Arba Minch University Arba Minch Institute of Technology Faculty of Electrical and Computer Engineering Modern Control System ECEE-4331 Chapter 5.Design Via State SpaceIntroduction + State-space techniques can be applied to a wider class of systems than transform methods, for example, a systems with nonlinearities, MIMO systems + Frequency domain methods of design, cannot be used to specify all closed loop poles of the higher order system + State-space technique allow to place all poles of the closed-loop system. + State-space methods do not allow the specification of closed-loop zero locations + Frequency domain methods do allow through placement of the lead compensator zero. + State-space design may prove to be very sensitive to parameter changes.Controller De: An nth order feedback control system has the following characteristic equation SHG, ST eee +as +a, =0 Pole Placement » Geet ep A plant shown in (a) is 4 represented in state space by r xX = Ax +Bu @ y =Cx rs ly e+] x cL! The state equation for ‘ the closed loop system “ in (b) can be written as -K . () X= Ax+Bu=AX+B(-Kx+r)=(A-BK)x+Br 4g. State-space representation of a plant: pax b. plant with state-feedbackController Design X= Ax +Bu = Ax +B( -Kx+r)4{ A-BK) x+Br y=Cx a. Phase-variable representation for plant: b. plant with state-variable feedbackPole Placement for Plants in Phase-Variable Form 1. Represent the plant in phase-variable from. 2. Feedback each phase variable to the input of the plant through a gain k; Find the char. Equation for the closed-loop system represented in step 2. 4. Decide upon all closed-loop pole locations and determine an equivalent char. Equation. 5. Equate like coefficients of the char. Equations from step 3 & 4 and solve for k;. 1. Following these steps, the phase-variable representation for the plant is a o 1 0 07 0 0 0 1 0 0 A= : : bB=loc=[e ¢ c.] ~a, -a, -a, Wa, , 1 The char. Equationis 5" +a, s"" +....+a,8 ta, =0Pole Placement for Plants in Phase-Variable Form 2. The system matrix for the closed-loop system is 0 1 0 ea 0 0 1 tar 0 A-BK = : : : i : —(ay+k,) —(a,+k,) -(a,+k3) .. a, +k,) 3. And the char. Equation by inspection is det(s! -(A-BK))=s" +(a,_,+k,)s"' +(4,5 +h, .)s">+...+ (a, +k,)s + (a, +k) =0 n-| 4. Assume the desired char. Equation for proper pole placement is s"+d,s"' +d, 8"° +....+d,s° +d,s+d, =0 5. The feedback coefficients can be found as Ki =d, —a,Controller design for phase-variable Example 5.1:Design the phase-variable feedback gains to yield 9.5% overshoot and settling time of 0.74 second. Gy = 208+) s(s +1)(s +4) Solution: Calculating the desired closed-loop char. Equation, the closed -loop poles are -5.4+/-j7.2. we select the third pole at -5.1 close to the zero.Example 5.1 The closed-loop system's state equation from previous figure is . 0 1 0 0 x=| 0 0 1 jx+fOfrs y =[100 20 ole -k, -(4+k,) -(5+k,) 1 The closed loop system matrix is 0 1 0 A-BK =| 0 0 1 -k, “4+k,) -G+k,) The closed loop char. Equation is det(sI —(4 —BK)) =s*° +(5+k,)s> +(44+k,)s +k, =0 This equation must match the desired char, Equation 5? +15,9s7 +136.08s +413.1=0 We obtain: k,=413.1, k,=132.08, k,=10.9Example 5.1 Substitute the values we obtain: 0 1 0 0 x=] 0 0 1 |x+fOlr; y» =[100 20 0]x 413.1 -136.08 -15.9 The transfer function is T (5) = +5) _ 5° +15.95° +136.08s +413.1 The simulation shows 11.5% overshoot and a settling time of 0.8. A redesign with 3°¢ pole canceling the zero af -5 will yield performance equal to the requirements. 