Tuned mass damper parameters design for structural systems via linear matrix inequality and genetic algorithm

Falu Weng1 , Yuanchun Ding2 , Weijie Mao3

1School of Electrical Engineering and Automation, Jiangxi University of Science and Technology, Ganzhou 341000, China

2Jiangxi Key Laboratory of Mining Engineering, Jiangxi University of Science and Technology, Ganzhou 341000, China

3State Key Lab of Industrial Control Technology, Institute of Cyber-Systems and Control, Zhejiang University, Hangzhou 310027, China

2Corresponding author

Journal of Vibroengineering, Vol. 18, Issue 6, 2016, p. 3768-3781. https://doi.org/10.21595/jve.2016.16629
Received 9 November 2015; received in revised form 30 April 2016; accepted 28 June 2016; published 30 September 2016

Copyright © 2016 JVE International Ltd. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Creative Commons License
Table of Contents Download PDF Acknowledgements References
Cite this article
Views 351
Reads 156
Downloads 1025
Abstract.

The tuned mass damper parameters designing for structural systems based on combining linear matrix inequality with genetic algorithm is of concern in this paper. Firstly, based on matrix transform, the novel model description with a singular style for structural systems is obtained, in which the possible coupling of those uncertainties is avoided. Secondly, an approach, which combines linear matrix inequality with genetic algorithm, is taken in this work to solving the optimization problems, and the optimized tuned mass damper parameters can be obtained by solving the optimization problems such that the tuned-mass-damper-controlled systems have a prescribed level of vibration attenuation performance. Furthermore, the obtained results are also extended to the uncertain cases. Finally, the effectiveness of the obtained theorems is demonstrated by numerical simulation results.

Keywords: tuned mass damper, linear matrix inequality, genetic algorithm, structural control, vibration attenuation.

1. Introduction

Since the concept of structural control was proposed by Yao et al. in 1972 [1], the research on structural control has been conducted by many scholars, and great strides have been made in advancing the theory and practice in this area. Generally speaking, the existing results can be classified into three types: passive control (without external energy input), semi-active or active control (need energy input) and intelligent-algorithm-based intelligent control [2, 3]. Active control can achieves a satisfying control result [4, 5], however, it needs continuous external energy supply which results in its low reliability. On the contrary, passive control needs no external energy input, thus, it can reach a much higher reliability. Tuned Mass Damper (TMD), as one kind of passive control device, has received considerable attention for its virtues, such as no energy consumption, low cost, easy installation, etc., and many results about TMD control have been achieved during the last decades, for example, the Trans-Tokyo Bay Highway Crossing, completed in1997, is 11 km in total length, and the vibration was significantly reduced by TMDs inside the box girder [6]. In order to diminish the vibrations in the structure during earthquakes and typhoons, the main tower of the Akashi-Kaikyo suspension bridge also contains TMDs [7]. There were two buildings in U.S. equipped with TMD, one is Citicorp Center, New York and the other is John Hancock Tower, Boston [8]. Moreover, 18 vertical and 18 lateral TMDs were installed in Dubai Meydan Racecourse Stadium to control the vibration induced by wind load in the two directions, and the results demonstrated a substantial reduction of the vertical and lateral wind vibration [9]. Some more achievements about TMD control can be found in [10-17] and those references therein. The TMD system is a well-accepted device for controlling flexible structures, particularly, tall buildings [10-12]. However, to the best of the authors’ knowledge, most of the TMD applications have been made towards mitigation of wind-induced motion, and seismic effectiveness of TMDs still has not been fully investigated. A typical kind of TMD consists of a mass block, a viscous damper and a spring connected to the main structure. The natural frequency of the TMD is tuned to the resonant frequency of the main structure, so, a large amount of entrance energy is transferred to the TMD [18]. Obviously, the performance of the TMD is based on the design of its parameters: mass, stiffness, and damping ratio.

The most common TMD designing methods are LQR, LQG, sliding mode control, pole assignment, H control, energy-to-peak control, fuzzy control, and so on [19-23]. Moreover, in order to achieve a better control performance, some optimization techniques are often used in optimizing those TMD parameters [4, 10, 24-26]. Such as, in [10], the parameters of the TMD were optimally designed using multi-objective genetic algorithms for a 12-story realistic building through both deterministic and robust design procedures. Yang et al. proposed an innovative practical approach to optimally design the TMD system in [26], and the effectiveness was illustrated by examples. It is worth pointing out that genetic algorithm has some virtues, such as global genetic optimization etc., and has been widely used for the parameter optimization in control systems. On the other hand, in recent years, LMI technique is widely researched and used in the system stability analysis and controller design. For example, based on LMI, references [27, 28] concerned the vibration-attenuation controller design for linear structure systems. In terms of LMI, Wu et al. [29] discussed the admissibility and dissipativity of singular systems, and some improved results were obtained. Some more results about LMI can be found in [30-32] and those references therein. Although the LMI technique has been widely used in the system stability analysis and controller design, LMI-based TMD parameters design also has not been fully investigated. Thus, there is still much room for improvement in this important issue.

