ra , Chapter Root Locus Techniques (8.1 Introduction 1B [Rov Tossa graphical presentation ofthe closed-loop poles asa system parameter is aried|s a powerful method of analysis and design for stability and transient ee oa) te amc corel ene ae Ae 16 conprebend (rattative point oftiew. and hence they rel heavily upon mathemati The rot leas Geer ints chapter sa graphical eehnige that gives us the qualitative description Stacntrol system's performance that we ae looking fr and also serves sa powerful sudan fol that ysis more information than the methods already disused. {Up to this point gains and other sytem parameters were designed to yeld a desired ransont response fron first and second-order ystems Eventhough the \ fot locus ean be used to solve the sme kind of problem, its real pone is in is ‘Ply to provige solutions for systems of order Righer than 2 For example, under the night conditions, a fourth-order syten’s parameters can be designed to yield a frerpercent overshoot and seting time using the concepts learned in Chapter 4 “The root loces can be used to describe qualitatively the performance of system as varios parameters are changed For example, the effect of varying gain iron percent overshoot, settling time, and peak time canbe vividly displayed. The Ghaltaive description can then be vred with quantitative analysis. ‘Jeide ramet response, the rot loc also gives a graphical representation ofa sytem salty. Weean clearly se ranges of stability. anges of instability, and the eondtions that cause a system to break into osilaton. Tefore presenting fot locus Ie us review two concepts that we need forthe ‘ensuing discussion: (1) the control system problem and (2) complex numbers and : their representation as vectors ; er of ines The Control System Problem a Alt Opole TF ‘We have previously encountered the control system problem in Chapter 6: Whereas the ‘poles of the open-loop transfer function are easily found | Can he G32) Fok inspection an donot change wi changes in sem ; transfer funeton re more fiat findypcally. they cannot nd fey der | Che 9% Ge ctosed-toop system's characteristic polynomial, the denominator of the closed-loop ‘Parsfer function), and further, the closedLoop poles change with changes in system gain. as £ cheys ‘Atypical closed loop feedback control system is shown in Figute 8.1(@). The open-loop tanster fonction was defined in Chapter Sas KG(s)A(). Ordinal, we the pre, f4 hed-lag » Formed rae libel Ae Plncans mae At Me nm and Leg Chee a: PQA koi NAS gai sv [no], Tekin mH Fain FIGURE R.A a.Closedoop — Speer beqivelet transfor tion function « »392 Chapter Root Locus Techniques (8.2. Defining the Root Locus {A security camera system similar to that shown in Figure 84(a) can automatically follow a subject. The tracking system monitors pixel changes and positions the camera to center the changes. “The root locus technique can be used to analyze and design the effect of loop ‘gain upon the system’s transient response and stability. Assume the block diagram Fepresentation ofa tracking system as shown in Figure 8.4(b), where the closed-loop ‘poles of the system change location as the gain, K, is varied. Table 8.1, which was formed by applying the quadratic formula to the denominator of the transfer function in Figure 8.4(c), shows the variation of pole location for different values of gain, K, The data of Table 8.1 is graphically displayed in Figure 8.5(a), which shows each pole and its gain. 