Integrated multi‐type sensor placement and response reconstruction method for high‐rise buildings under unknown seismic loading

Structural health monitoring system has been implemented on high‐rise buildings to provide real‐time measurement of structural responses for evaluating their serviceability, safety, and sustainability. However, because of the complex structural configuration of a high‐rise building and the limited number of sensors installed in the building, the complete evaluation of structural performance of the building in terms of the information directly recorded by a structural health monitoring system is almost impossible. This is particularly true when seismic‐induced ground motion is unknown. This paper thus proposes an integrated method that enables the optimal placement of multi‐type sensors on a high‐rise building on one hand and the reconstruction of structural responses and excitations using the information from the optimally located sensors on the other hand. The structural responses measured from multi‐type sensors are fused to estimate the full state of the building in the modal coordinates using Kalman filters, from which the structural responses at unmeasured locations and the seismic‐induced ground motion can be reconstructed. The optimal multi‐type sensor placement is simultaneously achieved by minimizing the overall estimation errors of structural responses at the locations of interest to a desired target level. A numerical study using a simplified finite element model of a high‐rise building is performed to illustrate the effectiveness and accuracy of the proposed method. The numerical results show that by using 3 types of sensors (inclinometers, Global Positioning System, and accelerometers), the proposed method offers an effective way to design a multi‐type sensor system, and the multi‐type sensors at their optimal locations can produce sufficient information on the response and excitation reconstruction.     

Optimal Control of Adaptive/Smart Multistory Building Structures

This research advances creation of a new generation of adaptive/smart structures with self-modification capability by designing a predetermined number of members as actively controlled members. The computational model and high-performance parallel algorithms for optimal control of large structures recently developed by the authors are applied to multistory buildings. Four different schemes for placement of controllers are investigated. Three types of dynamic loadings are considered: earthquake ground motion, periodic impulsive horizontal wind loading on the exterior joints of the structure, and asymmetric periodic impulsive wind loading on the exterior of the structure modeling a twister. Results are presented for three multistory building structures with curved beams and setback representing both space moment-resisting and braced frames. It is demonstrated that through a proper selection of the weighting factor, the response of a multistory building structure can be reduced substantially to a fraction of the response of the uncontrolled structure at a practically feasible maximum required actuator force. Recommendations are made on the placement of the controllers for various types of structural configurations.     

Optimal sensor placement methods and metrics – comparison and implementation on a timber frame structure

The current work aims at determining optimal sensor configurations for the modal identification of a post-tensioned timber frame structure. The objective is to maximise the information gained from the structural testing, while keeping the number of necessary sensors to a minimum. Three different and widely used optimal sensor placement (OSP) methods are investigated and evaluated based on appropriate metrics, namely the effective independence method, the modal kinetic energy method and the information entropy (IE) method. An enhanced IE variant is adopted, which resolves the problem of close positioning of sensors, which forms a common issue when sensor positions are selected from a dense grid. In terms of the adopted metrics, three different options are investigated, namely, the information entropy index, the Modal Assurance Criterion and a newly introduced metric, the relative dispersion index. For the quantification of trade-offs among the selected metrics a Pareto Front scheme is realised. The study indicates that the evaluation of different OSP configurations is strongly dependent on the employed metric. It is therefore of vital importance to select the appropriate metric when determining optimal sensor positions.     

Shaking table model test and numerical analysis of a long‐span cantilevered structure

This paper presents an earthquake‐resistance study program of a long‐span cantilevered story building. The program consists of a shaking table test study and nonlinear seismic analysis using finite element modeling technique. A 1/30 scale model of the prototype structure was designed and manufactured and then tested via the shaking table facility. Dynamic responses of the prototype structure under different earthquake excitation loadings were simulated. Dynamic properties, acceleration, and deformation responses of the scale down model under different intensity levels of earthquake were studied. The dynamic behavior, cracking pattern, and the likely governing failure mechanism of the structure were analyzed and discussed as well. The seismic responses of the prototype building were deduced and analyzed in terms of the similitude law. Furthermore, elaborate finite element models were established, and nonlinear numerical analysis of the prototype structure was conducted. The errors in the seismic response of the structure caused by structural simplification of scale down modeling are found small, and the dynamic behavior of the structure was not altered in the earthquake excitations. This test study provides a benchmark to calibrate the finite element model and a tentative guide in seismic design of such long‐span cantilevered story buildings.     

