This action might not be possible to undo. Are you sure you want to continue?

Welcome to Scribd! Start your free trial and access books, documents and more.Find out more

**JOURNAL OF SOUND AND VIBRATION
**

Journal of Sound and Vibration 311 (2008) 1391–1408 www.elsevier.com/locate/jsvi

**Active vibration suppression by pole-zero placement using measured receptances
**

John E. Mottersheada,Ã, Maryam Ghandchi Tehrania, Simon Jamesa, Yitshak M. Ramb

a

Department of Engineering, University of Liverpool, Brownlow Hill, Liverpool L69 3GH, UK b Department of Mechanical Engineering, Louisiana State University, Baton Rouge, USA

Received 28 November 2006; received in revised form 15 October 2007; accepted 16 October 2007 Available online 26 November 2007

Abstract The paper addresses the problem of pole-zero assignment using the receptance method in active vibration control and has applications particularly in vibration absorption and detuning of structures to avoid resonance. An output feedback approach is described that makes use of measured receptances, there being no requirement at all for the M, C, K matrices usually obtained by ﬁnite elements. Therefore, in the controller design, the approximations, assumptions and other modelling errors are largely eliminated. In addition, the method does not require the use of model reduction techniques or the estimation of unmeasured states by an observer. An advantage of the output feedback method, over state feedback, is that collocated actuator–sensor arrangements become possible. However this is achieved at the expense of creating characteristic equations nonlinear in the control gains. Numerical examples are provided to illustrate the working of the method. This is followed by a series of experimental tests carried out using collocated accelerometers and inertial actuators on a T-shaped plate. In a series of experiments poles and zeros are assigned separately and simultaneously. Stability robustness is demonstrated by applying a constraint to the singular values of the matrix return difference. r 2007 Elsevier Ltd. All rights reserved.

1. Introduction The problem of eigenvalue assignment was studied quite extensively within the active control community during the 1970s and 1980s. Wonham [1] had shown in 1967 that if a system was controllable, then its eigenvalues (or poles) could be assigned by appropriate choice of state feedback. Davison [2] determined the conditions under which output feedback could be applied for eigenvalue assignment. Kimura [3] studied the problem of incomplete state observation and Moore [4] considered the freedom offered by state feedback beyond the speciﬁcation of distinct closed-loop eigenvalues. Kautsky et al. [5] described numerical methods for determining robust (well conditioned) solutions to the state-feedback pole-assignment problem by deﬁning a solution space of linearly independent eigenvectors, corresponding to the required eigenvalues. The solutions obtained were such that the sensitivity of assigned poles to perturbations in the system and gain

ÃCorresponding author.

E-mail address: j.e.mottershead@liverpool.ac.uk (J.E. Mottershead). 0022-460X/$ - see front matter r 2007 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsv.2007.10.024

and (3) the number of eigenvalues to be assigned must be matched by the rank of the modiﬁcation. They showed.ARTICLE IN PRESS 1392 J. [6] developed the closed-form solution for the partial pole-assignment problem where some desired eigenvalues were relocated while keeping all the other eigenvalues unchanged. positive-deﬁniteness. derived from measured receptances. in a three-part paper [16–18] covering the theory. Datta et al. Meirovitch’s analysis is presented in terms of the physical mass. A good example is the ‘smart panel’ described by Gardonio et al. In practice. In recent times many of the disadvantages that prevented the practical application of active control to ﬂexible structures. K matrices are obtained by ﬁnite elements. [24]. and moving natural frequencies away from resonance with applied harmonic loads by applying a passive structural modiﬁcation. / Journal of Sound and Vibration 311 (2008) 1391–1408 matrices was minimised. Mottershead et al. Another way of implementing the independent modal-space control of Meirovitch is to use modal test data. There are in fact numerous examples of structures having natural frequencies close to ﬁxed excitation frequencies. Each control unit consisted of a collocated accelerometer-sensor and a piezoceramic patch actuator with a single channel velocity feedback controller to generate active damping. M. Examples include. matrices ﬁnally arranged in the form of ﬁrst-order state-space equations (when general viscous damping is included). the control force must be applied in the physical coordinates. Mottershead and Ram [13] reviewed the ﬁeld of inverse eigenvalue problems for vibration absorption by structural modiﬁcation and active control. More recently. design and application of the system. From the point of view of the structural dynamics it is preferable to work with the second-order matrix pencil [23]. In most cases the M. and may include the representation of distributed piezo-electric actuators and sensors. (2) rotational receptances are very difﬁcult to measure and require high levels of specialist expertise. and consequently the ﬁrst-order state-space model does not preserve any notion of the second-order nature of the system. assigning a zero at the tuned frequency of a classical vibration absorber. The state-of-art in structural vibration control however is the ‘independent modal-space control’ method developed by Meirovitch and his students and described in the book ‘Dynamics and Control of Structures’ [21]. this method preserved the second-order nature of the physical system since the receptances are given by. C. [10.E. as described for example by Lim et al. C. as described by Stobener and Gaul [25] in the active vibration control of a car body. However there are very considerable disadvantages: (1) the form of the modiﬁcation that can be realised physically (symmetry. K. Inertial actuators for the application of ‘sky-hook’ damping [14] and piezo-electric devices [15] for thin plate-like applications have been developed. Recently much attention has been focussed on inverse eigenvalue problems for assigning natural frequencies and mode shapes [7]. Active damping by velocity feedback has received much attention recently. In vibration analysis the purpose of assigning poles and zeros is to suppress vibration. Modern distributed actuators may be designed so that the vibration modes of a plate are controlled selectively in this way [22]. The advantage of the structural modiﬁcation approach is that the system is guaranteed to remain stable. deﬁniteness and bandedness. moving poles further to the left-hand side of the complex plane to increase damping. active constrained layer damping [19]. Mottershead and Tehrani [12] carried out the structural modiﬁcation of a helicopter tailcone also using measured rotational receptances obtained by means of an X-block attachment. how state-feedback control using measured receptances from the original (open-loop) system could be used to assign all the poles of the closed-loop system using just a single actuator. H(io) ¼ (Ào2M+ioC+K)À1 and the poles and zeros (anti-resonances) were assigned without the need to .11] measured rotational receptances and applied them to assign the natural frequencies and anti-resonances of a G-shaped structure by means of an added beam. Redeﬁning the second-order equations of motion into a ﬁrst-order realisation destroys the desirable physical matrix properties of symmetry. An alternative approach that uses the measured receptances directly and makes no assumption about damping was proposed by Ram and Mottershead [26]. pattern of non-zero matrix terms) is restrictive. in principle. This approach allows in principle for the control of one structural mode independently of the others. which comprised 16 decentralised units for the control of sound transmission. but these techniques are not applicable to large-scale built-up structures consisting of many components. have been overcome. A different approach. Additionally. Rayleigh (proportional) damping was assumed. with low damping and large numbers of modes. which means that sufﬁcient actuators must be used to ensure that the selected mode is controllable while the others are unaffected by the control force. damping and stiffness. vibration nodes [8] and anti-resonances [9] by structural modiﬁcation. For example Mottershead et al. can be particularly helpful in damping the higher-frequency closed-loop modes that might otherwise become unstable due to spillover [20].