'AS the gain, K, inereases in Table 8.1 and Figure 8.5(a), the closed-loop pole, ‘which is at -10 for K = 0, moves toward the right, and the closed-loop pole, which is tO for K = 0, moves toward the left. They meet at ~5, break away from the real axis, ‘and move into the complex plane. One closed-loop pole moves upward while the ‘other moves downward. We cannot tell which pole moves up or which moves down. In Figure 8.5(6), the individual closed-loop pole locations are removed and their paths are represented with solid lines. It i this representation of the paths of the Ek suns Mowe Camere yesene Sensors Amplifier —_amdcamera_ postion =a _, fs 7 aw, z er) plot the poles of the Ciofed- oop TF o as «choc m9 x ow, nes Ae Fire Tee (tgntne oF F ‘where K= Rik a inc(ee eS FIGURE 8.4 a Security cameras with auto tracking can be used 10 follow moving objects avtomaticallyb, Block diagram; closed-loop transfer function82. Defining the Root Locus 393 TABLE 8.1. Pole location as function of gain forthe systemof Figure 84 K Potet Pole? ° =10 0 5 947 033 0 887 “13 5 8:16 -18 0 14 276 2s I 5s 0 “34204 -5-R 8 5+ 16 -5- As 40 34887 5-780 4s -S+HA7 3-KaT 50 3458 5-18 o FIGURE S.S a. Poe plot from Table 8.1; be root locus closed-oop poles asthe gains varied that we call root los: For most ofour work, — gle the discussion will be limited to postive gain, or K > 0. Rewmik BO “The rat locus shows the Fo eee neem kc 75 2 oveldongal See ae eS aaitommamers io Aes > cCivoly available through the analytical techniques covered in Chapter 4, the following Ce ee i real ee con ee neg me uncer sears ofthe valu ft el part of he comps poise ana these. Remote HO: Foc the underdampal reqree CE>26), te teal pats of the compen ples at dliiys He seme: <> SEMIng ime (> tre Gene for all K > 25 - fhe K inrtases See damping (abe decttases and P.o. inrvcases. Mse as kK inecerses » the damped fieg monty (Wd) of 5c iMabans ‘net Reaik 4 Remik #3nay. 394 ue OK Fre tod Joeus never cdesses ON inyo the Conta alf_plone 9 She system cs always sheble for 4 all valnes of ke (Chapter Root Laces Teciques Since the seting ime is inversely proportional tothe real par ofthe complex poles for thissecond.ordersystem, the conclusions thatregardlessof the value of gain theseting time for the system remains the same under af conditions of underdamped responses “SE-KSo, as We increase the gain, the damping ratio diminishes, and the pereent overshoot increases. The damped frequency of oscillation, which is equal to the imaginary part of the pole, also increases with an increase in gan, resulting in a reduction ofthe peak time. Finally, since the root locus never crosses over into the Fight half-plane, the system i always stable, regardless of the value of gain, and can never break into a sinusoidal oseillation. “These conclusions for such a simple system may appear to be trivial. What we are about to se Is that the analysis is applicable to systems of order higher than 2. For these systems, itis cificult to He transient response characteristic to the polo location. The root locus will allow us to make that association and will become an important technique in the analysis and design of higher-order systems, (8.3. Properties of the Root Locus {In Section 8.2, we arrived at the root locus by factoring the second-order polynomial in the denominator of the transfer function. Consider what would happen if that polynomial were of fifth or tenth order. Without a computer, factoring the polyno- ‘mial would be quite a problem for numerous values of gain We are about to examine the properties of the root locus. From these ‘properties we will be able to make a rapid sketch ofthe root locus for higher-order Systems without having to factor the denominator of the closed-loop transfer function. "The properties ofthe root locus can be derived from the general control system. of Figure 8.1(a). The closed-loop transfer function for the system is 7) = KS) TIE KG), From Eq, (8:12), a poles, exists when the characteristic polynomial in the denor nator becomes zer0, or (812) (8.13) 1,22, KG@)H@) 1Z@k +1180" were Lis represented in polar form as 1 (2k-+ 1180". Alternately. a value of 2 dlosed-lop pole if i fy the pales of Th eS > [RGM = 1 (14) and & _s[ZRG@H® = Ge DEF (315) Equation (8.13) implies that if a value of s is substituted into the function KG(s)H(s), compiex number results Ifthe angle ofthe complex number isan odd ‘multiple of 180 that value of sis a system pole for some particular value of K. What pe iol | 16) pe84 Sketching the Root Locus |-Assessment Exercise 8.2 PROBLEM: Given a unity feedback system that has the forward transfer function K(s 