In this paper, the parameters of the TMD will be optimized by combining LMI technique with genetic algorithm. First, the novel model description for uncertain structural systems is obtained by introducing the singular system description, in which the coupling of uncertainties is avioded when the mass, damping and stiffness are subjected to possible perturbations. Then, in terms of combining LMI technique with genetic algorithm to solve the optimization problem, the optimized TMD parameters can be easily achieved such that the TMD-controlled system has a prescribed level of vibration attenuation performance, and the obtained results are also extended to the uncertain cases. In the end, the effectiveness of the obtained theorems is illuminated by numerical examples.

Notation: Throughout this paper, for real matrices X and Y, the notation XY (respectively, X>Y) means that the matrix X-Y is semi-positive definite (respectively, positive definite). I is the identity matrix with appropriate dimension, and Ik is the k×k identity matrix. 0n×m and 0k represent the n×m zero matrix and the k-dimensional zero vector, respectively. The superscript “T” represents transpose. x2 expresses the 2-norm of x. We define MH=MT+M. For a symmetric matrix, * denotes the symmetric terms. The symbol Rn stands for then-dimensional Euclidean space, and Rn×m is the set of n×m real matrices.

2. Problem formulation and dynamic models

Consider anndegree-of-freedom structural system with k TMDs installed on k storeys respectively, which are shown in Fig. 1, where mj j=1,2,,n, cj j=1,2,,n and kjj=1,2,,n are the mass, damping and stiffness of storeyj, respectively; mdi i=1,2,,k, cdi i=1,2,,k andkdii=1,2,,k are the mass, damping and stiffness of TMDi, respectively. The structural system subjected to horizontal earthquake excitation can be expressed in the following model equations:

(1)
M - x - ¨ ( t ) + C - x - ˙ ( t ) + K - x - ( t ) = i = 1 k B i c d i x ˙ d i + k d i x d i + B - ω ω ( t ) ,
(2)
m d i x ¨ d i t + c d i x ˙ d i t + k d i x d i t = - m d i B i x - ¨ t - m d i ω t ,         i = 1,2 , , k ,

where: 

M - = d i a g m 1 , m 2 , m n ,         B - ω = - m 1 ,     m 2 , , m n T ,
C - = c 1 + c 2 - c 2 0 0 0 0 - c 2 c 2 + c 3 - c 3 0 0 0 - c n 0 0 0 0 - c n c n ,
K - = k 1 + k 2 - k 2 0 0 0 0 - k 2 k 2 + k 3 - k 3 0 0 0 - k n 0 0 0 0 - k n k n .

Fig. 1. n degree-of-freedom structural system controlled by TMDs

n degree-of-freedom structural system controlled by TMDs

M - , K-, C-, are the mass, damping and stiffness matrices of the structure, respectively. B-ω is the external excitation vector. x-(t)=xm1(t),,xmj(t),,xmn(t)T, xmj(t) is the relative displacement of storey j to the ground, xdi i=1,2,,k donates the relative displacement of the TMDi to the storey j, on which the TMDi is installed. Bi=00,1,00T is a vector with the jth item to be 1 and others to be 0, which means TMDi is installed on storey j. Based on Eqs. (1)-(2), the following system equation is obtained:

(3)
M x - ~ ¨ ( t ) + C x - ~ ˙ ( t ) + K x - ~ ( t ) = B ω ω ( t ) ,

where x-~(t)=xm1(t),xm2(t),,xmn(t),xg1(t),xg2(t),,xgk(t)T,xgi(t)denote the relative displacement of the ith TMD to the ground. Obviously xgi(t)=xdi(t)+xmj(t), and:

B ω = B - ω 0 k - i = 1 k m d i 0 n + i - 1 1 0 k - i ,         M = M - 0 n × k 0 k × n 0 k × k + i = 1 k m d i 0 n + i - 1 1 0 k - i 0 n + i - 1 1 0 k - i T ,
C = C - 0 n × k 0 k × n 0 k × k + i = 1 k c d i 0 j i - 1 1 0 n - j i + i - 1 - 1 0 k - i 0 j i - 1 1 0 n - j i + i - 1 - 1 0 k - i T ,
K = K - 0 n × k 0 k × n 0 k × k + i = 1 k k d i 0 j i - 1 1 0 n - j i + i - 1 - 1 0 k - i 0 j i - 1 1 0 n - j i + i - 1 - 1 0 k - i T .

In terms of Eq. (3), it is easy to obtain the following singular description of the system model:

(4)
E x ˙ t = A x t + H ω ω t ,       z ( t ) = C z x ( t ) ,

where x(t)=x-~(t)T,x-~˙(t)T,x-~¨(t)TT, Cz is a problem based matrix with suitable dimension, and:

E = I 2 n + k 0 2 n + k × n + k 0 n + k × 2 n + k 0 n + k × n + k ,       H ω = 0 2 n + k B ω 0 k - i = 1 k m d i 0 3 n + 2 k + i - 1 1 0 k - i ,
A = 0 n + k × n + k I n + k 0 n + k × n + k 0 n + k × n + k 0 n + k × n + k I n + k - K - 0 n × k 0 k × n 0 k × k - C - 0 n × k 0 k × n 0 k × k - M - 0 n × k 0 k × n 0 k × k
            - i = 1 k k d i 0 2 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 k - i 0 j i - 1 1 0 n - j i + i - 1 - 1 0 2 n + 3 k - i T - i = 1 k c d i 0 2 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 k - i 0 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 n + 2 k - i T
            - i = 1 k m d i 0 3 n + 2 k + i - 1 1 0 k - i 0 3 n + 2 k + i - 1 1 0 k - i T .