020 oas 0s 0 05 10 Time (seconds)Controllability If an input to a system can be found that takes every state variable from desired initial state to a desired final state, the system is said to be controllable; otherwise, the system is uncontrollable. Pole placement is viable design technique only for systems that are controllable Controllability by inspection A system with distinct eigenvalues and diagonal system matrix is controllable if the input coupling matrix B does not have any rows that are zero. -a, 0 0] [0 x=| 0 -a, 0 |xtft fu 3 SH 1 -4 0 0 1 x=| 0 -a, 0 |x+|i fu JU CS "a. controllable @ b. uncontrollable systems ®Controllability matrix An nth-order plant whose state equationis x» = 4x +Bu Is completely controllable if the matrix c,, =[2 AB AB... A™B | Is of rank n, where Cy is called controllability matrix. Example 5.2 Given the system in SS, determine its controllability -1 1 0° fo x=]0 -1 0 x41 lu 002 1 Solution: the state equation for the system is The controllability matrix is 0 Cy =[B AB en 1 The rank of Cy is 3, so The system is controllableController design by matching coefficients Pole Placement for Plants NOT in Phase-Variable Form Example 5.3 Given a plant Y(s)/U(s)= 10/[(s + 1)(s + 2)], design state feedback to yield a 15% overshoot with settling time of 0.5 sec. Solution: writing the state equation for (b) we have. -f2 1 ol. no oy | "Oe y Tl en tL le oT The characteristic equation is s? +(S+k,)s + (242k, +k) =0 to Hi 1 Fo ’ O- O Compare it to the transient z YD CF response requirements 8? +16s + 239.5 =0 We get k2=13 and k,=211.5 * (6) a. Signal-flow graph in cascade form for G(s ) =10/[(s + 1)(s + 2): b. system with state feedback addedObserver Design If the sate variables are not available because of system configuration or cost, it is possible to estimate the states. Estimated states are then fed to the controller. An observer, sometimes called an estimator, is used to calculate state variables that are not accessible from the plant. Here the observer is a model of the plant. Assume a plant And an observer Subtract the observer from S A n the plant, we obtain x =Ax + Bu x =Ax+Bu “A A y =Cx x A x-x =A(x -x) y=Cx y-y =C(@e—x)| Observer Design State-feedback design using an observer to estimate Unavailable state variables: Plant Plant Pant [2 re} mm fF roi # Bsimicd 1-0} } Same SP } p= a. Open-loop observer; b. Closed-loop observer; ee es cater fo o Estimied Pat “8 TK c. Exploded view of a closed-loop observer, showing feedback Esimated arrangement to soa reduce state-variable estimation error To contoletObserver Design Writing the state equations of the observer from figure A a A Estimated Plant x =Ax+Bu+lL(y-y) , i eee a A so { Qe — y=Cx But the state equation for plant i A x =Ax +Bu Estimated ae y =x i out Subtract, we get Tocuntolr 0 (x=) = A(x-)-L(y- 9) y-y=C(x-3) Where x —x is the error between actual and estimated state vectorsObserver Design Substitute the output equation into the state equation, we get (e-x) =(A -LCY(x -2) yay =C(e-3) Lete, =x —x we have, e, =(4-LC)e, and y —y =Ce, Solving for L, the char. Equation is det[AJ —(4 —LC)]=0 For an nth-order plant, the char. Equation A-LC is Ss" +(a,,+h)s"' +(a,5+h)s"?+..4+(a, +1, 8 + (ay +1,) =0 This can be written by inspection if compared to char. Equation for plant, if the plant is represented in observer canonical form. n nl n2 det(s] -A)=s" +a, s"" +a,_.8"~ + +as +a, =0 We equate this to desired char. Equation, and solve for ||, n-2 s"+d,s"'+d, 8"? + .