42) Grae a {do the following: ‘a, Calculate the angle of G(s atthe point ( ~3 + j0)by finding the algebraic sum of angles ofthe vectors drawn from the zeros and poles of G(s) to the given point. », Determine ifthe point specified in a is on the root locus. ¢ If the point specified in a is on the root locus find the gain, K, using the lengths of the vectors. ANSWERS: 1. Sum of angles = 180" bs, Point is on the root locus 10 ‘The complete solution is at www.wiley.com/college/nise. 397 wes Tyla {Use MATLAB and the fo lowing statements sate ‘ShileAstenment Exercise 2 +031 Gueezianat4eatiyr ‘Toeten(200/p6)" =~ ‘angielo) weabotsie mam (8.4 Sketching the Root Locus It appears from our previous discussion that the root locus can be obtained by ‘sweeping through every point in the s-plane to locate those points for which the ‘angles as previously deseribed, add up to an odd multiple of 180°, Although this task is tedious without the aid of a computer, the concept can be used to develop rules, that ean be used to sketch the root locus without the effort required to plor the locus. (Once a sketch is obtained, itis possible to accurately plot ust those points that are of interest to us for a particular problem. ‘The following five rules allow us to sketch the root locus using minimal calculations. The rules yield a sketch that gives intuitive insight into the behavior ‘of a control system. In the next section, we refine the sketch by finding actual points (or angles on the root locus. These refinements, however, require some calculations or the use of computer programs, sich as MATLAB. 9 Riot ) vcs sketcle 1. Number of branches. Each closed-loop pole moves as THe gain is varied. If we define a branch as the path that one pole traverses, then there will be one branch for each closed-loop pole. Our frst rule, then, defines the number of branches of, the root locus! The number of branches of the root locus equals the number of closed-loop poles. [As an example, look at Figure 8.5(b), where the two branches are shown. One originates at the origin, the other at —10. 2, Symmetry. [fcomplex closed-loop poles do not exist in conjugate pairs the resulting polynomial, formed by multiplying the factors containing the closed-loop poles TheFIGURE 8.8 Poles and zeros ofa general ‘open-loop system with test points, ,,on the real axis, ts Chapter $ Root Locus Techniques ‘would have complex coefficients Physically realizable systems cannot have complex Coefficients in their transfer functions Thus, we conclude: |The root locus is symmetrical about the rea axis. ‘An example of symmetry about the real axis is shown in Figure 8.5(b). 3. Real-axis segments, Let us make use of the angle property, Eq. (8.15), of the points on the root locus to determine where the real-axis segments of the root locus exist. Figure 8.8 shows the poles and zeros of a general open-loop system, Ifan attempt is made to calculate the angular contribution of the poles and zeros at each point, Ps, Pz, P:.and Ps, along the real axis, ‘we observe the following: (1) At each point the angular contribution of a pair of open-loop complex poles or zeros is zero, and (2) the F,*+? contribution of the open-loop poles and open-loop zeros to the left of the respective point is zero. The conclusion is that the only contri- bution tothe angle at any of the points comes from the open-loop, real- ‘axis poles and zeros that exist to the right of the respective point. Ifwe calculate the angle at each point using only the open-loop, real-axis ‘poles and zer0s to the right of each point, we note the following: (1) The “angles on the real axis alternate between 0° and 180, and (2) the angle {is 180° for regions ofthe real axis that exist to the left of an odd number of poles and/or zeros. The following rule summarizes the findings: |i, finite open-loop poles andor finite open-loop zeras— (On the real ats, for K > Othe root locus exits to the left of an odd numberof real- Examine Figure 86(b). According to the rule just developed, the real-axis segments of the root locus are between ~1 and —2 and between ~3 and —4 ‘as shown in Figure 89. 