Consider the uncertainties existing in the mass, damping and stiffness matrices of the structural systems, and the uncertainties satisfy mjΔ=mj+θ1jmj, kjΔ=kj+θ2jkj, cjΔ=cj+θ3jcj, mdiΔ=mdi+θ4imdi, kdiΔ=kdi+θ5ikdi, cdiΔ=cdi+θ6icdi, where j= 1, 2,…, n, i= 1, 2,…, k, and 0θlsθ-ls<1. s= 1, 2,…, n, while l= 1, 2, 3, and s= 1, 2,…, k, while l= 4, 5, 6. Then, the uncertain structural systems can be expressed in the following form:

(5)
E x ˙ t = A Δ x t + H ω Δ ω t ,         z t = C z x t ,

where:

H ω Δ = H ω - j = 1 n θ 1 j m j 0 2 n + 2 k + j - 1 1 0 n + k - j - i = 1 k θ 4 i m d i 0 3 n + 2 k + i - 1 1 0 k - i ,
A Δ = A + θ 21 k 1 0 2 n + k 1 0 n + k - 1 1 0 3 n + k - 1 T + j = 2 n θ 2 j k j 0 2 n + k + j - 1 1 - 1 0 n + k - j - 1 0 j - 1 1 - 1 0 3 n + 3 k - j - 1 T
            + θ 31 c 1 0 2 n + k 1 0 n + k - 1 0 n + k 1 0 2 n + k - 1 T + j = 2 n θ 3 j c j 0 2 n + k + j - 1 1 - 1 0 n + k - j - 1 0 n + k + j - 1 1 - 1 0 2 n + 2 k - j - 1 T
            + j = 1 n θ 1 j m j 0 2 n + k + j - 1 1 0 n + k - j 0 2 n + k + j - 1 1 0 n + k - j T + i = 1 k θ 5 i k d i 0 2 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 k - i 0 j i - 1 1 0 n - j i + i - 1 - 1 0 2 n + 3 k - i T
            + i = 1 k θ 6 i c d i 0 2 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 k - i 0 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 n + 2 k - i T + i = 1 k θ 4 i m d i 0 3 n + 2 k + i - 1 1 0 k - i 0 3 n + 2 k + i - 1 1 0 k - i T .

Remark 1: It has been shown in the references [13, 33] that the linear structural system Eq. (3) can be described in the following state equation:

(6)
q ˙ t = A q t + H ω ω t ,       z t = C z q ( t ) ,

where:

A = 0 n × n I n - M - 1 K - M - 1 C 2 n × 2 n ,         H ω = 0 n × r M - 1 B ω 2 n × r .

It is obvious that there exists a coupling of the uncertainties when the mass, damping and stiffness are subjected to possible perturbations, for example, while uncertainties exist in the matrices M and K, the uncertain description of the matrix block M-1K can be expressed as M+ΔM-1K+ΔK, which results in a serious conservatism in the system analysis and synthesis obviously, however this is not existing in the state Eq. (5). That is to say, Eq. (5) has less conservatism than Eq. (6).

Lemma 1 [34]: given matrices χ, μ and ν with appropriate dimensions and with χ symmetrical, then χ+μF(t)ν+νTF(t)TμT<0 holds for any F(t) satisfying F(t)TF(t)I, if and only if there exists a scalar δ>0 such that χ+δμμT+δ-1νTν<0.

In the following content, we consider how to obtain the suitable TMD parameters such that the system with the designed TMDs has a satisfying disturbance attenuation performance. Then, the robustness of the TMD control system will be considered.

3. TMD parameters design for structural systems

Theorem 1: The system Eq. (4) is admissible with performance z2<γω2 for all non-zero ωL20,  , and constant γ>0, if there exist a positive definite symmetric matrix P and matrix S satisfying the following matrix inequality:

(7)
Ψ = A T E T P + S L T T + E T P + S L T A + C z T C z E T P + S L T H ω * - γ 2 I < 0 ,

where LR3n+k×n+k is any full-column rank matrix satisfying ETL=0.

Proof:First, system Eq. (4) is regular and impulse free under the condition of Theorem 1 is proved. Form Eq. (7), we have:

(8)
A T E T P + S L T T + E T P + S L T A < 0 .

According to [35], it can be obtained that the system (4) is regular and impulse free. Then, system Eq. (4) is stable under the condition of Theorem 1 will be shown. Choose a Lyapunov functional candidate as:

(9)
V t = x T t E T P E x t .

The derivative of V(t) along the trajectories of Eq. (4) satisfies:

(10)
V ˙ ( t ) = 2 x T ( t ) E T P A x ( t ) + H ω ω ( t ) .