+d,s* +d,s +d, =0Observer design for observer canonical form Example 5.4: Design an observer for the plant. Which is represented in observer canonical form, the observer will respond 10 times faster than the controller loop designed the closed-loop poles at -1+/2 (s+4) s+4 Gs) = —_——_—_— = = GGT DETS SFR + ITs +10 Solution: the canonical form representation is first shown (a). Now form the difference between actual plant and observer outputs and add the feedback to the derivative of each state variable as shown in (b).Next find the char. Polynomial. The state equation for the estimated plant is and the observer error is . -8 1 0] fo . ~8+1,) 1 0 x =Ax+Bu=|-17 0 I|x+]1 Jue, =(4-LC)e, =|-(174+1,) 0 IIe, OO. ~(10+/,) 0 0 peCx=fl 0 o]x The char. Polynomial is det[Al -(A -LC)] =s° +(8+1,)s? +(17 +1,)s + (10 +/,) =0 To make the observer 10 times faster, we design the observer poles at We select the 3" pole at -100, the desired char. Is then -10+ 20 (s +100)(s* + 20s +500) =s* +120s* + 2500s + 50000 Compare and solve, we get |,=112, |.=2483, and 1,=49990Simulation showing response of observer for an input r(t) =100t: a. closed-loop; b. open-loop with observer gains disconnected ae * The response is slower in (b), since the observer is a copy of the plant with different initial conditions oO Ol 02 03 Time (seconds) @ On 02 03 Time (seconds) ©Observobility If the initial sate vector, x(to), can be found from u(t) and y(t) measured over a finite interval of time from to, the system is said to be observable; otherwise, it is said tg be unobservable. Observability is the ability to deduce the state variables from a knowledge of the input, u(t), and the output, y(t). By inspection, for diagonalized systems, (represented in parallel form) with distinct eigenvalues, if any column of the output matrix is zero the system is not observable. An nth order system is completely observable if the observability matrix Cc is of rank n. CA Oy = - 4 a. observable and CA" b. Unobservable systemsExamples Example 5.6 Determine if the system in figure is observable. Solution: the state and output equations are 1 . o 1 0 0 x=dx+Bu=|0 0 1 |x+lolu | 4-3 2 1 y=[0 5 Ix c oO 5 4 0, =| cA |=|4 3 3 ca™| |-12 -13 -9 4 Example 5.7 Determine if the system in figure is observable. Solution: the state and output equations are e 0 1 0 x =Ax+Bu= +] lee [’ ual Hi rb ae aH J ca} |-20 -16nt: Observer design by matching coeffi Example 5.8 Design an observer for the phase variable with transient response described by ¢=0.7 and «, =100 for the plant Go)= 407(s +0.916) _ 407s +372.81 (s +1.27)(s +2.69) 3.965 +3.42 Solution: first represent the system in phase-variable form as shown in figure (a). For the plant we have [ 0 1 | C =[372.81 407] 3.42 3.96 Calculating the observability matrix, 0,, =[C CA ] shows the plant is observable. Next find the characteristic equation, first we have, 0 1 1, A-LC = -|/'|[372.81 407] 3.42 3.96] [I , _ 372.811, a-407,) ] °° “| -G.42+372.81,) -G.96 +407/,)Observer design by matching coefficients- Example Now evaluate det(ar —(4 —LC)]=0 A+372.81), —(-407/,) det[A/ —(A—LC)] = det (3.42+372.81,) (A +3.96 +4071,) = A? +(3.96+372.81l, +4971, )A + (3.42 + 84.39/, +372.81/,) =0 From the problem statement we want ¢=0.7 and @, =100 Thus, 2° +140A + 10000 =0 comparing coefficients, we find I, and I, to be -38.397 and 35.506, respectively. Finally, we implement the observer as shown in figure (b) and using 0 1 “ A= ; C =[372.81 407]; o- > 0 —38.397 B= 3 L= " | 35.506 |Example: Design of Controller and Observer Using the simplified block diagram of the plant for the antenna azimuth position control system shown in Figure, design a controller to yield a 10% overshoot and a settling time of 1 second. Place the third pole 10 times as far from the imaginary axis as the second-order dominant pair. Assume that the state variables of the plant are not accessible and design an observer to estimate the states. The desired transient response for the observer is a 10% overshoot