4. Starting and ending points. Where does the root locus begin (zero gain) and end (inti fe gain)? The answer to this question will enable us to expand the sketch of the root locus beyond the real-axis segments. Consider the closed-loop transfer function, T\s), deseribed by Eq. (8.3). T(s) can now be evaluated for both large and small gains, K. As K approaches zero (small gain), KNo(s)Duls) 19) = a ()\Duls)+€ (623) From Eq, (8.23) we see that the closed-loop system poles at small gains approach the combined poles of G(s) and H(s). We conclude that the root locus begins at the poles of G(s)#(). the open-loop transfer function. FIGURE 8.9. Real-axis segments of the root locus fr the system of Figure 8.6 yxy‘84 _ Sketching the Root Locus At high gains, where K is approaching infinity, KNe(s)Duls) TW) * RNG GINA on ee ne eee eee the combined zeros of G(s) and H(s). Now we conclude that the root locus ends at the zeros of G(s)H(3), the open-loop transfer function. ee cal | The root locus begins atthe finite and infinite poles of G(s)H(s) and ends atthe ) (Raveena cng 2 ‘Remember that these poles and zeros are the open-loop poles and zeros. igs San oeeensec eet Pee tape roar eects er imag striae cre See eames ss sacrum 3 and —4 (see Figure 8.10). Thus, the poles start out at ~1 and ~2 and move Sapper areca coe Sees acre eee ee eee eee Sede ara ae eerie See oe ae eee cea Sena KGI)H(s) (825) SF DETD) ‘There are three finite poles, at s= 0,~1, and ~2, and no finite zeros. ‘A function can also have infinite poles and zeros. Ifthe function approaches infinity as s approaches infinity, then the function has a pole at infinity. If the function approaches zero as s approaches infinity, then the function has a zero at infinity, For example, the function G(s) =s has a pole at infinity, since G(s) approaches infinity ass approaches infinity. On the other hand, G(s) = 1/s has a zer0 at infinity, since G(s) approaches zero as s approaches infinity Every function of s has an equal number of poles and zeros if we include the infinite poles and zeros as well as the finite poles and zeros. In this example, wo FIGURE 8.10 Complete oot locus forthe system of Figure 86 399400 Chapters Root Locus Techniques Eq. (825) contains three finite poles and three infinite zeros. To illustrate, let s approach infinity. The open-loop transfer function becomes Kawi) =% (826) Each s in the denominator causes the open-loop function, KG(s)H(s), to become zero as that s approaches infinity. Hence, Eq. (8.26) has three zeros at infinity. “Thus, for Eq, (825), the root locus begins atthe finite poles of KG(s)H() and tends at the infinite zeros. The question remains: Where are the infinite zeros? We ‘must know where these zeros are in order to show the locus moving from the three finite poles to the three infinite zeros. Rule 5 helps us locate these zeros a infinity. Rule also helps us locate poles a infinity for functions containing more finite zeros than finite poles? We now state Rule 5, which will tell us what the root locus looks like as it approaches the zeros at infinity or as it moves from the poles at infinity. The derivation can be found in Appendix M.1 at www.wiley:com/college/nise, [The root locus approaches straight ines as asympioter as the Tocus approaches \ and angie, 0, as follows: Snite potes — finite zeros ae. “linite poles — #inite zeros é (26+ 1) —> | Fisite poles — #finite zeros Jains. Further the equation ofthe asymptotes is given by thereal-axsintercep, os} (27) (8.28) ” : where =0, 41,42, 13 andthe angles given in radians with respect tothe poste extension ofthe real axis Notice that the running index, k, in Eq. (8.28) yields a multiplicity of lines that account for the many branches of a root locus that approach infinity. Let us demonstrate the concepts with an example. Example 8.2 Sketching a Root Locus with Asymptotes PROBLEM: Sketch the root locus forthe system shown in Figure 8.11 Kaew cy, mo a Baber aero, FIGURE 8.11 System for Example 82. 