Full dynamic analysis of Tehran telecommunication tower; linear and nonlinear responses

Dynamic response of large concrete tower structures with passive or active damping is important in terms of performance under dynamic loads. Structural instability mostly arises as a result of losing the energy capacity in the system, which forces the system to act as a mechanism. The purpose of this paper is to present three different finite element models to investigate the linear and nonlinear response of the Tehran telecommunication tower. The tower is ranked the fourth tallest structure in the world and has been under construction since 1997. Using a mathematical model, an efficient method is used to take into account the cracking and crushing of the reinforced concrete material as well as large deformation effects. The linear and nonlinear dynamic characteristics of the large tower may be defined by using an advanced formulation for the concrete element. Then, the proposed advanced element is used to model a full‐scale finite element mesh to derive the dynamic response of the Tehran telecommunication tower structure with use of a computer program. Copyright © 2001 John Wiley & Sons, Ltd.     

Impact wave propagation in tall buildings using advanced spectral element method

One of the major divisions in analysis and design of tall buildings is to account for forces induced by flying‐object impact and the subsequent progressive collapse. During the past three decades, problems of contact mechanics have been considered by some authors, and special attention has been devoted to high‐speed impact. Impact wave propagation in tall buildings may be analysed using the conventional finite element method. However, in order to guarantee stability and accuracy of the solution, the number of elements used to model the structure may be very large indeed; more precisely, accurate results can be obtained after a substantial computational effort. In this respect, an improved complex spectral element method is developed for analysing the wave propagation problems in large structures. The purpose of this paper is to obtain the resulting dynamic responses of a tall building induced by a high‐speed impact of a flying object. Copyright © 2003 John Wiley & Sons, Ltd.     

Substructure Identification and Health Monitoring Using Noisy Response Measurements Only

Abstract:   A probabilistic substructure identification and health monitoring methodology for linear systems is presented using measured response time histories only. A very large number of uncertain parameters have to be identified if one considers the updating of the entire structure. For identifiability, one then would require a very large number of sensors. Furthermore, even when such a large number of sensors are available, processing of vast amount of the corresponding data raises computational difficulties. In this article a substructuring approach is proposed, which allows for the identification and monitoring of some critical substructures only. The proposed method does not require any interface measurements and/or excitation measurements. No information regarding the stochastic model of the input is required. Specifically, the method does not require the response to be stationary and does not assume any knowledge of the parametric form of the spectral density of the input. Therefore, the method has very wide applicability. The proposed approach allows one to obtain not only the most probable values of the updated model parameters but also their associated uncertainties using only one set of response data. The probability of damage can be computed directly using data from the undamaged and possibly damaged structure. A hundred-story building model is used to illustrate the proposed method.     

A cross‐scale refined damage evolution analysis of large commercial aircraft crashing into a nuclear power plant

Since the 9/11 attacks, whether nuclear power plants (NPPs) can withstand the malicious impact of large commercial aircraft has become a question that must be considered in the design phase. Refined model and fine‐scale element reflect more accurately the failure procedure and behaviour. However, due to the exceedingly complex structure of NPP, the traditional meshing method requires a substantial number of geometric operations involving point, line, area, and volume, leading to a tedious modelling process and very high computational costs, as well as making it difficult to modify and optimize the model. This paper combines the scaled boundary finite element method with the octree technique to refine nuclear engineering damage evolution analysis and establish a refined numerical model of the Generation III⁺ NPP, considering the detailed equipment hatch, air intake, and internal steel containment vessel structures. Subsequently, the refined damage evolution analysis of a large commercial aircraft crashing into an NPP was developed. The mesh size sensitivity, impact region shape comparison, and the influence of different impact heights were discussed. The results indicate that the cross‐scale refined meshing and analysis method provides a high‐quality discrete grid with fewer elements. Furthermore, this method is highly flexible, more accurately simulating the damage evolution and gradual destruction process. We recommend future structural optimization for the air intake and conducting detailed analyses at this location. The cross‐scale analysis method presented in this paper enables a rapid, refined simulation of a malicious impact by a large commercial aircraft. Additionally, it provides technical support for future studies of the responses of NPPs' internal structures, systems, and equipment under extreme earthquake and other disaster conditions.     