.ARTICLE IN PRESS J. r . ðs2 M þ sC þ KÞxðsÞ ¼ BuðsÞ þ pðsÞ. C. the other poles of the system at higher frequencies are found to remain stable. Mottershead et al. KAR and mass matrices. Likewise. A necessary and sufﬁcient condition is that the matrices M. two zeros are assigned. In addition to velocity feedback. the method uses displacement feedback for active stiffness. for active damping. each to the two-pointreceptances corresponding to the locations of collocated proof-mass actuators and accelerometers.FARm Â l. In all the experiments. C. In the ﬁrst experiment. This means that by the receptance method there is no need to estimate the unmeasured states using an observer. Then. (C+BFD) and (K+BGD) should be real. Active stiffness is more difﬁcult to achieve than active damping [28] as can be appreciated when considering the location of a modal circle on the Nyquist diagram. . nÂn T T T (1) . described in the following section. whereas for displacement feedback the circle occupies the lower two quadrants and can therefore be very close to the threshold of instability at À1. On the other hand. K matrices.E. M40. In the case of velocity feedback the modal circle is conﬁned to the right-hand half of the diagram. four poles are assigned. combining Eqs. . The very considerable advantage of the output feedback method over state feedback is that it allows the use of collocated actuators and sensors in multiple-input–multiple-output (MIMO) systems. In the third experiment poles and zeros are assigned simultaneously. . thereby enabling the assignment of both poles and zeros to desired locations in the complex s-plane. C. j ¼ 1. . in order to make the system equations complete. It is well known that the transfer function of a lightly damped system with a collocated actuator–sensor pair displays a line of interlacing poles and zeros just to the left of the imaginary axis. (1)–(3) leads to. K ¼ K . . In this paper the receptance method for output-feedback is described in detail. BARn Â m is the control force distribution matrix. The method. p(s)ARn Â 1 represent the displacement states and external forces. The beneﬁts of this arrangement were discussed in detail by Preumont [27] who showed the trajectory of one such pole to be into the left-hand half-plane. and u(s)ARm Â 1 is the control force. r p 2n j j (4) (5) which we assume to be distinct and closed under conjugation. when using the receptance method only the available states are needed to complete the equations. The assignment of zeros is of special interest in vibration analysis because the vibration can be made to vanish at chosen frequencies and locations. may be cast in the form of the second-order matrix equation. x(s). damping where M. respectively. Our purpose is to assign poles lj of the closed-loop system determined by the solution of. or estimated using an observer. / Journal of Sound and Vibration 311 (2008) 1391–1408 1393 know or evaluate the M. where DAR expressed as lÂn (2) lÂ1 is the sensor distribution matrix and the output is denoted by yAR uðsÞ ¼ ÀðG þ sFÞyðsÞ. ðs2 M þ sC þ K þ BðG þ sFÞDÞxðsÞ ¼ pðsÞ. KX0 are the usual structural stiffness. and in the second. in the Laplace frequency domain. A further signiﬁcant point is that the familiar state-space approach requires a dynamic stiffness model and all the states must be measured. det l2 M þ l ð C þ BFD Þ þ ð K þ BGD Þ ¼ 0. 2. The feedback law is (3) so that the output and rate gains are given by the terms in the matrices G. Poles and zeros The dynamical equations. Neither is there any need for model reduction. therefore remaining stable under closed-loop control. is applied to both numerical and experimental examples. Three experiments on a ‘T’-shaped plate are described. C ¼ C . the output equation may be written as yðsÞ ¼ DxðsÞ. 2. M ¼ M .