7 Physical stems however, have more finite poles than finite zero since the implied ferentiation yes infinite ouput for dtcomtinaous input fenetions suchas step ips‘84 Sketching the Root Locus SOLUTION: Let us begin by calculating the asymptotes. Using Eq, (8.27), the real- axis intercept is evaluated as ee z 4) (829) “The angles ofthe lines that intersect at ~4/3, given by Eq. (828) are = inp pao co xf} form=0 (8300) = fork=1 (8.30) Sn/3 forth 2 (300) Ifthe value for k continued to increase, the angles would begin to repeat, The ‘number of lines obtained equals the difference between the number of finite poles and the number of finite zeros. Rule 4 states that the locus begins at the open-loop poles and ends at the ‘open-loop zeros. For the example there are more open-loop poles than open-loop zeros. Thus, there must be zeros at infinity. The asymptotes tell us how we get to these zeros at infinity. Figure 8.12 shows the complete root locus as well as the asymptotes that were just caleulated. Notice that we have made use of all the rules learned so far. The real-axis segments lie to the eft of an odd number of poles and/or zeros. The locus starts at the open-loop poles and ends at the open-loop zeros. For the example there is only one open-loop finite zero and three infinite zeros. Rule 5, then, tells us that the three zeros at infinity are at the ends of the asymptotes. aompuee / 401 FIGURE 8.12 Root locus and asymptots forthe sytem of Figure 811402 ‘Chapter 8 Root Locus Techniques |-Assessment Exercise 8.3 PROBLEM: Sketch the root locus and its asymptotes for a unity feedback system that has the forward transfer function K OO = FES NEFH) ANSWER: The complete solution is at www: wiley comicollege/nise. 8.5 Refining the sketch ‘The rules covered in the previous section permit us tosketch aroot locus rapidly. If we ‘want more detail, we must be able to accurately find important points on the root locus ‘along with their associated gain. Points on the real axis where the root locus enters oF leaves the complex plane—real-anis breakaway and break-in points—and the jo-axis ‘crossings are candidates. We can also derive a better picture ofthe root locus by finding, the angles of departure and arrival from complex poles and zeros, respectively In this section, we discuss the calculations required to obtain specific points on the root locus, Some of these calculations can be made using the basic root locus relationship that the sum of the zero angles minus the sum of the pole angles equals ‘an odd multiple of 180°, and the gain ata point on the root locus found asthe ratio ff (1) the product of pole lengths drawn to that point to (2) the product of zero Tengths drawn to that point. We have yet to address how to implement this task. In the past, an inexpensive tool called a Spirule™ added the angles together rapidly ‘and then quickly multiplied and divided the lengths to obtain the gain. Today we ean rely on hand-held or programmable calculators as well as personal computers. Students pursuing MATLAB will lam how to apply it to the root locus at the ‘end of Section 8.6. Other alternatives are discussed in Appendix H.2 at wwwwiley. ‘com/college/nise. The discussion can be adapted to programmable hand-held calcu- Tators All readers are encouraged to select a computational aid at this point. Root Tocus calculations can be labor intensive if hand calculations are used. ‘We now discuss how to refine our root locus sketch by calculating real-axis, breakaway and break-in points, jo-axis crossings, angles of departure from complex poles, and angles of arrival to complex zeros. We conclude by showing how to find ‘accurately any point on the root locus and calculate the gain. Real-Axis Breakaway and Break-In Points [Numerous root loci appear to break away from the real axis as the system poles rove from the real axis to the complex plane. At other times the loci appear to return to the real axis as a pair of complex poles becomes real. We illustrate this in Figure 8.13. This locus is sketched using the first