Noting ETL=0, we obtain:

(11)
2 x T ( t ) S L T E x ˙ ( t ) = 2 x T ( t ) S L T A x t + H ω ω t = 0 .

Next, the z2<γω2 performance of the system under zero initial condition (x(t) t=0=0, and V(t) t=0=0) will be established. Consider the following index:

(12)
J = 0 z T ( t ) z ( t ) - γ 2 ω ( t ) T ω ( t )   d t .

Then, for any non-zero ω(t)L20,, there holds:

(13)
J 0 z T ( t ) z ( t ) - γ 2 ω ( t ) T ω ( t ) d t + V ( t )   t = - V ( t )   t = 0
            = 0 z T ( t ) z ( t ) - γ 2 ω ( t ) T ω ( t ) + V ˙ ( t ) d t .

Noting Eqs. (10)-(13), after some algebraic manipulations, the following results can be obtained:

(14)
z T ( t ) z ( t ) - γ 2 ω ( t ) T ω ( t ) + V ˙ ( t ) x t T         ω t T Ψ x t ω t .

Then, 0zT(t)z(t)-γ2ω(t)Tω(t)+V˙(t)dt<0 is obtained from Eq. (7). Thus J<0, and z2<γω2 is satisfied for any non-zero ωL20,. Assume the zero disturbance input, i.e. ω(t)0. If Ψ<0, it is easy to obtain V˙(t)<0, and the asymptotic stability of system Eq. (4) is established. This completes the proof.

Remark 2: The optimal TMD parameters can be obtained by solving the following optimization problems, such as, according to solving the optimization 1, the TMD parameters, which guarantee the minimum γ, can be obtained. For a given γ=γ- (γ- is a constant), a set of TMD parameters, which has the minimum TMD masses, is obtained by solving the optimization 2.

Optimization 1:

(15)
m i n P > 0,0 < m d i m - d i , 0 < k d i k - d i , 0 < c d i c - d i γ ,         ( s . t .   >   E q .   7 ) .

Optimization 2:

(16)
m i n P > 0 , γ = γ - , 0 < m d i m - d i , 0 < k d i k - d i , 0 < c d i c - d i i = 1 k m d i ,       ( s . t . > E q .   7 ) .

Theorem 2: The system Eq. (5) is robustly admissible with performance z2<γω2 for all non-zero ωL20,  , and constant γ>0, if there exist positive definite symmetric matrix P, matrix S and positive scalars υi1,υi2,,υin i=1, 2, 3, υi1,υi2,,υik i=4, 5, 6 satisfying the following LMI:

(17)
Ψ ~ = Ψ ~ 11 Ψ ~ 12 * Ψ ~ 22 < 0 ,

where LR3n+k×(n+k) is any full-column rank matrix satisfying ETL=0, and:

Ψ ~ 11 = Ψ + i = 1 n υ 1 i θ - 1 i 2 m i 2 Γ 1 i Γ 1 i T + i = 1 n υ 2 i θ - 2 i 2 k i 2 Γ 2 i Γ 2 i T + i = 1 n υ 3 i θ - 3 i 2 c i 2 Γ 3 i Γ 3 i T
            + i = 1 k υ 4 i θ - 4 i 2 m d i 2 Γ 4 i Γ 4 i T + i = 1 k υ 5 i θ - 5 i 2 k d i 2 Γ 5 i Γ 5 i T + i = 1 k υ 6 i θ - 6 i 2 c d i 2 Γ 6 i Γ 6 i T ,
Γ 1 i = 0 2 n + k + i - 1 1 0 n + k - i 1 ,         Γ 21 = 1 0 3 n + 3 k ,         Γ 2 i = 0 i - 1 1 - 1 0 3 n + 3 k - i ,         Γ 31 = 0 n + k 1 0 2 n + 2 k ,
Γ 3 i = 0 n + k + i - 1 1 - 1 0 2 n + 2 k - i ,         Γ 4 i = 0 3 n + 2 k + i - 1 1 0 k - i 1 ,         Γ 5 i = 0 j i - 1 1 0 n - j i + i - 1 - 1 0 2 n + 3 k - i + 1 ,         Γ 6 i = 0 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 n + 2 k - i + 1 ,
Ψ ~ 12 = Λ 11 , Λ 12 , , Λ 1 n , Λ 21 , Λ 22 , , Λ 2 n , Λ 31 , Λ 32 , , Λ 3 n
            Λ 41 , Λ 42 , , Λ 4 k , Λ 51 , Λ 52 , , Λ 5 k , Λ 61 , Λ 62 , , Λ 6 k ,
Λ 1 i = E T P + S L T 0 2 n + k + i - 1 1 0 n + k - i + 1 0 ,         Λ 21 = E T P + S L T 0 2 n + k 1 0 n + k - 1 0 ,
Λ 2 i = E T P + S L T 0 2 n + k + i - 1 1 - 1 0 n + k - i - 1 0 ,         Λ 31 = E T P + S L T 0 2 n + k 1 0 n + k - 1 0 ,
Λ 3 i = E T P + S L T 0 2 n + k + i - 1 1 - 1 0 n + k - i - 1 0 ,         Λ 4 i = E T P + S L T 0 3 n + 2 k + i - 1 1 0 k - i 0 ,
Λ 5 i = E T P + S L T 0 2 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 k - i 0 ,         Λ 6 i = E T P + S L T 0 2 n + k + j i - 1 1 0 n - j i + i - 1 - 1 0 k - i 0 ,
Ψ ~ 22 = d i a g - υ 11 , - υ 12 , - υ 1 n , - υ 21 , - υ 22 , - υ 2 n , - υ 31 , - υ 32 , - υ 3 n ,
            - υ 41 , - υ 42 , - υ 4 k , - υ 51 , - υ 52 , - υ 5 k , - υ 61 , - υ 62 , - υ 6 k .