. ðI þ HðsÞDZðsÞÞxðsÞ ¼ HðsÞpðsÞ. r. 3. . are given by the solution of a different eigenvalue problem. 6 .pþ1 .1 6 6 . where DZðsÞ ¼ BðG þ sFÞD xðsÞ ¼ ðI þ HðsÞBðG þ sFÞDÞÀ1 HðsÞpðsÞ or xðsÞ ¼ adjðI þ HðsÞBðG þ sFÞDÞ HðsÞpðsÞ: detðI þ HðsÞBðG þ sFÞDÞ (14) (12) and ﬁnally the closed-loop receptance equation can be expressed in terms of the open-loop receptances as (13) (11) (10) The closed-loop system poles may be assigned by selecting real-valued gains G. j ¼ 1. (16) .pÀ1 . mn1 2 ÁÁÁ m1. 6 . mn. .E. mp.pþ1 mpþ1. Mottershead et al.n 7 7 7 . det m2 (6) k Mp þ mk ðC þ BFDÞp þ ðK þ BGDÞp ¼ 0. denoted hpp(s). The zeros of the closed-loop system occur when terms in the numerator matrix product. Premultiplying both sides of Eq. / Journal of Sound and Vibration 311 (2008) 1391–1408 The closed-loop zeros. 2. to satisfy the nonlinear characteristic equations. mT M¼ p mp Mp and m11 6 . .. mn. The zeros of the ppth receptance may therefore be assigned by selecting gains such that.8) ÁÁÁ ÁÁÁ ÁÁÁ ÁÁÁ ÁÁÁ ÁÁÁ 3 m 1n 7 . . mpÀ1.pÀ1 mpþ1. As with the poles. . where Mp is the submatrix corresponding to the ppth term of M.pþ1 ÁÁÁ i . À Á det I þ Hðlj ÞBðG þ lj FÞD ¼ 0. 4 . adj(I+H(s)B(G+sF)D)H(s). rp2n. Thus. (4) by H(s) then gives.pÀ1 mp. . " # h mpp mT p ¼ mp1 .pþ1 . Â À Á Ã adj I þ Hðmk ÞBðG þ mk FÞD Hðmk Þ pp ¼ 0. (15) where the assigned poles are distinct and closed under conjugation. vanish to zero. 6 . . . .pÀ1 m1. .n 7 7 7 mpþ1.pþ1 . Since the equations are nonlinear in the gains there may be one or more strictly real solutions G. . we assume the zeros to be distinct and closed under conjugation. F or there may be no solution. 6 6 mpÀ1. It is seen that the zeros are the eigenvalues of the system grounded at the pth coordinate. . mpÀ1.1 6 Mp ¼ 6 6 mpþ1. . . of the ppth point receptance. .pÀ1 . mpn (7. 7 5 mnn (9) with similar deﬁnitions for the submatrices (C+BFD)p and (K+BGD)p. Pole and zero assignment by using receptances The receptance matrix of the open-loop system is now deﬁned as HðsÞ ¼ ðs2 M þ sC þ KÞÀ1 . . . in which case the closed-loop system is uncontrollable.ARTICLE IN PRESS 1394 J. mk. . F. 7 7 7 mpÀ1.

0170. neglected in the above theory. . Mottershead et al. C ¼ 4 0 0 0 3 0 À0:1 0 :1 6 6 À K¼4 2 À1 2 À2 4 À2 3 À1 7 À2 5 3 f1 k1 m1 c1 x1 k2 m2 x2 c2 k4 k3 m3 f3 x3 Fig. FX0. If the sensors and actuators are collocated. must be considered. the fully populated matrices should allow more poles and zeros to be assigned than the number of actuators and sensors. Numerical examples The method is illustrated by a series of eigenvalue assignment exercises based upon the system shown in Fig. damping and stiffness matrices are given by 2 3 2 3 0 :1 0 0 2 0 0 6 7 6 7 0 :1 À 0 :1 5 . 1.7i and l3. The parameters of the system have the following values: k1 ¼ 3.8i are assigned using two actuators supplying feedback control forces at masses m1 and m3. k2 ¼ 2. 2. (5) and (6) combined with Eqs. M ¼ 4 0 1 0 5. The mass. c1 ¼ 0.4 ¼ À0. G. (18) (17) then it is seen from Eq. m1 ¼ 2. m3 ¼ 3.E. Example 1 (Assignment of poles). a greater number of solutions to the nonlinear characteristic equations may become available. G ¼ diagðgi Þ. m2 ¼ 1. as indeed they are in the experimental study that follows in Section 5. D ¼ BT 2 <mÂn and the gain matrices are symmetric and positive semi-deﬁnite.0671. 1. so that if a solution (or solutions) exist then the gains are found to be strictly real. In principle. k4 ¼ 1. m. (18). Two pairs of complex conjugate poles at l1. (17) and (18)) retain considerably greater freedom in the choice of control gains than does a passive structural modiﬁcation by the adjustment of physical parameters such as beam cross sections or added masses. c2 ¼ 0. Three degree-of-freedom system.1. In the following numerical and experimental examples the gain matrices are diagonal. (19) This is perhaps the simplest form for the gain matrices. k3 ¼ 2. but it should be noted that in more complex or larger structural problems there may be advantages in retaining the fully populated symmetric form of Eqs. 4. (5) that the closed-loop system matrices are symmetric with unchanged deﬁniteness properties. À Á F ¼ diag f i . The resulting characteristic equations (Eqs. thereby allowing the selection of gains that result in the least cost of control or that render those assigned poles and zeros least sensitive to small changes in the gain terms. In practice the actuator and sensor dynamics. .ARTICLE IN PRESS J. F ¼ FT 2 <mÂm . .2 ¼ À0.1. i ¼ 1. / Journal of Sound and Vibration 311 (2008) 1391–1408 1395 We consider zeros that are distinct and closed under conjugation. The use of collocated actuator–sensor systems was discussed in detail by Preumont [27]. . Alternatively. G ¼ GT 2 <mÂm . which means that the poles have strictly negative real parts. .