and a natural frequency 10 times as great as the system response above. As in the case of the controller, place the third pole 10 times as far from the imaginary axis as the observer's dominant second-order pair. U(s) = E(s) Y(s) = 4,(s) s+ 1.71)(s + 100)Example: Design of Controller and Observer SOLUTION: Controller Design: We first design the controller by finding the desired characteristic equation. A 10% overshoot and a settling time of 1 second yield € = 0.591 and wn= 6.77. Thus, the characteristic equation for the dominant poles is s2 + 8s + 45.8 = 0, where the dominant poles are located at -4 + j5.46. The third pole will be 10 times as far from the imaginary axis, or at - 40. Hence, the desired characteristic equation of the closed-loop system is:Example: Design of Controller and Observer (s? + 8s + 45.8)(5 + 40) = 5? + 485? + 365.85 + 1832 = 0 Next, we find the actual characteristic equation of the closed-loop system. The first step is to model the closed-loop system in state space and then find its characteristic equation. From Figure, the transfer function of the lant i ae oe 1325 _ 1325 8) = S(s+1.71(s+ 100) s(? + 101.71 + 171) Using phase variables, this > 7 : 5 1325 transfer function is converted “ to the signal-flow graph shown in Figure, and the state equations are o 1 0 0 written as follows: x=|0 0 1 |x+|0|u=Ax+Bu 0 -171 101.71 1 y= [1325 0 Ojx=CxExample: Design of Controller and Observer We now pause in our design to evaluate the controllability of the system. The controllability matrix, Cy, is 0 0 1 Cm=[B AB A’B] f 1 =101.71 1 -101.71 10,173.92 The determinant of Cy is - 1; thus, the system is controllable. Continuing with the design of the controller, we show the controller's configuration with the feedback from all state variables in Fiaure. We now find the characteristic equation of the system of Figure, the system matrix, A - BK, is 0 1 0 A-BK= | 0 0 1 ky —(I7L+k2) -(101.71 +s)Example: Design of Controller and Observer Thus, the closed-loop system's characteristic equation is det(sI — (A — BK)] = s° + (101.71 + ks)s? + (171 + k)s + ky =0 Matching the coefficients, we evaluate the k;'s as follows: K=1832, K=194.8, K=-53.71 Observer Design: Before designing the observer, we test the system for observability. Using the A and C matrices, the observability matrix, O, is c 1325 0 0 Om=|CA}=] 0 132 0 cA’ 0 0 1325 The determinant of Oy is 13253. Thus, Oy is of rank 3, and the system is observable. We now proceed to design the observer. Since the order of the system is not high, we will design the observer directly without first converting to observer canonical form. We need first to find A - LC.Example: Design of Controller and Observer -1325 1 0 A-LC= | -13252 0 1 -1325I; -171 -101.71 The characteristic equation for the observer is now evaluated as det[al — (A — LC)] = a3 + (13254 + 101.71)a? + (134,800/, + 1325/2 + 171) + (226,600/; + 134,800/. + 1325/3) =0 From the problem statement, the poles of the observer are to be placed to yield a 10% overshoot and a natural frequency 10 times that of the system's dominant pair of poles. Thus, the observer's dominant poles yield [s? + (2 x 0.591x 67.7)s + 67.77] = (s? + 80s + 4583). The real part of the roots of this polynomial is -40. The third pole is then placed 10 times farther from the imaginary axis at -400. The composite characteristic equation for the observer isExample: Design of Controller and Observer (s? + 80s + 4583)(s + 400) = s° + 480s? + 36,580s + 1,833,000 = 0 Matching coefficients, we solve for the observer gains: |,=0.286, 1,=-1.57, |=1494 plane Figure shows the completed design, including the controller and the observer