Proof: Replacing A and Hω with AΔ and HωΔ, Eq. (7) can be expressed as:

(18)
Ψ + i = 1 n θ 1 i m i Λ 1 i Γ 1 i T + θ 2 i k i Λ 2 i Γ 2 i T + θ 3 i c i Λ 3 i Γ 3 i T H
            + i = 1 k θ 4 i m d i Λ 4 i Γ 4 i T + θ 5 i k d i Λ 5 i Γ 5 i T + θ 6 i c d i Λ 6 i Γ 6 i T H < 0 .

By Lemma 1, Eq. (18) holds if and only if there exist positive scalars υi1,υi2,,υin (i= 1, 2, 3), υi1,υi2,,υik i=4, 5, 6, such that:

(19)
Ψ + i = 1 n υ 1 i - 1 Λ 1 i Λ 1 i T + υ 1 i θ - 1 i 2 m i 2 Γ 1 i Γ 1 i T + υ 2 i - 1 Λ 2 i Λ 2 i T + υ 2 i θ - 2 i 2 k i 2 Γ 2 i Γ 2 i T
            + υ 3 i - 1 Λ 3 i Λ 3 i T + υ 3 i θ - 3 i 2 c i 2 Γ 3 i Γ 3 i T + i = 1 k υ 4 i - 1 Λ 4 i Λ 4 i T + υ 4 i θ - 4 i 2 m d i 2 Γ 4 i Γ 4 i T
            + υ 5 i - 1 Λ 5 i Λ 5 i T + υ 5 i θ - 5 i 2 k d i 2 Γ 5 i Γ 5 i T + υ 6 i - 1 Λ 6 i Λ 6 i T + υ 6 i θ - 6 i 2 c d i 2 Γ 6 i Γ 6 i T < 0 .

Applying the Schur complement, equation (17) is equivalent to equation (19). This completes the proof.

Obviously, the optimal TMD parameters for the uncertain systems can be obtained by solving the following optimization problems.

Optimization 3:

(20)
m i n P > 0,0 < m d i m - d i , 0 < k d i k - d i , 0 < c d i c - d i γ ,         s . t . > E q . 17 .

Optimization 4:

(21)
m i n P > 0 , γ = γ - , 0 < m d i m - d i , 0 < k d i k - d i , 0 < c d i c - d i i = 1 k m d i ,         s . t .   >   E q .   17 .

Remark 3: It is worthy to point out that there exist nonlinear properties in Eq. (7) and (17) for the coupling of the parameters mdi, cdi, kdi (i= 1, 2,…, k) with the matrices P and S. Thus, the traditional LMI solving methods is useless to solve Eq. (7) and (17). Fortunately, there are lots of intelligent algorithms (such as, genetic algorithm, neural network, particle swarm optimization and ant colony algorithm, etc.), which have been certified to be effective for treating the nonlinear properties of all kinds of equations. In this paper, a method of combining genetic algorithm with LMI toolbox is introduced to solve the optimizations 1 to 4, and the algorithm is shown as following:

Algorithm.

Step 0 (initialization): Set the maximum iteration step T, population size N, crossover rate Pc, mutation rate Pm, and the steps counter t=0. Then, the individual solutions are randomly generated to form an initial population, for example, in optimization 1, the population size is l, that is, the initial population are mdij, cdij, kdij, i= 1, 2,…, k, j= 1, 2,…, l.

Step 1 (fitness of the individual solutions): The fitness is the value of the objective function in the optimization problem being solved. Based on solving of the optimization 1, 2, 3 or 4, the fitness of the individual solutions can be obtained.

Step 2 (selection, reproduction and mutation): The more fit individuals are stochastically selected from the current population, and each individual’s genome is modified (recombined and possibly randomly mutated) to form a new generation. The new generation of candidate solutions is then used in the next iteration of the algorithm.

Step 3 (production of the new generation): Based on solving of the optimization 1, 2, 3 or 4, the fitness of the individuals in the new generation can be obtained, and according to fitness of the individuals in the new generation, the candidate individual solutions can be obtained and then used in the next iteration of the algorithm.

Step 4 (Termination): The algorithm terminates when t=T, or a maximum number of generations has been produced, or a satisfactory fitness level has been reached for the population, and the individual solution with the maximum fitness is chosen as the optimal solution. Otherwise, set t=t+1, and go back to step 2.