. 0:048Þ. The open-loop system receptances are at the values of s corresponding to the chosen eigenvalues. Example 2 (Assignment of zeros). / Journal of Sound and Vibration 311 (2008) 1391–1408 and the matrices B and D are chosen to be. ÀMÀ1 ðK þ B diagðgi ÞBT Þ ÀMÀ1 ðC þ B diagðf i ÞBT Þ where 2 3 0 7 À 0 :1 5 0:1475 3 À1 7 À2 5 3:3401 0:1999 6 T ðC þ B diagðf i ÞB Þ ¼ 4 0 0 and 2 0 0 :1 À0:1 8:3873 6 T ðK þ B diagðgi ÞB Þ ¼ 4 À2 À1 which yields the following poles: À2 4 À2 l1. 4 which yields F ¼ diagð0:0493. j ¼ 1. are then solved numerically using a Gauss–Newton method. G ¼ diagð2:3873. the eigenvalues of the closed-loop system are obtained by the state-space method.2 ¼ À 0:0100 Æ 0:7000i. . The initial and modiﬁed receptances at m1 are plotted in Fig. . 1:5223Þ: . 0:3401Þ: In order to validate the results. l3. À À Á Á det I À Hðlj ÞB diag f i þ lj gi BT ¼ 0. j ¼ 1. generally nonlinear in the gains. 4. G ¼ diagð1:8691.02571. The sets of nonlinear equations (18) are solved to obtain the gains fi and gi Á Â À Ã adj I þ Hðmj ÞB diagðgi þ mj f i ÞBT Hðmj Þ 22 ¼ 0. l5. F ¼ diagð0:0999. j j Four characteristic equations (17).03772i are assigned using the feedback control forces at masses m1 and m3.4 ¼ À0. " # 0 I A¼ . . j ¼ 1.ARTICLE IN PRESS 1396 J. Mottershead et al.2i and m3. gi. D¼B ¼ T 1 0 0 0 0 1 ! . . so that y1 ¼ x1. The zero assignment is considered for the point receptance h22.6 ¼ À 0:0546 Æ 2:3599i: The ﬁrst two pairs exactly replicate the values assigned and the third pair of poles are stable. . . . Two pairs of complex conjugate zeros m1. À1 H ð lj Þ ¼ M l2 þ C l þ K . 4 with the result that. 2. . y2 ¼ x3. . represented by the solid (blue) and dashed (red) line. fi. . respectively. 0:0475Þ. .E.2 ¼ À0.4 ¼ À 0:0600 Æ 1:8000i.

5 1 1. Example 1—initial receptance (solid line) and modiﬁed receptance (dashed line). Mottershead et al. Example 2—initial receptance (solid line) and modiﬁed receptance (dashed line).5 4 4. Thus. 102 Amplitude 100 10−2 0 0 −50 Phase −100 −150 −200 0 0.E.5 4 4. " A¼ À Á 1 T À MÀ 2 ðK þ B diagðgi ÞB 2 0 À Á .5 5 Frequency (rad/s) Fig. 0. T 1 À MÀ 2 ðC þ B diagðf i ÞB 2 I # 102 Amplitude 100 10−2 10−4 0 0 −50 1 2 3 4 5 Phase −100 −150 −200 0 1 2 3 4 5 Frequency (rad/s) Fig. may be obtained by the state-space method using the gains determined above.5 2 2. 2. 3.5 3 3.5 5 .5 3 3.ARTICLE IN PRESS J.5 1 1.5 2 2. being the eigenvalues of the system grounded at the second coordinate. / Journal of Sound and Vibration 311 (2008) 1391–1408 1397 The zeros.

The initial and modiﬁed receptances.2 ¼ À0.5 3 3.5 5 . 0:0404Þ.E. Example 3—initial receptance (solid line) and modiﬁed receptance (dashed line). plotted in Fig.7i and m1. Four nonlinear equations are solved simultaneously À Á det I À Hðl1 ÞB diagðf i þ l1 gi ÞBT ¼ 0. Poles and zeros are assigned to h11 at l1. m1. À Á det I À Hðl2 ÞB diagðf i þ l2 gi ÞBT ¼ 0. respectively.5 3 3. respectively.5 4 4.2 ¼ À 0:025 Æ 1:2i. Â À Ã Á adj I þ Hðm2 ÞB diagðgi þ m2 f i ÞBT Hðm2 Þ 11 ¼ 0 which yields to the following gains: F ¼ diagð0:4023. are represented by the solid (blue) and dashed (red) lines. for the point receptance h22. m3.9i.4 ¼ À 0:037 Æ 2i.5 4 4. G ¼ diagð0:3720.5 1 1.5 5 Frequency (rad/s) Fig. 3. 4.ARTICLE IN PRESS 1398 J. Example 3 (Assignment of poles and zeros together). 0:6836Þ: 102 Amplitude 100 10−2 10−4 0 0 −50 Phase −100 −150 −200 0 0.2 ¼ À0.5 1 1.0270. Á Â À Ã adj I þ Hðm1 ÞB diagðgi þ m1 f i ÞBT Hðm1 Þ 11 ¼ 0. 0.5 2 2.5 2 2. Mottershead et al.00870. / Journal of Sound and Vibration 311 (2008) 1391–1408 where À C þ B diagðf i ÞB T Á 2 0:1493 ¼ 0 0 0:1480 ! and À K þ B diagðgi ÞBT Á 2 ¼ 7:8691 À1 À1 4:5223 ! which yields the zeros.

2 ¼ À 0:0200 Æ 0:7000i. Mottershead et al. represented by the solid (blue) and dashed (red) line. 5. using these gains. 6. (a) First mode (stem bending) and (b) second mode (stem twisting).E.2 ¼ À 0:0080 Æ 0:9000i.4 ¼ À 0:0992 Æ 1:6497i. l5. respectively. 250 mm 3 mm 60 mm 30 mm 12.ARTICLE IN PRESS J. Fig. .6 ¼ À 0:0798 Æ 2:2755i and zeros. / Journal of Sound and Vibration 311 (2008) 1391–1408 1399 State-space analysis. m3. l1. T-shaped plate: (a) dimensions and (b) experimental arrangement. 4. m1. l3.4 ¼ À 0:0654 Æ 2:1007i: The initial and modiﬁed receptances h11 are plotted in Fig.5 mm 100 mm 50 mm Fig. results in the poles.