Predicting acceleration response of super-tall buildings by support vector regression

Recovering missing data of defective sensors is an important challenge for reliability of structural health monitoring systems and misjudgment of structural conditions. The present study concerns predicting corrupted data of lost sensors by support vector regression (SVR). The method is tuned via optimizing their parameters by observer–teacher–learner-based optimization as a powerful meta-heuristic algorithm. Their performances are compared in predicting the acceleration responses of two real-world super-tall buildings: Milad Tower, located in Tehran, and Canton Tower in Guangzhou. Also the minimum required of sensors to predict the acceleration responses are investigated. The results are evaluated by five statistical indices exhibiting that the optimized SVR has sufficient capacity to predict acceleration responses of both towers with limited number of sensors. The proposed method is of practical interest as it does not require finite element modeling of the structure to derive its dynamic responses.     

Sensor Attitude Correction of Wireless Sensor Network for Acceleration‐Based Monitoring of Civil Structures

Wireless sensors are now becoming practical alternatives to traditional wired sensors in monitoring civil structures. Though many have been reported on acceleration‐based monitoring of civil structures using wireless sensor networks, sensor attitude that may be different from instrumentation plan has been a seemingly overlooked issue behind performance validation of the network. In this article, a technique to correct the sensor attitude is proposed for the wireless sensor network that measures 3D acceleration of civil structures. Six simple formulas to assess the well‐known 3D Euler angles (i.e., roll, pitch, and yaw) are derived using the gravity extracted from measured 3D acceleration and nonchanging direction of sensors on a stationary structure. The proposed technique is validated at a large‐scale wireless sensor network with 22 sensors in the respective attitudes on a truss bridge. First, attitudes assessed by the proposed method are compared with instrumentation plan. Then, mode shapes obtained before and after the correction are compared with those from finite element model. Comparison shows that quality of the mode shapes improves significantly by small amount of attitude correction less than 7°.     

Uncertainty and Sensitivity Analysis of Damage Identification Results Obtained Using Finite Element Model Updating

Abstract: A full‐scale seven‐story reinforced concrete shear wall building structure was tested on the UCSD‐NEES shake table in the period October 2005–January 2006. The shake table tests were designed so as to damage the building progressively through several historical seismic motions reproduced on the shake table. A sensitivity‐based finite element (FE) model updating method was used to identify damage in the building. The estimation uncertainty in the damage identification results was observed to be significant, which motivated the authors to perform, through numerical simulation, an uncertainty analysis on a set of damage identification results. This study investigates systematically the performance of FE model updating for damage identification. The damaged structure is simulated numerically through a change in stiffness in selected regions of a FE model of the shear wall test structure. The uncertainty of the identified damage (location and extent) due to variability of five input factors is quantified through analysis‐of‐variance (ANOVA) and meta‐modeling. These five input factors are: (1–3) level of uncertainty in the (identified) modal parameters of each of the first three longitudinal modes, (4) spatial density of measurements (number of sensors), and (5) mesh size in the FE model used in the FE model updating procedure (a type of modeling error). A full factorial design of experiments is considered for these five input factors. In addition to ANOVA and meta‐modeling, this study investigates the one‐at‐a‐time sensitivity analysis of the identified damage to the level of uncertainty in the identified modal parameters of the first three longitudinal modes. The results of this investigation demonstrate that the level of confidence in the damage identification results obtained through FE model updating, is a function of not only the level of uncertainty in the identified modal parameters, but also choices made in the design of experiments (e.g., spatial density of measurements) and modeling errors (e.g., mesh size). Therefore, the experiments can be designed so that the more influential input factors (to the total uncertainty/variability of the damage identification results) are set at optimum levels so as to yield more accurate damage identification results.     

