Decentralized control of tall buildings

The response of large‐scale building structures can be actively reduced using an appropriate control algorithm and a number of actuators and sensors located within the building. This paper presents a decentralized control approach for controlling the response of tall buildings. The proposed method aims to divide the structural model into several substructures, each one performing on its own controller algorithm. This approach has already been used in controlling large‐scale systems such as power transmission systems and space structures. An instantaneous optimal control scheme is used as the control algorithm with different control feedbacks such as velocity feedback alone and displacement and velocity feedback. The stability issue of this method is addressed. A numerical example is used to investigate the performance of the proposed method compared to the case with centralized control. Copyright © 2005 John Wiley & Sons, Ltd.     

Optimized and decentralized pulse control of seismically excited steel structures

The idea of decomposing a centralized complicated system into several synchronous decentralized subsystems has been resulted into the development of decentralized control methods. In this study, a decentralized pulse control scheme is presented based on the theory of Inclusion Principle for steel systems comprising a multi-overlapping structure. The proposed control algorithm is basically an active control system that generates corrective pulses at each moment when the displacement or acceleration exceeds the predefined threshold. In order to evaluate the performance of proposed control system, a numerical study is conducted. The control system is implemented for two linear structural models of five- and twenty-story steel buildings. An optimization algorithm (PSO) is also used to determine the locations of required control inputs in the 20-story building. The results show that the proposed algorithm can substantially mitigate the structural response under different earthquake records (for 5-story model) and different subsystem configurations (for 20-story model).     

Decentralized Networked Control of Building Structures

The article presents a new method for the design of decentralized networked switched controllers to mitigate the response of building structures under earthquakes. It consists of two phases. The first phase generates low‐order gain matrices based on the linear quadratic Gaussian (LQG) control design. It includes a substructural approach when the equations of motion of substructures are extracted from the equation of motion of the overall structures described by a finite element in‐plane (2‐D) model. Appropriate model reduction procedures, determination of damping as well as the selection of sensor and actuators locations and models are applied to each substructure. The sensors and actuators are implemented into the design model. Both resulting reduced‐order state space models of substructures are used for the proper LQG control design. The obtained local controllers are implemented into the overall structure to evaluate the performance of the closed‐loop system. Displacement, drift, acceleration, maximal actuator forces as well as dynamic responses on selected locations are checked. The computational originality is the method derived under the subsequent second phase, where the gain matrices computed in the first phase serve as a tool for the design of decentralized networked switched controller. The switching is realized periodically between two switched modes. Each mode corresponds with only one active local feedback loop for a certain period of time. The network parameter is the time interval of activity of each mode determined by a given protocol. Robustness of performance against packet dropouts and sensor faults is tested. A numerical example of the decentralized networked switched controller design applied on the 20‐story high‐fidelity building benchmark model is supplied. The simulation tests show the proposed method exhibits acceptable performance.     

Decentralized robust control of building structures based on sliding mode theory

In this paper, a decentralized control algorithm is proposed for actively controlling the response of the flexible tall building structures under earthquake excitations. In the proposed approach, tall building structure was divided into some substructures in the form of state equation. The interaction of the subsystems and external excitations is conducted as bounded generalized force acting on the subsystems. A decentralized control algorithm of tall building structures is established based on the sliding model control theory. The control structure is described based on unit vector control. The control law consists of two parts: a linear control law u_L and a nonlinear law u_N. The linear control is merely a linear state feedback controller, whereas the nonlinear feedback controller incorporates the discontinuous or continuous nonlinear elements of the control law. Using the advantage of match conditions of sliding mode theory and the bounded feature of generalized force, the overall stability of decentralized control is also investigated. The actuator arrangement and matching conditions are discussed. The effectiveness of the proposed method is demonstrated by the numerical simulation of the decentralized control of a 20-story benchmark structure under seismic excitations.     

Decentralized sliding mode control of multistory buildings

In the present paper, a decentralized control algorithm is considered for actively controlling the response of flexible tall building structures under earthquake excitations. Sliding mode control is used as a base control algorithm and a decentralized scheme of that is presented. In the proposed approach, the building is divided into several substructures and for each subsystem a separate local control algorithm is designed. The control algorithm consists of the sliding surface and the reaching law equations. A numerical example is used to demonstrate the efficiency of the control algorithm in guaranteeing the system's global stability under the effects of subsystem interconnections and earthquake excitations. Copyright © 2007 John Wiley & Sons, Ltd.     