125 Hz.ARTICLE IN PRESS 1400 J. (b) h12(io) measurement—solid line. . frequency resolution 0. ﬁtted curve—dashed line. Mottershead et al. 6. implemented using MATLAB/Simulink and dSPACE. signals from the Kistler accelerometers were integrated twice by digital means. 7. The third natural frequency. thereby enabling both velocity and displacement feedback. number of hits 20 and an exponential window with a decay to 1% applied to the measured accelerations.E. The internal active damper that forms part of the inertial actuator. based on analogue integration of acceleration was not used. Instead. Rational fraction curve ﬁt (a) h11(io). in-phase arm bending. The ﬁrst two mode shapes are shown in Fig. The ﬁrst mode was a stem bending mode and the second showed stem twisting with the two arms in anti-phase. (a) 10−3 Amplitude 10−4 150 200 250 300 350 Frequency (rad/s) 400 450 (b) 10−3 Amplitude 10−4 150 200 250 300 350 Frequency (rad/s) 400 450 Fig. occurred at 125 Hz (785 rad/s) some distance away from the ﬁrst two modes. / Journal of Sound and Vibration 311 (2008) 1391–1408 5. The ﬁrst two natural frequencies of the open-loop system were at 40 Hz (252 rad/s) and 52 Hz (325 rad/s). The H1 estimator was applied using the following test parameters: sample rate 256 Hz. Open-loop receptances were measured by means of a modal test using hammer excitation with the inertial actuators in place but not operational. Experimental examples Experiments were carried out on the T-shaped plate shown in Fig. 5 using two sets of collocated sensors (Kistler accelerometer type 8636C50) and inertial actuators (Micromega Dynamics type IA-01).

5 0. The receptances h11(io) and h22(io) were almost identical because of geometric symmetry of the T-plate and due to linearity h12(io) and h21(io) were found to be very similar. approximations and assumptions associated with ﬁnite element models are avoided completely. 9. as required by the theory.4 0.ARTICLE IN PRESS J.2 0. Mottershead et al. / Journal of Sound and Vibration 311 (2008) 1391–1408 1401 The purpose of using receptances instead of the M.E. The coefﬁcients were found by solving a least-squares problem. Actuator force–voltage transfer function. Solid line: open-loop receptance. Dot-dashed line: closed-loop receptance for the second test. 7 where measurements are represented by full lines (blue) and ﬁtted curves are shown as dashed lines (red). 100 200 300 400 500 600 400 500 600 10−3 Receptance (Amplitude) 10−4 200 250 300 350 Frequency (rad/s) 400 Fig. K matrices is that uncertainties. which should be well conditioned so that the coefﬁcients are not sensitive to small changes in the measurements. C. . 8. Dashed line: closed-loop receptance for the ﬁrst test. The ﬁtted receptances are presented in Fig. The good agreement shown in Force/Voltage (Amplitude) 0. However H(io) is available from vibration experiments and not H(s). Assignment of poles. Rational fraction polynomials were ﬁtted to the measured terms in H(io) and the coefﬁcients of the numerator and denominator polynomials determined [29].1 0 0 Force/Voltage (Phase) 200 100 0 −100 −200 0 100 200 300 Frequency (rad/s) Fig.3 0.

1:476 Â 10À10 s4 þ 4:987 Â 10À9 s3 þ 2:515 Â 10À5 s2 þ 0:0003862s þ 1 h12 ðsÞ ¼ h21 ðsÞ ¼ It should be pointed out that the rational fraction polynomials represent a model of the system. In the case of lightly damped systems it is likely that the identiﬁed rational fraction polynomial will be accurate if the curve ﬁt agrees closely with the measured terms in H(io). 7. An expression for the force–voltage transfer function of an inertial actuator with a ﬁxed base is given by Preumont [27]. Assignment of zeros: (a) h11 and (b) h22. 10. The following polynomials were identiﬁed: h11 ðsÞ ¼ h22 ðsÞ ¼ 7:956 Â 10À10 s2 þ 1:382 Â 10À8 s þ 6:438 Â 10À5 . was similarly obtained for phase. However all that is required of this model is that it is accurate at the chosen location of the poles and zeros to be assigned. . for amplitude.ARTICLE IN PRESS 1402 J. Above a critical frequency it is shown that the actuator behaves as an ideal force generator (a) 10−3 Unmodified H11 Zero assignment at −10 + −300i Amplitude 10−4 200 250 300 Frequency (rad/s) 350 400 (b) 10−3 Receptance (Amplitude) Unmodified H22 Zero assignment at −10 + −300i 10−4 200 220 240 260 280 300 320 340 Frequency (rad/s) 360 380 400 Fig. Mottershead et al.E. / Journal of Sound and Vibration 311 (2008) 1391–1408 Fig. 1:476 Â 10À10 s4 þ 4:987 Â 10À9 s3 þ 2:515 Â 10À5 s2 þ 0:0003862s þ 1 À1:334 Â 10À10 s2 À 4:678 Â 10À9 s þ 5:208 Â 10À6 .

. 12. (a) 0 −10 Imaginary −20 −30 −40 −50 −20 −10 0 10 Real 20 30 40 (b) 0. / Journal of Sound and Vibration 311 (2008) 1391–1408 1403 10−3 Receptance (Amplitude) Unmodified Zero Assignment at −15+ −295i .5 0 Real 0. Simultaneous assignment of poles and zeros. Polar plot of det[I+0.37 Â H(io)B diag(gi+iofi)BT]: (a) full polar plot and (b) magniﬁed view close to the origin.ARTICLE IN PRESS J.2 −0. Pole Assignment at −12 + −265i 10−4 200 250 300 Frequency (rad/s) 350 400 Fig.5 Fig. Mottershead et al. 11.1 0 −0.1 −0.E.2 0.

