Signal Processing and Linear Systems-B.P.Lathi copy

However if the mode were to decay a t a r ate less t

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: uld be designed so t hat t heir characteristic roots are n ot close t o t he engine's frequency of vibration. P roper design of this platform can not only avoid resonance, b ut also a ttenuate v ibrations if t he system roots are placed far away from t he frequency of vibration. 2 .8 Appendix 2.1: Determining t he Impulse Response We now derive t he u nit impulse response of a n LTIC s ystem S specified by t he n th-order differential equation Fig. 2 .21 Build up of system response in resonance. Q (D)y(t) = P (D)f(t) or Recall t hat w hen>. = j w, t he system is marginally stable. As we have indicated, t he full effect of resonance cannot be seen for an asymptotically stable system; only in a marginally stable system does t he resonance phenomenon boost t he s ystem's response t o infinity when t he s ystem's input is a c haracteristic mode. B ut even in a n a symptotically stable system, we see a manifestation of resonance if its characteristic roots are close t o t he i maginary axis, so t hat Re >. is very small. We c an show t hat when t he c haracteristic roots of a system are I j ± j wo, t hen, t he system response t o t he i nput ejwot or the sinusoid cos wot is very large for small I j. t T he r esponse drops off rapidly as t he i nput signal frequency moves away from woo T his frequency-selective behavior can be studied more profitably after a n understanding of frequency-domain analysis is acquired. For this reason we p ostpone full discussion of t his s ubject until C hapter 7. Importance o f t he Resonance Phenomenon T he r esonance phenomenon is very i mportant because it allows us t o design frequency-selective systems by choosing their characteristic roots properly. Lowpass, bandpass, highpass, a nd b andstop filters are all examples of frequency selective networks. I n mechanical systems, t he i nadvertent presence of resonance can cause signals of s uch t remendous magnitudes t hat t he system may fall a part. A musical note (periodic vibrations) of proper frequency can s hatter a glass if t he frequency is m atched t o t he c haracteristic root of the glass, which acts as a mechanical system. Similarly, a company of soldiers marching in step across a bridge t This follows d irectly from Eq. (2.74) w ith .\ = (j + j wo a nd < = (j (Dn + a n_ID n- 1 + ... + a ID + ao)y(t) = (bnDn + bn_ID n - 1 + ... + bID + b o)f(t) (2.77a) (2.77b) In Sec. 2.3 we showed t hat t he impulse response h (t) is given by h (t) = A 0 8(t) + characteristic modes (2.78) We now show t hat in t he above equation Ao = bn where bn is t he coefficient of t he n th-order t erm on t he r ight-hand side of Eq. (2.77b). W hen t he i nput f (t) = 8(t), t he response y (t) = h (t). Therefore, from Eq. (2.77b), we o btain ( Dfi + a n_I Dn - 1 + ... + a ID + ao)h(t) = (bnDn + bn_ID n- 1 + ... + bID + bo)8(t) In this equation we s ubstitute h (t) from Eq. (2.78) a nd compare t he coefficients of similar impulsive terms on b oth sides. T he highest order of t he derivative of impulse on b oth sides is n , w ith its coefficient value as Ao o n the left-hand side a nd bn on the right-hand side. T he two values must be...
View Full Document

This note was uploaded on 04/14/2013 for the course ENG 350 taught by Professor Bayliss during the Spring '13 term at Northwestern.

Ask a homework question - tutors are online