强震作用下超高层建筑动力响应的分析

采用大型有限元软件ANSYS对阜矿高层框架—剪力墙结构建模,通过对模型输入天津波,对高层结构在地震荷载作用下结构各层的变形情况和各层结构之间的变形量进行动力学分析。结果表明:高层框架—剪力墙结构,剪力墙对于中底部侧向位移有很大的限制作用,较上部结构对侧向位移限制能力强。     

基于径向基神经网络的桥梁有限元模型修正

基于某预应力混凝土大跨刚构-连续梁桥的ANSYS有限元模型,提出一种基于径向基神经网络的有限元模型修正方法。该方法以不同设计参数条件下有限元模型模态分析频率作为输入向量,以对应的桥面单元、中墩、边墩的弹性模量、密度等设计参数修正值作为输出向量,利用径向基神经网络来逼近两者之间的非线性映射关系。结合该桥梁结构健康监测系统中加速度传感器监测的桥梁结构动力反应的加速度数据,利用神经网络的泛化特性,直接计算出有限元模型设计参数的修正值。研究结果表明:修正后的有限元模型能更真实地反映结构的物理状态,较好地反映该桥梁结构的真实动力特性。     

Seismic damage identification of moment frames based on random forest algorithm and enhanced gray wolf optimization

The present study aims to identify damage in two-dimensional (2-D) moment frames using seismic responses by combining the random forest (RF) machine classifier and the enhanced gray wolf optimizer (EGWO) metaheuristic algorithm. First, a 2-D moment frame for the dynamic analysis is simulated using the finite element method (FEM). Then, the placement of sensors is optimized using a proposed optimal sensor placement (POSP) method, which is a combination of the iterated improved reduced system (IIRS) and the binary differential evolution (BDE) optimization algorithm. The acceleration responses of the moment frame having damaged elements under 1995 Kobe earthquake are measured at the optimal sensor placement. Then, the natural frequencies and mode shapes of the structure are extracted using the auto-regressive model with exogenous input method (ARX) as a system identification method. The natural frequencies are exploited to train an RF machine learning network that can find the damaged story of the moment frame. Then, EGWO is employed to accurately locate and quantify the damaged elements of the structure. The efficiency of the proposed method is assessed through considering a six-story frame with 18 elements, a seven-story frame with 49 elements, and the experimental data of an eight-story frame for various conditions. The results show that the RF algorithm has an outstanding performance to correctly find a damaged story. Furthermore, the location and severity of damaged elements are precisely determined by EGWO algorithm. As a final outcome, it is demonstrated that the two-step proposed method is very effective in seismically identifying damage to such structures.     

Two-dimensional computational framework of meso-scale rigid and line interface elements for masonry structures

In this paper, rigid elements along with nonlinear line interface elements are utilized to model masonry structures. The modeling approach focuses on two dimensions (2D) whereby the in-plane behavior of components is represented by rigid elements and nonlinear line interfaces instead of modeling by a traditional finite element method. In this approach, the component will be allowed to crack in predefined paths which have more likelihood for propagation. The paper discusses the model derivation and implementation. Moreover, the mesh sensitivity of this method is assessed by using different mesh sizes, and it is shown that the model captures response obtained by the experimental tests. The traditional finite element method is indeed capable of predicting the behavior of large scale masonry component, but the computational time is very high. In this study it has been shown that using rigid elements along with nonlinear line interfaces leads to a reduced number of degrees-of-freedom, which consequently reduces the computational time. The material model is implemented in a user-defined subroutine that is compiled with DIANA. The algorithms and material models are validated with well-documented experimental studies, and results clearly show the capabilities of the proposed procedures.     

A graph deep learning method for short‐term traffic forecasting on large road networks