Experiment 3 (Simultaneous assignment of poles and zeros).8) were found and in the second test G ¼ diag(124. respectively. To maintain geometrical symmetry of the T-plate identical zeros were assigned to h11 and h22 at m1. agree closely with the imaginary part of the assigned poles and zeros.4 ¼ À257375i. .10955) and F ¼ diag(4.2 ¼ À127284i and l3. Gains were determined as F ¼ 1. Mottershead et al.E.75) were found and the resulting open-loop and closed-loop receptances were found as shown in Fig. The open-loop transfer function between the input 30 20 10 0 −10 −20 −30 −40 −50 −60 0 100 200 300 400 Frequency (Hz) 500 600 700 Fig. 10 where the frequency of the zeros is seen to agree very well with the assigned value of 300 rad/s and the poles remain stable. 5. A gain of 0. 11.4.8I. the peaks of the dashed and dot-dashed lines can be seen to agree very well with the imaginary parts of the assigned poles for the ﬁrst and second test. the force being measured by a force sensor inserted between the base of the actuator and a heavy rigid mass.11.2 ¼ À107290i and l3.14339) and F ¼ diag(0. In all of the three experiments the force and sensor distribution matrices were set to B ¼ DT ¼ I.2 ¼ À127265i.2 ¼ À157295i and m1.m2 ¼ À107300i. Poles were ﬁrstly assigned at l1. for the ﬁrst and second tests.37 Â H(io)B diag(gi+iofi)BT. 13. respectively.1. being the average value of the almost constant pure gain (zero phase) at frequencies greater than the natural frequency of the actuator at around 65 rad/s. The problem considered is that of Example 2 in Section 4 above. In the ﬁrst test. The receptance for the ﬁrst test is shown as a dashed line (red) and for the second test as a dot-dashed line (green). As expected. h11.m2 ¼ À107300i. where I denotes the identity matrix. Spectrum of eigenvalues l1 (dashed line) and l2 (solid line) of the open-loop transfer function 0. Stability robustness In this section the stability robustness of the experimental T-plate structure to active vibration control by pole-zero placement is addressed. / Journal of Sound and Vibration 311 (2008) 1391–1408 with constant amplitude and zero phase.ARTICLE IN PRESS 1404 J.11. 8. Fig. Experiment 2 (Assignment of zeros). that represents the closed-loop receptance.4 ¼ À227365i and then in a second test at l1. Experiment 1 (Assignment of poles). 8.24. 11 that the frequencies of the ﬁrst peak and dip of the dashed (red) line.5. Preumont’s analysis is in fact a simpliﬁcation of the real situation since the actuator was ﬁxed to the ﬂexible T-plate in our control application. Gain values of G ¼ diag(3720. The actuators were operational during the closed-loop modal tests used to determine the receptances from excitation by an instrumented hammer. This is the case shown in Fig. It can be seen in Fig. 5. G ¼ 3735I. However a small distance away from the T-plate resonances the expression was found to hold good. namely the assignment of zeros of h11 and h22 to m1. 2355) and F ¼ diag (2.19). 9 shows experimental receptances for the open-loop system as the full line (blue) and closed-loop receptances. Poles and zeros were assigned to h11 at l1.37 was obtained from the curve shown in Fig. Experimental open-loop and closed-loop receptances are shown in Fig. gains with the values of G ¼ diag(205.

. calibrated for displacement. Mottershead et al. dashed line—after added constraint. thereby conﬁrming that in practice spillover does not lead to instability at higher frequencies. It can be seen from Figs. 14.37 Â H(s)B diag(gi+sfi)BT. 0oooN. The spectrum in Fig. Solid line—without constraint.37 Â H(io)B diag(gi+iofi)BT]. dashed line—closed-loop receptance for s 40:6 from 50 to 100 Hz.5 1 0.5 2 1. (a) 3. does not enclose the point (0. where the gain of 0.E. 13 shows good roll-off of the eigenvalues l1 and l2 of the open-loop transfer function. / Journal of Sound and Vibration 311 (2008) 1391–1408 1405 voltage to the actuators and the output displacement y(s) is then deﬁned as 0. 0). respectively.ARTICLE IN PRESS J. 12(a) and (b) that this appears to be the case. The frequency-domain requirements for the stability of single-input–single-output (SISO) systems take the form of the standard Nyquist criterion and in the MIMO case they involve its multivariable generalisation [30].5 0 0 20 40 60 80 100 Frequency (Hz) 120 140 160 (b) 102 Receptance (Amplitude) 101 Open–loop response Closed–loop response (MSV > 0.5 3 Minimum singular value 2. Open-loop experiments were carried out by applying random voltages to the actuators in the range of 0–1000 Hz and measuring the output voltages. Robustness (a) minimum singular value. Solid line—open-loop receptance. from the dSPACE board.6) Closed–loop response (MSV > 0. (b) Receptance.37 represents the actuator dynamic as described previously.7) 100 200 250 300 Frequency (rad/s) 350 400 Fig. Thus the system may be considered stable if det[I+0. although the system is extremely close to being marginally stable. indicated by the dashed line (red) and solid line (blue). dot-dashed line—closed-loop receptance for s 40:7 from 50 to 100 Hz.

37 Â H(s)B diag(gi+sfi)BT] as explained by Maciejowski [31]. . Robustness (a) minimum singular value. The resulting control gains were found to be.5 2 1. 15.ARTICLE IN PRESS 1406 J. Solid line—open-loop receptance. dashed line—closed-loop receptance for s 40:8 from 50 to 100 Hz.45 (6 –10 Hz) 100 200 250 300 350 400 Frequency (rad/s) Fig.8) Closed–loop MSV > 0. a constraint of s 40:7 was applied over the frequency range of 50–100 Hz.5 0 0 20 40 60 80 100 Frequency (Hz) 120 140 160 (b) 102 Receptance (Amplitude) Minimum singular value 101 Open–loop receptance Closed–loop with constraint (MSV > 0. / Journal of Sound and Vibration 311 (2008) 1391–1408 The robustness of the closed-loop system can be improved by increasing the minimum singular value s of [I+0. Mottershead et al. dashed line—after added constraint.5 3 2. (b) Receptance. ! ! 12:7 0 3734:9 0 F¼ .E. thereby deﬁning the following constraint: Â Ã s I þ 0:37 Â HðsÞB diagðgi þ sf i ÞBT 4r on the solution of the nonlinear characteristic equations. dot-dashed line—closed-loop receptance for s 40:9 from 50 to 100 Hz and s 40:45 from 6 to 10 Hz. Solid line—without constraint. In our particular example.5 1 0. G¼ . 0 13:8 0 3734:9 (a) 4 3.9 (50 –100 Hz) and MSV > 0.