Decentralized Parametric Damage Detection Based on Neural Networks

In this paper, based on the concept of decentralized information structures and artificial neural networks, a decentralized parametric identification method for damage detection of structures with multi-degrees-of-freedom (MDOF) is conducted. First, a decentralized approach is presented for damage detection of substructures of an MDOF structure system by using neural networks. The displacement and velocity measurements from a substructure of a healthy structure system and the restoring force corresponding to this substructure are used to train the decentralized detection neural networks for the purpose of identifying the corresponding substructure. By using the trained decentralized detection neural networks, the difference of the interstory restoring force between the damaged substructures and the undamaged substructures can be calculated. An evaluation index, that is, relative root mean square (RRMS) error, is presented to evaluate the condition of each substructure for the purpose of health monitoring. Although neural networks have been widely used for nonparametric identification, in this paper, the decentralized parametric evaluation neural networks for substructures are trained for parametric identification. Based on the trained decentralized parametric evaluation neural networks and the RRMS error of substructures, the structural parameter of stiffness of each subsystem can be forecast with high accuracy. The effectiveness of the decentralized parametric identification is evaluated through numerical simulations. It is shown that the decentralized parametric evaluation method has the potential of being a practical tool for a damage detection methodology applied to structure-unknown smart civil structures.     

An adaptive numerical scheme based on the Craig–Bampton method for the dynamic analysis of tall buildings

A new numerical scheme is proposed to perform a nonlinear dynamic analysis for tall buildings. The structural components (beams and columns) of tall buildings gradually enter the inelastic phase under strong seismic excitation. Because the distribution of nonlinear components is initially unknown due to the randomness of earthquake inputs, a group of linear and nonlinear substructures are automatically figured out during the time‐history analysis of a structure. Then a modified Craig–Bampton method is proposed to condense the DOFs of the linear substructures in modal coordinates at each time step while keeping the governing equation of the nonlinear substructure in physical coordinates. The dominant modes of the linear substructures are selected to capture the main dynamic characteristics of the structure. The time step integration analysis is used to solve the governing equation of the structures in hybrid coordinates. A 20‐story building is employed as the numerical simulation test to validate the feasibility and effectiveness of the proposed numerical scheme. This scheme provides a new method for the nonlinear dynamic analysis of tall buildings with acceptable simulation accuracy and high computational efficiency.     

高层建筑地震反应的分散最优迭代学习控制研究

针对地震作用下高层建筑结构的振动控制方法进行研究,引入分散控制的策略,将线性二次型最优控制(LQR)与迭代学习控制(ILC)算法原理相结合,提出分散最优迭代学习控制(Decentralized OptimalIterative Learning Control,DOILC)算法并应用于高层结构振动控制中。对某20层钢结构Benchmark结构模型进行数值计算与分析,结果表明,采用分散控制策略的DOILC算法与传统集中控制策略一样可以有效地抑制结构地震反应,相对于集中控制的单一失效分散控制系统的可靠性更强且数据实时处理效率大为提高。     

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.     

Identification of Instantaneous Modal Parameter of Time‐Varying Systems via a Wavelet‐Based Approach and Its Application

This work presents an efficient approach using time‐varying autoregressive with exogenous input (TVARX) model and a substructure technique to identify the instantaneous modal parameters of a linear time‐varying structure and its substructures. The identified instantaneous natural frequencies can be used to identify earthquake damage to a building, including the specific floors that are damaged. An appropriate TVARX model of the dynamic responses of a structure or substructure is established using a basis function expansion and regression approach combined with continuous wavelet transform. The effectiveness of the proposed approach is validated using numerically simulated earthquake responses of a five‐storey shear building with time‐varying stiffness and damping coefficients. In terms of accuracy in determining the instantaneous modal parameters of a structure from noisy responses, the proposed approach is superior to typical basis function expansion and regression approach. The proposed method is further applied to process the dynamic responses of an eight‐storey steel frame in shaking table tests to identify its instantaneous modal parameters and to locate the storeys whose columns yielded under a strong base excitation.     