Short‐term traffic flow prediction on a large‐scale road network is challenging due to the complex spatial–temporal dependencies, the directed network topology, and the high computational cost. To address the challenges, this article develops a graph deep learning framework to predict large‐scale network traffic flow with high accuracy and efficiency. Specifically, we model the dynamics of the traffic flow on a road network as an irreducible and aperiodic Markov chain on a directed graph. Based on the representation, a novel spatial–temporal graph inception residual network (STGI‐ResNet) is developed for network‐based traffic prediction. This model integrates multiple spatial–temporal graph convolution (STGC) operators, residual learning, and the inception structure. The proposed STGC operators can adaptively extract spatial–temporal features from multiple traffic periodicities while preserving the topology information of the road network. The proposed STGI‐ResNet inherits the advantages of residual learning and inception structure to improve prediction accuracy, accelerate the model training process, and reduce difficult parameter tuning efforts. The computational complexity is linearly related to the number of road links, which enables citywide short‐term traffic prediction. Experiments using a car‐hailing traffic data set at 10‐, 30‐, and 60‐min intervals for a large road network in a Chinese city shows that the proposed model outperformed various state‐of‐the‐art baselines for short‐term network traffic flow prediction.     

基于有限测点信息的大型空间网架结构智能工作状态评估技术研究

针对大型空间网架结构的特点,建立基于有限测点信息的智能工作状态评估技术。该技术是首先通过测量少数网架结构杆件的响应信息,获得结构响应的峰因子;再通过安置在网架结构上的风压力传感器获得结构的荷载信息,采用BP神经网络法和加权荷载本征模态分解法分别精确识别作用在网架结构上的平均风荷载和脉动风荷载;并通过安置在网架结构上的加速度传感器获得结构在环境激励下的加速度响应信息,对结构进行有限元模型修正;然后采用正向演绎的方法,得到整个网架结构的工作状态,根据预先设定的安全级别就能对其进行工作状态的评估。最后,通过该技术成功地应用于深圳市民中心屋顶网架结构的智能健康监测系统中,说明该技术的可行性和有效性。     

An integrated simulation method for coupled dynamic systems

Partitioned methods have been widely used in multiphysics and large‐scale structure‐media problems since they allow decomposition of a complex system into smaller subsystems. Although they have been considered to be superior to monolithic methods in terms of software reuse, difficulties still exist in the implementation process. This paper addresses these difficulties and proposes a new method to ease the coupling of the dynamic subsystems analyzed with different finite element codes. This is enabled by the development of a new staggered approach such that each involved program acts as a black box that is accessible only through model input and output, that is, displacements and forces, at the interface boundary. The accuracy and stability of the proposed method are numerically evaluated. A practical method to determine the maximum time step for stable solutions is also proposed. Two application examples are presented to verify the algorithm and demonstrate potential of the proposed method.     

Response covariance-based sensor placement for structural damage detection

One important function of a structural health monitoring system is to detect structural damage in a structure. However, this is a very challenging task since the measurement is often incomplete in a civil structure due to a limited number of sensors. This paper presents a response covariance-based sensor placement method for structural damage detection with two objective functions for optimisation. The relationship between the covariance of acceleration responses and the covariance of unit impulse responses of a structure subjected to multiple white noise excitations is first derived. The response covariance-based damage detection method is then presented. Two objective functions based on the response covariance sensitivity and the response independence are, respectively, formulated and finally integrated into a single objective function for optimal sensor placement. Numerical studies are conducted to investigate the feasibility and effectiveness of the proposed method via a three-dimensional frame structure. Numerical results show that the proposed method with the backward sequential sensor placement algorithm is effective for damage detection.     

Seismic analysis of Tehran Telecommunication Tower using complex fractional modulus

Seismic response of large concrete tower structures with passive or active damping is important in terms of performance under earthquake loads. The conventional finite element method has been used successfully in linear and nonlinear analyses in large concrete structures. The method can be performed by subdividing the large structure into small uniform elements having approximate shape functions. Although this replaces a single complicated structural system with a number of simple uniform elements, in cases of tall concrete tower structures with cracking and crushing behaviour in the concrete material and yielding in the reinforcement, the computer time and memory can be large. Hence, it is desirable to search for a procedure requiring fewer elements and also less computer time and effort to model the structure. In this respect, attention is paid to the advanced complex damped spectral element method, which benefits from the more accurate and also mathematically complicated shape functions. Use of the advanced spectral element method can help engineers to design a complex structure, such as a tall concrete tower, with lower cost and lower weight. Using a computer program, the proposed formulation has been used to derive the nonlinear dynamic response of the 435‐m Tehran Telecommunication Tower. Copyright © 2002 John Wiley & Sons, Ltd.