The solid (blue) line in Fig. The stability robustness may be improved by applying a constraint to the singular values of the matrix return difference. On the ﬂexibility offered by state feedback in multivariable systems beyond closed loop eigenvalue assignment. now with two constraints. [9] J. P. Ram. Mottershead et al. M. 14(b) represents the unconstrained open-loop receptance.M. Mares. . Linear Algebra and its Applications 257 (1997) 29–48. 14(a) the solid (blue) line denotes s versus frequency without constraint. [3] H. Illustrative numerical examples are presented. Robust pole assignment in linear state feedback. obtained by applying gains determined from the analysis. 15 shows the minimum singular values and the closed-loop receptances for the system. / Journal of Sound and Vibration 311 (2008) 1391–1408 1407 In Fig. [5] J. Nichols. C.J. In the range 50–100 Hz the lowest singular value was constrained such that s 40:9. as are the results of a physical experiment using collocated inertial actuators and accelerometers. The structural modiﬁcation inverse problem.E. Kimura.C. Wonham. [4] B. Transactions of the American Society of Mechanical Engineers. On pole assignment in linear systems with incomplete state feedback. [2] E. Mechanical Systems and Signal Processing. S. IEEE Transactions on Automatic Control AC-15 (1970) 348–351. 14(a) at low frequencies. The dashed (red) and dot-dashed (green) lines represent the closed-loop receptances after the added constraints s 40:6 and 0:7 were applied over the range of 50–100 Hz. Datta. An inverse method for the assignment of vibration nodes. IEEE Transactions on Automatic Control AC-21 (1976) 689–692. Gardonio (Southampton University). N. IEEE Transactions on Automatic Control AC-12 (1967) 660–665. It is seen that the effect of the constraints is to add damping and thereby improve the stability robustness of the T-plate system. Stanway (Shefﬁeld University) and Professor G.E. [8] J. Davison. [6] B. 0 22:64 0 2164 Fig. G¼ . Of course. Dr. [7] I. R. Structural modiﬁcation for the assignment of zeros using measured receptances. Elhay. C. Acknowledgements The authors wish to acknowledge helpful conversations with Professor P. International Journal of Control 41 (5) (1985) 1129–1155. Yao (Jilin University). A series of modal tests using hammer excitation were carried out with the feedback control gains obtained with the added constraints s 40:6 and 0:7. Pole assignment by gain output feedback. receptances H(s) are determined from the measured H(io) by ﬁtting rational fraction polynomials. Closed-loop receptances. Y. S. this is achieved at the cost of reduced accuracy in the placement of the zeros. Mechanical Systems and Signal Processing 7 (1993) 217–238. Kautsky.M. This results in considerably increased damping over the previous test and further deterioration in the accuracy of placing the zeros. In the experiments. Journal of Applied Mechanics 68 (5) (2001) 791–798. Control gains were then obtained as ! ! 17:35 0 1709:5 0 F¼ .N. Van Dooren. Special Issue on Inverse Methods in Structural Dynamics 15 (1) (2001) 87–100. Braun. The method does not require the usual M. Friswell. Dr.-F.K.ARTICLE IN PRESS J. References [1] W. have peaks and dips that agree very closely with the imaginary parts of the assigned poles and zeros. Orthogonality and partial pole placement for the symmetric deﬁnite quadratic pencil. Mottershead. 6. was then added to the range 6–10 Hz to address the sharp ‘dip’ in the dashed line of Fig. Bucher. to assign poles and zeros (natural frequencies and anti-resonances). Conclusions Active vibration suppression by eigenvalue assignment using output feedback and the receptance method is presented.T. s 40:45. A. IEEE Transactions on Automatic Control AC-20 (1975) 509–516. Shenton (Liverpool University). Mottershead. The dashed (red) line is the case of s 40:7 for the range of 50–100 Hz. On pole assignment in multi-input controllable linear systems.I. K matrices but uses measured receptances from the open-loop system instead. A further constraint.E. Moore.