4. Illustrative example

Consider a 3 degree-of-freedom structural system, which has a TMD installed on the third storey. The structural parameters are as following: mi= 345.6 ton, ci= 2973 kNs/m-1, ki= 3.404×105 kN/m, i= 1, 2, 3, are the mass, damping and stiffness of each storey respectively [19]. md, cd and kd, which are the parameters of the TMD, need to be designed. Choose zt=x-~(t)T,x-~˙(t)TTas the controlled output, that is, Cz=I808×4. Firstly, the system without uncertainties is considered. Set γ= 0.5, and choose the minimum md as the optimization objective. Then, by setting the maximum iteration step T= 100, we use the Matlab GAOPT Toolbox to solve the Algorithm mentioned above (the values of md obtained in the iteration are shown in Fig. 2), and obtain the optimal TMD parameters, which are shown in Table 1. For description in brevity, this TMD is denoted as TMD1 thereafter.

Table 1. The final results after 100 steps

k d
c d
m d
γ
21244.88
2632.25
218.28
0.5

Fig. 2. The md obtained in the iteration

The md obtained in the iteration

In order to verify the dynamics of the TMD-controlled system, a time history of acceleration from North Korea seismic excitation is applied to this system (see Fig. 3, http://www. vibrationdata.com/newsletters.html), and the storey 3’ displacements of the system with and without TMD1 are shown in Fig. 4. The displacements of the other two storeys and the velocities of three storeys have a similar varying trend, which is omitted here for brevity. It is shown from Fig. 4 that TMD 1 is effective in attenuating the vibration of the structural systems. The maximum displacements and velocities of the three storeys are shown in Table 2. The Percentage of the Maximum-displacement-reduced Value (PMV) of the system with TMD1 is ((0.0599-0.0185))/0.0599 ≈ 33.06 % in storey 1, and the other PMVs are shown in Table 2. That is, the maximum displacements and velocities are all attenuated when the system is controlled by TMD1. Thus, the effectiveness of TMD1 is obvious.

Fig. 3. North Korea seismic excitation

North Korea seismic excitation

Fig. 4. The storey 3’ displacements of the structural system

The storey 3’ displacements of the structural system

Table 2. the maximum displacements and accelerations of the structural system

Displacements (m)
Velocities (m/s)
Floor 1
Floor 2
Floor 3
Floor 1
Floor 2
Floor 3
Without TMD1
0.0599
0.1078
0.1348
0.9108
1.6771
2.1212
with TMD1
0.0401
0.0775
0.1019
0.5594
1.0249
1.3168
PMV
33.06 %
28.11 %
24.41 %
38.58 %
38.89 %
37.92 %

Then, the uncertain case is considered. Consider the uncertainties as applied to the mass, stiffness and damping coefficients of the first storey, and assume the parameter uncertainties satisfying m1Δ=m1+θ11m1, k1Δ=k1+θ21k1, c1Δ=c1+θ31c1, 0θ11θ-11=0.2, 0θ21θ-21=0.2, 0θ31θ-31=0.2. Choose the minimum γ as the optimization objective. By setting the maximum iteration step T= 50, we use the Matlab GAOPT Toolbox to solve the Algorithm mentioned above (the values of γ obtained in the iteration are shown in Fig. 5), and obtain the TMD parameters which are shown in Table 3. For description in brevity, this TMD is denoted as TMD2 thereafter.

Table 3. The final result after 50 steps in the uncertain case

k d
c d
m d
γ
7135.642
7797.655
1620.312
0.302577

Under the excitation mentioned above, the displacements of storey 3 for the system with TMD2 are shown in Fig. 6. The displacements of the other two storeys and the velocities of the three storeys have a similar varying trend, which is omitted here for brevity. It is shown from Fig. 6 that TMD 2 is effective in attenuating the vibration of the uncertain structural systems obviously. The maximum displacements and velocities of the three storeys are shown in Table 4. From Table 4, it is easy to get that the maximum displacements and velocities are all attenuated when the system is controlled by TMD2. Thus, it is validated that TMD2 is robust to parameter uncertainties.

Table 4. The maximum displacements and accelerations of the uncertain structural system

Displacements (m)
Velocities (m/s)
Floor 1
Floor 2
Floor 3
Floor 1
Floor 2
Floor 3
Without TMD2
0.0643
0.1061
0.1290
1.0542
1.5584
1.8664
With TMD2
0.0226
0.0413
0.0570
0.3583
0.6283
0.8277
PMV
64.85 %
61.07 %
55.81 %
66.01 %
59.68 %
55.65 %

Fig. 5. The γ obtained in the iteration for the uncertain case

The γ obtained in the iteration for the uncertain case

Fig. 6. The storey 3’ displacements of the uncertain structural system

The storey 3’ displacements of the uncertain structural system

Table 5. The maximum displacements and accelerations of the uncertain structural system

Displacements (m)
Velocities (m/s)
Floor 1
Floor 2
Floor 3
Floor 1
Floor 2
Floor 3
Without TMD2
0.0098
0.0189
0.0240
0.1488
0.2748
0.3579
With TMD2
0.0084
0.0138
0.0195
0.0713
0.1321
0.1786
PMV
14.29 %
26.98 %
18.75 %
52.08 %
51.93 %
50.10 %

In order to further verify the effctiveness of the obtained TMD controller, another seismic excitation (see Fig. 7), which was adopted as an excitation in [2, 19-21], is applied to this system. The storey 3’ displacements of the system with and without TMD2 are shown in Fig. 8. The displacements of the other two storeys and the velocities of three storeys have a similar varying trend, which is omitted here for brevity. It is shown from Fig. 8 that TMD 2 is still effective in attenuating the vibration of the uncertain structural systems which is excited by the seismic excitation shown in Fig. 7. The maximum displacements, velocities and PMVs of the three storeys are shown in Table 5. From Table 5, it is easy to get that the maximum displacements and velocities are all attenuated when the system is controlled by TMD2. Thus, the effectiveness of TMD2 is validated again.