Building temperature regulation using a distributed model predictive control

This paper presents a predictive control structure for thermal regulation in buildings. The proposed method exploits the intermittently operating mode of almost all types of buildings. Usually the occupation profile can be known in advance and this fact will be used to reduce the energy consumption without decreasing the thermal comfort during the occupation. For that purpose, the predictive control strategy is first presented for a single zone building then extended to a multizone building example. Two opposite control strategies commonly exists: the decentralized control structure, which does not offer good performances especially when the thermal coupling among adjacent rooms is not negligible, and on the other hand, the centralized control for which the computational demand grows exponentially with the size of the system, being very expensive for large scale buildings. Our solution is based on a distributed approach which takes the advantages of both methods mentioned above. A distributed MPC algorithm with one information exchange per time step is proposed with good control performances and low computational requirements. Simulations and a comparison performance table end the article.     

地震作用下建筑结构的分散控制研究

对地震作用下建筑结构的振动控制方法进行研究,指出传统的集中控制策略在高层结构控制器设计中的局限性,阐述采用分散控制策略进行控制器设计的必要性及适用性.基于经典最优控制算法原理,导出两种分散控制算法--分散次优控制(Decentralized Sub-Optimal Control,DSOC)算法及分散经典最优控制(Decentralized Classic 0Dtimal Control,DCOC)算法.引入多市场概念,提出基于多市场机制的控制(Multi-Market Based Control,MMBC)算法并应用于高层结构分散控制中.对一高层受控结构进行数值计算与分析,结果表明采用分散控制策略与集中控制策略一样能有效地抑制结构振动反应,相对于集中控制的单一失效分散控制使系统的可靠性更强;显示MMBC算法较DSOC与DCOC算法具有参数选取简便、控制效果显著等优点,能较好地适用于高层结构分散控制器设计.     

基于分布式传感网络的结构模态识别方法

结构的模态参数识别是结构健康监测系统的基本任务。随着工程结构的日益大型化和复杂化,振动测试时需要布置大量的传感器。传统的集中采集和处理技术将难以胜任海量数据的处理要求,采用无线智能传感器的结构健康监测系统正是应运而生的新方向,而分布式采集和处理是其特点。在无线智能传感网络拓扑结构中采用分布式算法求解结构整体振型,利用随机子空间法识别各子结构模态,结合粒子群优化算法调整子振型获取结构整体振型。通过混凝土钢管拱桥模型试验验证了分布式算法的可行性,并利用模态置信度(MAC)对比分析了由分布式模态识别方法和集中式模态识别方法得到的结果,结果表明两种算法吻合较好。     

Shaking table model test of a mega‐frame with a vibration control substructure

A mega‐frame with a vibration control substructure (MFVCS) is a tuned mass damper system that converts substructures into a tuned mass. In this study, a kind of MFVCS using lead–rubber bearings (LRBs) to connect the vibration control substructure to the mega‐frame was proposed. To investigate the damping effect of this MFVCS, a series of shaking table tests were conducted, and the seismic responses of the MFVCS were compared with those of the traditional mega‐frame structure (TMFS). The results show that the seismic responses of the MFVCS are clearly smaller than those of the TMFS; additionally, the proposed MFVCS can provide a sufficient damping effect under different ground motions. Finite element (FE) models of the TMFS and MFVCS were established and validated by experimental results. Finally, the simulation results adopting different LRB models (equivalent linear and nonlinear elements) were compared, and the results indicate that simulation results can be obtained with greater accuracy from the FE model with a nonlinear LRB model than that with a linear LRB model.     

基于自适应RBF神经网络算法的建筑结构递阶分散控制研究

针对建筑结构振动控制的递阶分散控制问题进行研究。首先,通过设置全局控制器消除子系统间的关联耦合;在此基础上,结合Lyapunov稳定性理论和RBF神经网络理论设计了仅依赖于子系统位移和速度响应反馈信息的自适应控制律,并利用差分进化（DE）算法对自适应RBF神经网络局部子控制器相关参数进行优化,建立了适用于建筑结构振动控制的自适应RBF神经网络递阶分散控制（ARBFHDC）算法。对ASCE 9层Benchmark模型进行递阶分散控制设计、优化及仿真分析。结果表明,不同地震激励下,基于ARBFHDC算法设计的递阶分散控制较传统集中控制而言有更好的控制效果,且能保障各子系统作动器处于最大功效工作状态。     

Experimental evaluation of the seismic performance of steel MRFs with compressed elastomer dampers using large-scale real-time hybrid simulation