Tisseur. E. W. [21] L. Parameter estimation from frequency response measurements using rational fraction polynomials. P. 1990. K.V. Journal of Sound and Vibration 284 (1–2) (2005) 249–265. Elliot. [12] J. Fuller. Mottershead et al. [18] E.R. Nelson. D. Multivariable Feedback Design. second ed. Active constrained layer damping: a state-of-the-art review. Mechanical Systems and Signal Processing 15 (1) (2001) 173–188. L. Receptance method in active vibration control.J. Addison-Wesley. Feedback control of ﬂexible structures.H. New York. S. 1989.G. Gopinathan.A. J. Ram. [22] K.K. Mechanical Systems and Signal Processing 20 (8) (2006) 2290–2304. Wiley. The quadratic eigenvalue problem. Adaptive Structures. Benassi. The stability of multivariable systems. Academic Press. Mottershead. Maciejowski. Richardson. [28] C. Elliot. G. IEEE Transactions on Automatic Control AC17 (1972) 105–107. Jian. . Bianchi. P. Ouyang. Wiley/Interscience. / Journal of Sound and Vibration 311 (2008) 1391–1408 [10] J. Elliot. Gardonio. Gibbs. S. Journal of Sound and Vibration 274 (2004) 193–213. Proceedings of the IMechE.M. Structural modiﬁcation. Dordrecht. V. [14] L. P. [15] R. Structural modiﬁcation. Ouyang. Stancioiu.E. [11] A. SIAM Review 43 (2) (2001) 235–286. 1982.J. Balas. [19] R. [16] P. part 2: assignment of natural frequencies and antiresonances by an added beam. 1996.P. [30] H. Structural modiﬁcation of a helicopter tailcone. San Diego. J. N. Active vibration isolation using an inertial actuator with local force feedback control. Sims. Mottershead. M. Wokingham.D. Bianchi. Part II: design of decentralised control units. Smart panel with multiple decentralised units for the control of sound transmission. [26] Y. Mottershead. E. Preumont. Rongong. S. [23] F. Dynamics and Control of Structures.E. Journal of Sound and Vibration 284 (1–2) (2005) 267–281.E.R.E. Part III: control system implementation.I.J. Mottershead. S. Part I: theoretical predictions. [13] J. Active Control of Vibration. M. Mechanical Systems and Signal Processing 20 (1) (2006) 5–44. Vibration Control of Active Structures. Proceedings of the First International Modal Analysis Conference. Kyprianou.M. Bianchi. Elliot. Dynamics and Control. Clark. Ram. Rosenbrock. Journal of Systems and Control 217 (6) (2003) 437–456. H.M. American Institute of Aeronautics and Astronautics Journal 45 (3) (2007) 562–567. Journal of Sound and Vibration 274 (2004) 163–192.E. Lim. FL. Friswell. H.J. Mottershead.E. Journal of Sound and Vibration 276 (2004) 157–179. Active vibration control of a car body based on experimentally evaluated modal parameters. 2002.A. Varadan. Journal of Sound and Vibration 298 (1–2) (2006) 366–384. 1998. IEEE Transactions on Automatic Control AC-23 (1978) 673–679. Y. Designing distributed modal sensors for plate structures using ﬁnite element analysis. S. A. [17] P. [31] J. Elliott.. Stanway. Meirovitch. S. Smart panel with multiple decentralised units for the control of sound transmission. Stobener.J. Saunders. [27] A.ARTICLE IN PRESS 1408 J. Kyprianou. New York. Kluwer Academic Publishers. Part I.L. Gardonio. S. Formenti.J. England. Smart Materials and Structures 8 (1999) 324–337. [20] M. Meerbergen. Smart panel with multiple decentralised units for the control of sound transmission.V. Finite element simulation of smart structures using an output feedback controller for vibration and noise control. Gardonio. H. Inverse eigenvalue problems in vibration absorption: passive modiﬁcation and active control. [25] U. James. Orlando. Journal of Sound and Vibration 274 (2004) 215–232. M. [29] D. Gaul. part 1: rotational receptances. Tehrani. V.-H. Varadan. Shahverdi. Gardonio. J. [24] Y.

- The Sensitivity Method in Stochastic Model Updating
- Frequency Response Matching to Optimise Wind Turbine Test Data Correlation
- 2003 IMAC Oresund A4
- The Stochastic Finite Element Method_Theory and Applications
- The Stochastic Finite Element Method_PAST, PRESENT and FUTURE
- 7_Identification of Multi-Degree-Of-freedom Non-linear Systems Under Random Excitations by the Reverse Path Spectral Method
- 6_Past_present and Future of Nonlinear System Identification in Structural Dynamics
- Review and Assessment of Model Updating for Non-linear, Transient Dynamics
- 298_1
- ACM Weld
- Simulation and Experimental Validation of Modal Analysis for Non-linear Symmetric Systems
- System Identification Using Associated Linear Equations
- Past_present and Future of Nonlinear System Identification in Structural Dynamics
- 4_Identification of Multi-Degree of Freedom Non-linear Systems Using an Extended Modal Space Model
- Fagan Et Al - J of Eng in Medicine 2002a
- Adaptive vibration control through a SMA embedded panel.pdf
- Mechanical Daming Simulation in MSC Nastran
- Finite Element Analysis of the Dynamic Behaviour of an Engine Block and Comparison With Experimental Modal
- Component Mode Synthesis of Space Station Freedom Using Msc Nastran Superelement Architecture
- Active Control of Plate Like Structure for Vibration and Sound Suppression
- A Comparison of MSC NASTRAN With Model Solution
- On the Fundamental Transverse Vibration Frequency of a Free-free Thin Beam With Identical End Masses
- Smart Control of Cantilever Structures Using
- Conversational
- Tom W B Kibble, Frank H Berkshire-Classical Mechanics-World Scientific Publishing Company(2004)

Active Control

Active Control

- IEEE_IE03
- MAEG3050 (2)
- Chapter 10
- Benchmark Examples
- Lecture 14
- 09 Power Method
- Cookbook Matrix
- S06 - 2D Linear Dynamic Systems
- Orthogonal Matrices
- 2 Ising Spins
- Eigenvalues and Eigenvectors
- hw4ss
- Eigen Beamforming
- 1464_IEEETAC05Fev.pdf
- Bode Compensador Phaselag1web
- 04586595
- AHP Technique
- SCHEME - E Sixth Semester (IE).pdf
- 2000
- M.M.Deminov and A.A.Slavnov- A practical gauge invariant regularization of the SO(10) grand unified model
- Errores SL
- trom pho tan so
- Gabor-Kernel Fisher Analysis
- on the norm
- Quantum Assignment -Aakash
- Dmitry E. Pelinovsky and Panayotis G. Kevrekidis- Dark solitons in external potentials
- 63 Equation Post Processing
- Confessions
- re30131_0203
- Cellular manufacturing – A time based analysis to the layout problem (Seminar report)

Are you sure?

This action might not be possible to undo. Are you sure you want to continue?

We've moved you to where you read on your other device.

Get the full title to continue

Get the full title to continue reading from where you left off, or restart the preview.

scribd