Fig. 7. EI Centro 1940 earthquake excitation

EI Centro 1940 earthquake excitation

Fig. 8. The storey 3’ displacements of the uncertain structural system

The storey 3’ displacements of the uncertain structural system

5. Conclusions

The TMD parameters designing for structural systems based on combining LMI technique with genetic algorithm is of concern in this paper. Firstly, based on matrix transform, the singular description of the structural systems controlled by TMDs is obtained. Secondly, in terms of combining LMI toolbox with genetic algorithm to solving the optimization problem, the optimized TMD parameters are obtained such that the TMD-controlled system has a prescribed level of vibration attenuation performance. Thirdly, the obtained results are also extended to the uncertain cases. Finally, examples are given to show the effectiveness of the obtained theorems.

Acknowledgements

This research has been supported by the National Natural Science Foundation (No. 61463018), Jiangxi Provincial Natural Science Foundation (No. 20151BAB207046) and the Natural Science Foundation of Jiangxi University of Science and Technology (No. NSFJ2014-K16) of China.

References

  1. Yao J. Concept of structural control. Journal of Structural Division, Vol. 98, Issue 7, 1972, p. 1567-1574. [Search CrossRef]
  2. Weng F., Ding Y., Liang L., Yang G., Ge J. Finite-time vibration-attenuation controller design for structural systems with actuator faults and parameter uncertainties. Journal of Vibration Engineering and Technologies, Vol. 4, Issue 2, 2016, p. 117-129. [Search CrossRef]
  3. Soong T., Cimellaro G. Future directions in structural control. Structural Control and Health Monitoring, Vol. 16, Issue 1, 2009, p. 7-16. [Publisher]
  4. Du H., Zhang N., Samali B., Naghdy F. Robust sampled-data control of structures subject to parameter uncertainties and actuator saturation. Engineering Structures, Vol. 36, 2012, p. 39-48. [Publisher]
  5. Ding Y., Weng F., Tang M., Ge J. Active vibration-attenuation controller design for uncertain structural systems with input time-delay. Earthquake Engineering and Engineering Vibration, Vol. 14, Issue 3, 2015, p. 477-486. [Publisher]
  6. Fujino Y. Vibration, control and monitoring of long-span bridges-recent research, developments and practice in Japan. Journal of Constructional Steel Research, Vol. 58, Issue 1, 2002, p. 71-97. [Publisher]
  7. Koshimura K., Tatsumi M., Hata K. Vibration control of the main towers of the Akasi Kaikyo Bridge. Proceedings of the First World Conference on Structural Control, Los Angels, California, Vol. 2, 1994, p. 98-106. [Search CrossRef]
  8. Ou J., Wu B. Recent advances in research on and applications of passive energy dissipation Systems. Earthquake Engineering and Engineering Vibration, Vol. 16, Issue 3, 1996, p. 72-96. [Search CrossRef]
  9. Chen Y., Cao T. Tune mass damper system for roof wind control of Meydan Racecourse Stadium in Dubai. Building Structure, Vol. 42, Issue 3, 2012, p. 49-53. [Publisher]
  10. Pourzeynali S., Salimi S., Kalesar H. Robust multi-objective optimization design of TMD control device to reduce tall building responses against earthquake excitations using genetic algorithms. Scientia Iranica, Vol. 20, Issue 2, 2013, p. 207-221. [Search CrossRef]
  11. Morga M., Marano G. Optimization criteria of TMD to reduce vibrations generated by the wind in a slender structure. Journal of Vibration and Control, Vol. 20, Issue 16, 2014, p. 2404-2416. [Publisher]
  12. Yang Y., Dai W., Liu Q. Design and implementation of two-degree-of-freedom tuned mass damper in milling vibration mitigation. Journal of Sound and Vibration, Vol. 335, Issue 20, 2015, p. 78-88. [Publisher]
  13. Debnath N., Deb S., Dutta A. Multi-modal vibration control of truss bridges with tuned mass dampers under general loading. Journal of Vibration and Control, 2015, p. 1-20. [Publisher]
  14. Xiang P., Nishitani A. Seismic vibration control of building structures with multiple tuned mass damper floors integrated. Earthquake Engineering and Structural Dynamics, Vol. 43, 2014, p. 909-925. [Publisher]
  15. Soto M., Adeli H. Optimum tuning parameters of tuned mass dampers for vibration control of irregular highrise building structures. Journal of Civil Engineering and Management, Vol. 20, Issue 5, 2014, p. 609-620. [Publisher]
