柱子试验数据库,将三种单元数值模拟结果与相应的试验结果进行对比,并以峰值强度之比、初始刚度之比和能量耗散能力之比三个指标统计评估了三种单元数值模拟的精确性.研究结果表明,PH单元计算结果的初始刚度之比和能量耗散能力之比均好于DB和FB单元,三种单元计算结果的峰值强度之比较接近,PH单元模拟精度最好.

The key of automatic modal parametric identification is automatically eliminating spurious modes from identified modes.Here, methods eliminating spurious modes in automatic modal parametric identification were reviewed.Aiming at how to automatically eliminate spurious modes, the existing automatic modal parametric identification methods were divided into three categories including automatic analysis methods of stabilization diagrams based on intelligence algorithms, true and false modes automatic distinction methods based on index threshold values and automatic analysis techniques getting clearer stabilization diagrams based on improved identification algorithms.Detailed and comprehensive introductions to principles of the above three methods were presented.The characteristics, problems and future study directions of these methods were discussed.At last, combined with the measured acceleration data, representative automatic identification methods were used to automatically eliminate false modes of a reinforced concrete structure.The identified results using these methods were compared.