four rules: (1) number of branches, (@) symmetry, (3) real-axis segments and (4) starting and ending points. The figure ‘hows a root locus leaving the real axis between ~1 and ~2 and returning tothe real axis between +3 and +. The point where the locus leaves the real axis{= Pecrsd wan eal A dhe SW crossmy Potals 9 He Feat part oF Sf ster ch- eq, 14+ KOH S435 +e 4 K(5~ 25 1d =0 > Cea) st (3° BENS + (2 41Ek) => 241K - EO Ne cae {see (3-Ble\ew) = 0 —@ | jw tj296\Z 406 ‘Chapter 8 Root Locus Techniques TABLE 82. Data for breakaway and break-in points for the rot locus of Figure 8:13 ‘eal-axs value Gain Comment “Lat (0.008557 1a 0.008585 -Ls ‘0.008603 -1s4 0.008617 =14s 0.008623 - Max. gine breakaway 146 0.008622 33 4685 34 srs 3s 33.000 36 30667 31 29440 38 23.00 = ‘Min. gain: break-in 39 20am ee ee illustrate instability at small values of gain and stability at large values of gain. These ‘ystems have a root locus starting in the right-half-plane (unstable at small values of ‘gain) and ending in the left-half-plane (stable for high values of gain). ‘To find the ja-axis crossing, we can use the Routh-Hurwitzcriterion, covered in (Chapter 6, as follows: Forcing a row of zeros in the Routh table will yield the gain; ‘going back one row to the even polynomial equation and solving forthe roots yields the frequency at the imaginary-axis crossing, Example 8.5 Frequency and Gain at Imaginary-Axis Crossing PROBLEM: For the system of Figure 811, find the frequency and gain, K, for which the root locus crosses the imaginary axis For what range of Kis the system stable? SOLUTION: The closed-loop transfer function forthe system of Figure 8.11 is eee + (640) TO) = aes iat + OT RETR Using the denominator and simplifying some of the entries by multiplying any row by a constant, we obtain the Routh array shown in Table 8.3. 'A complete row of zeros yields the possibilty for imaginary axis roots. For positive values of gain, those for which the root locus is plotted, only the s! row can yield a row of zeros. Thus, -K? 65K +720 (eal) From this equation K is evaluated as K a) el Ros) =o5_ Refining the Sketch ‘TABLE 8.3. Routh table for Eg. (840) # 1 14 3K 4 &@? 7 ee 9. 4 2 s0-K i eee Bre gf 2R=esk +700 ° 0 WK & # 2K o 0 Forming the even polynomial by using the »* row with K = 9.65, we obtain (90~ Kj? +21K = 80.35"? + 202.7 643) and sis found to be equal to +j1.59. Thus the root locus crosses the j-axis at +4/1.59 ata gain of 9.65. We conclude thatthe system is stable for 0 < K < 9.65. Another method for finding the jo-axis crossing (or any point on the root locus, for that matter) uses the fact that atthe je-axis crossing, the sum of angles from the finite open-loop poles and zeros must add to (2k-+ 1)180°. Thus, we ean search jo-axis until we find the point that meets this angle condition. A computer ‘program, such as the root locus program discussed in Appendix H.2 at www.wiley -comlcollege/nise or MATLAB, can be used for this purpose. Subsequent exam- piles in this chapter use this method to determine the jo-axis crossing. Angles of Departure and Arrival In thts subsection, we further refine ou sketch of the root oeus by finding angles eee aia fre clot ei eos, Comte Fa, AMR ope oy poet tcc musi Stet veomples Theron Teese te cpecbroy polit al Gaical WH open loop seroe In Orde 1. Sketch the root locas more ccortely, we want (0 calculate the root locis Gepartare ange from the complex poles and the arrival ange t the complex If we asus 1 point on the oot lous ¢ cose to a comple poe the sum of angles drawa fom al finite poles and zeros to this point is an odd nulipe of 180 Except for the pole that isc close to the point, we sstme all angles avn from all Cae esx er drewa recty otnapoa ster toe peet Tht nly unknown angle inthe suri the angle dram from the pole thats cose, We Se ig eee epee a a cee ee complex pole. Hence, from Figure 8.15(a), 01+ + 05 — 04 ~ 65+05 = Ok + DIST (44a) = 85 405 8485+ 85 — (2k + 1180" (e440) we assume a point on the root locus «close toa complex zero, the sum of ‘angles drawn from al finite