  16. Zhou D., Li J., Hansen C. Suppression of the stationary maglev vehicle-bridge coupled resonance using a tuned mass damper. Journal of Vibration and Control, Vol. 19, Issue 2, 2013, p. 191-203. [Publisher]
  17. Lin G., Lin C., Chen B., Soong T. Vibration control performance of tuned mass dampers with resettable variable stiffness. Engineering Structures, Vol. 83, 2015, p. 187-197. [Publisher]
  18. Wang A., Lin Y. Vibration control of a tall building subjected to earthquake excitation. Journal of Sound and Vibration, Vol. 299, 2007, p. 757-773. [Publisher]
  19. Du H., Zhang N. H∞ control for buildings with time delay in control via linear matrix inequalities and genetic algorithms. Engineering Structures, Vol. 30, Issue 1, 2008, p. 81-92. [Publisher]
  20. Zhang W., Chen Y., Gao H. Energy-to-peak control for seismic-excited buildings with actuator faults and parameter uncertainties. Journal of Sound and Vibration, Vol. 330, Issue 4, 2011, p. 581-602. [Publisher]
  21. Samali B., Al-Dawod M. Performance of a five-story benchmark model using an active tuned mass damper and a fuzzy controller. Engineering Structures, Vol. 25, 2003, p. 1597-1610. [Publisher]
  22. Li L., Song G., Ou J. Nonlinear structural vibration suppression using dynamic neural network observer and adaptive fuzzy sliding mode control. Journal of Vibration and Control, Vol. 16, Issue 10, 2010, p. 1503-1526. [Publisher]
  23. Rathi A., Chakraborty A. Robust design of TMD for vibration control of uncertain systems using adaptive response surface method. Advances in Structural Engineering, 2015, p. 1505-1517. [Publisher]
  24. Aldermir U. Optimal control of structures with semiactive tuned mass dampers. Journal of Sound and Vibration, Vol. 266, Issue 4, 2003, p. 847-874. [Publisher]
  25. Alexandros A. Optimal probabilistic design of seismic dampers for the protection of isolated bridges against near-fault seismic excitations. Engineering Structures, Vol. 33, Issue 12, 2011, p. 3496-3508. [Search CrossRef]
  26. Yang F., Sedaghati R., Esmailzadeh E. Optimal design of distributed tuned mass dampers for passive vibration control of structures. Structural Control and Health Monitoring, Vol. 22, Issue 2, 2015, p. 221-236. [Publisher]
  27. Ding Y., Weng F., Ge J., Nguyen J. Displacement-constrained vibration-attenuation controller design for linear structure systems with parameter uncertainties. International Journal of Computer Applications in Technology, Vol. 53, Issue 1, 2016, p. 82-90. [Publisher]
  28. Ding Y., Weng F., Jiang X., Tang M. Vibration-attenuation controller design for uncertain mechanical systems with input time delay. Shock and Vibration, 2016, p. 9686358. [Search CrossRef]
  29. Wu Z., Park J., Su H., Chu J. Admissibility and dissipativity analysis for discrete-time singular systems with mixed time-varying delays. Applied Mathematics and Computation, Vol. 218, Issue 13, 2012, p. 7128-7138. [Publisher]
  30. Ahmadi A., Aldeen M. An LMI approach to the design of robust delay-dependent overlapping load frequency control of uncertain power systems. International Journal of Electrical Power and Energy Systems, Vol. 81, 2016, p. 48-63. [Publisher]
  31. Long S., Zhong S. H∞ control for a class of discrete-time singular systems via dynamic feedback controller. Applied Mathematics Letters, Vol. 58, 2016, p. 110-118. [Publisher]
  32. Wu Z., Shi P., Su H., Lu R. Dissipativity-based sampled-data fuzzy control design and its application to truck-trailer system. IEEE Transactions on Fuzzy Systems, Vol. 23, Issue 5, 2015, p. 1669-1679. [Publisher]
  33. Karimi H., Zapateiro M., Luo N. An LMI approach to vibration control of base-isolated building structures with delayed measurements. International Journal of Systems Science, Vol. 41, Issue 12, 2010, p. 1511-1523. [Publisher]
  34. Weng F., Ding Y., Liang L. Acceleration-based vibration control for structural systems with actuator faults and finite-time state constraint. Noise and Vibration Worldwide, Vol. 46, Issue 4, 2015, p. 10-21. [Publisher]
  35. Dai L. Singular control systems. Lecture Notes in Control and Information Sciences, Springer-Verlag, New York, 1989. [Search CrossRef]

Cited By

Falu Weng, Huan Wang, Yuanchun Ding
2020
Proceedings of the Institution of Mechanical Engineers, Part C: Journal of Mechanical Engineering Science
Yuanchun Ding, Falu Weng, Feiyue Geng
2019
Asian Journal of Control
Falu Weng, Yi Guo, Yuanchun Ding, Shuren Han, Feiyue Geng
2018
ISA Transactions
Falu Weng, Mingxin Liu, Weijie Mao, Yuanchun Ding, Feifei Liu
2018