Real-time hybrid simulation combines experimental testing and numerical simulation, and thus is a viable experimental technique for evaluating the effectiveness of supplemental damping devices for seismic hazard mitigation. This paper presents an experimental program based on the use of the real-time hybrid simulation method to verify the performance-based seismic design of a two story, four-bay steel moment resisting frame (MRF) equipped with compressed elastomer dampers. The laboratory specimens, referred to as experimental substructures, are two individual compressed elastomer dampers with the remainder of the building modeled as an analytical substructure. The proposed experimental technique enables an ensemble of ground motions to be applied to the building, resulting in various levels of damage, without the need to repair the experimental substructures, since the damage will be within the analytical substructure. Statistical experimental response results incorporating the ground motion variability show that a steel MRF with compressed elastomer dampers can be designed to perform better than conventional steel special moment resisting frames (SMRFs), even when the MRF with dampers is significantly lighter in weight than the conventional MRF.     

Agent-based system identification for control-oriented building models

The paper presents a general agent-based system identification framework as potential solution for data-driven models of building systems that can be developed and integrated with improved efficiency, flexibility and scalability, compared to centralized approaches. The proposed method introduces building sub-system agents, which are optimized independently, by solving locally a maximum likelihood estimation problem. Several models are considered for the sub-system agents and a systematic selection approach is established considering the root mean square error, the parameter sensitivity to output trajectory and the parameter correlation. The final model is integrated from selected models for each agent. Two different approaches are developed for the integration; the negotiated-shared parameter model, which is a distributed method, and the free-shared parameter model based on a decentralized method. The results from a case-study for a high performance building indicate that the model prediction accuracy of the new approach is fairly good for implementation in predictive control.     

Chiller sequencing control with enhanced robustness for energy efficient operation

This paper presents a strategy for improving the reliability and the energy efficiency of chiller sequencing control based on the total cooling load measurement of centralized multiple centrifugal chiller plants. The improvement is achieved as follows. Firstly, a fused measurement of building cooling load is used to replace the direct/indirect measurement. Secondly, the maximum cooling capacity of individual chillers is computed online using a simplified centrifugal chiller model. Thirdly, the online computed maximum cooling capacity is calibrated according to the quality of the fused measurement in order to deal with the possible misbehaviours in measurement instruments. A simplified model for computing the maximum cooling capacity is developed and validated using field data. The performance of the proposed chiller sequencing control strategy is tested and compared with a conventional chiller sequencing control algorithm. Test results are presented showing that the proposed strategy can effectively improve the reliability of chiller sequencing control and reduce the energy consumption of chiller plants.     

Model‐Reference Health Monitoring of Hysteretic Building Structure Using Acceleration Measurement with Test Validation

A model‐reference health monitoring algorithm with two damage sensitive features is presented in this study, utilizing structural acceleration measurements from earthquake‐damaged building structure. A virtual linear healthy model, representing linear behavior of the instrumented structure, is used to generate real‐time reference response signals for health monitoring during a disastrous earthquake. The tracking error of acceleration and a relevant statistical factor are first proposed for identifying damage occurrence and location at story level. The severity of the hysteretic damage is estimated numerically using a model‐based prediction curve in an equivalent stiffness reduction manner with the implementation of robust Kalman filtering. The performance of damage detection and evaluation in the presented algorithm are illustrated by numerical simulation of structural models with different hysteretic characteristics, and further validated by experimental investigation employing a base‐isolated three‐story structure and real‐world case study of a seven‐story frame structure. The influence of measurement noise and uncertain stiffness in linear healthy model is also discussed through a parametric study.     

柱间连接型消能子结构设计方法与抗震性能试验

消能减震技术能显著提高建筑结构的抗震能力,但目前对消能子结构的性能目标和设计方法尚未达成共识。针对工程中较常见的柱间连接型消能子结构,提出基于多遇地震作用下弹性分析的设计方法,给出设计流程和设计要点。对1个钢筋混凝土纯框架子结构试件和2个消能子结构试件进行低周反复加载试验,考察消能子结构的抗震性能,检验所提设计方法的合理性。研究结果表明：消能子结构承载力约为纯框架与阻尼器承载力之和,单周耗能能力约等于纯框架与阻尼器各自耗能之和;阻尼器损伤累积直至完全剪断过程中,消能子结构承载力和刚度逐渐下降;阻尼器破坏后,周围框架抗震性能与加载同阶段的纯框架性能基本一致。按所提设计方法设计的消能子结构在罕遇地震作用下破坏模式合理,能保证阻尼器充分发挥消能效果;增大消能子结构纵向钢筋配筋量对改善其塑性铰状态和结构破坏模式无显著效果。