poles and zeros to this point is an odd multiple of 180°. Except for the zero that i close tothe point, we can assume all angles drawn from. all other poles and zeros are drawn directly to the zero that is near the point. Thus,Ss pule_G: Root Locus Plotting Procedare Role es SiBpot: Angles of Departure/Arrival Sea Ante ss deen cata rs te cee pots oie rember ander ery Estes etnetal iapstsaennte opin cial Sls — | sa yanaee accone sisi Malini Po pn arirtil Icleplamrndefephl To tee BE encye ples Gr NBO g E-cnshesfore boot TAN9| 6s rem olber pels | (Am ged, Nock dy gi st Seeks oo eee Mi cach rove: G,-= \Bo+ Eawples hom poles A anples Gam cede angle ig with the perdve «oN ont. Root Locus Plotting Procedure Step 4: Angles of Departure/Arrival ‘By convention, te arial and depanure angles ae measured relative fo the ral nis, otha the positive rel axis i. Graphically, the angles look ike thismor Sa | se yea 2 = NBC 4 (Ona, + Pye, + Be +%,,,) — 1 es aha 123 + (lyn ~ Pern Pen) (ye, + Pan® Sp) —§ Root Locus Plotting Procedure ‘Step 4: Angles of Departure/Arrival .. Example For the system with forward: op transfer function se ee z+ y\oRet) | @riyerassy the angle of departure for he poe located a (2, 1) i equal to Op = 180-90" 135° 14 = STE Gmc hss pod ae eg = \80" oh = 0° et = tres hia 4) - or HF OB = wor (-Iee (a) -(00s#- of Op = — Of (From syynmelyy) Root Locus Plotting Procedure Step 4: Angles of Departure/Arrival .. Example. Forth sytem wth fora lop taser fncion (@ +8368 + 1796) +248) ices ese cee rely the angle of departure forthe pole located at (5,15) sequal 10 (= 180-0" [48-113 + 110" + 136° + 149° = 164") ae ae 5 (ate + Ona) 8D lé‘The ROOT LOCUS Procedure ‘The ROOT LOCUS Problem Problem description Given the closed lop system shown ‘below and the transfer function Mie Se eG Re) 1+ KG) HI) “The response ofthe system depends mainly fon the closed loop pole ofthe system ‘Those poles locations depend onthe value of k aS sre om 0< Kew ‘Thus we need to study the locations ofthe Closed loop ples as k varies fom 0 1 ‘Tensclectastble ange fOr pe 6@}-" iss ~~ ei ° characters 7The ROOT LOCUS Problem EXAMPLE ke s(s? +6s +25) 1. Poles of the open-loop system (k= 0) are s=0 ,-3+j4 2.n=3, m=0 = ice. there are 3 branches ending at infinity G(s) H(s) = 3. Asymptotes center and angles are 0-3-3 3 4. There are no breakaway or break-in points, =2 ©, = 60, 180, 300 §. The j@ axis crossings occur at s = +j5 when k= 150 6. The angles of departure are calculated as 6,,=—36.8 and 0,)= 36.8The ROOT LOCUS Problem EXAMPLE (continued) s=0 ,-34j4 0-3-3 oF —_ -2 a= 60, 180,300 jo axis crossings at s = j5 when k=150 The angles of departure are @4,=~ 36.8 and @,)= 36.8The ROOT LOCUS Problem Design Problem The control system of the previous example is needed to satisfy the following specifications for a ramp input R(s) = 1/s? : (a) steady state error to be < 10% of the input magnitude. (b) damping ratio of the dominant roots to be > 0.707. (c) settling time to be <3 sec. G(s) RG) =@22 + Y(s) H(s)The ROOT LOCUS Problem Design Problem Solution (a) E=R-Y =R-EG = F(s)--*_- Gy yk 5 (s*-+65 +25) 1 i > 5 = sE()=$ 8, = 1 = S12 Le 3 (57 +65 +25) +4 5(s? +65 +25) SSF 65425 . _28 Se, =limsk (s)=F =k 2250 fore, <01The ROOT LOCUS Problem Design Problem So/ution .. continued (b) For & > 0.707, the angle of the vector from the origin to the dominant poles locations of the closed-loop system should be > 45° (©) T,=4/20,=4/0 <3 seo —+ o> 4/3 SotaThe ROOT LOCUS Problem ROOT-LOCUS PLOTS IN MATLAB Use one of the following MATLAB commands to generate root-locus plots. rlocus(num,den) rlocus(num,den,K) rlocus(A,B,C,D,K) Note: num and den are the numerator and denominator polynomials of the transfer function. A, B, C, and D are state-variable representation matrices. K is the user- supplied gain vector. IfK is not given, the gain vector is automatically determined.The ROOT LOCUS Problem ROOT-LOCUS PLOTS WITH MATLAB EXAMPLE K(s +844) SOHO) = s+ 46+ 4145+) num={1 2 4]: den=conv({1 0],conv({1 4], conv([1 6},[1 1.4 1); rlocus(num,den) axis({-10 10 -10 10])