2E1262

2006

2E1262 Nonlinear Control
Automatic Control

• Disposition
5 credits, lp 2 28h lectures, 28h exercises, 3 home-works

• Instructors
Bo Wahlberg, professor,

bo@ee.kth.se
Krister Jacobsson, teaching assistant,

krister.jacobsson@ee.kth.se
STEX (entrance floor in S3 building), administration,

stex@s3.kth.se
Lecture 1 1

2E1262

2006

Course Goal
To provide participants with a solid theoretical foundation of nonlinear control systems combined with a good engineering understanding You should after the course be able to

• understand common nonlinear control phenomena • apply the most powerful nonlinear analysis methods • use some practical nonlinear design methods

Lecture 1

2

2E1262

2006

2E1262 Nonlinear Control
Lecture 1

• Practical information • Course outline • Nonlinear control phenomena • Nonlinear differential equations

Lecture 1

3

relay.2E1262 2006 Today’s Goal You should be able to • Recognize some phenomena in nonlinear systems • Transform differential equation to first-order form • Mathematically describe saturation. back-lash • Derive equilibrium points Lecture 1 4 . dead-zone.

s3.kth.se/control/kurser/2E1262 • Get an E computer account for the computer exercise (see STEX) • Homeworks are mandatory and have to be handed in on time Lecture 1 5 .2E1262 2006 Course Information • All info and handouts are available at http://www.

Nonlinear Systems. Only references to Khalil will be given. Nonlinear Systems: Analysis. Slotine & Li. Nonlinear Systems Analysis. Prentice Hall. Glad & ¨ Ljung. flervariabla och olinjara metoder. Stability and Control. Reglerteori.2E1262 2006 Material • Textbook: Khalil. • Lecture notes: Copies of transparencies • Exercises: Class room and home exercises • Homeworks: 3 computer exercises to hand in • Software: Matlab (provided by KTH Library) Alternative textbooks (decreasing mathematical brilliance): Sastry. Lecture 1 6 . 3rd ed. Vidyasagar.. Two course compendia sold by STEX. 2002. Applied Nonlinear Control. Optional but highly recommended.

stability theory. describing function (L3-L6) • Control design: compensation. computer simulation (L1-L2) • Feedback analysis: linearization. fuzzy control (L11-L13) • Summary (L14) Lecture 1 7 . optimal control. high-gain design. neural networks.2E1262 2006 Course Outline • Introduction: nonlinear models. Lyapunov methods (L7-L10) • Alternatives: gain scheduling.

y(t) = Cx(t). v ∈ M and α ∈ R : M → M is S(αu) = αS(u) S(u + v) = S(u) + S(v) Example: Linear time-invariant systems scaling superposition x(t) = Ax(t) + Bu(t). x(0) = 0 ˙ y(t) = g(t) u(t) = Y (s) = G(s)U (s) t 0 g(τ )u(t − τ )dτ Notice the importance to have zero initial conditions Lecture 1 8 . The system S linear if for all u.2E1262 2006 Linear Systems Definition: Let M be a signal space.

2E1262 2006 Linear Systems Have Nice Properties Local stability=global stability Stability if all eigenvalues of A (or poles of G(s)) are in the left half-plane Superposition Enough to know a step (or impulse) response Frequency analysis possible Sinusoidal inputs give sinusoidal outputs: Y (iω) = G(iω)U (iω) Lecture 1 9 .

2E1262 2006 Linear Models are not Enough Example:Positioning of a ball on a beam φ x Nonlinear model: m¨(t) = mg sin φ(t). Linear model: x(t) = gφ(t) x ¨ Lecture 1 10 .

Possibly also include other nonlinearities such as centripetal force. Lecture 1 11 .1 φ0 ≈ = 2 rad = 114◦ 0. saturation.2E1262 2006 Can the ball move 0. Clearly outside linear region! Linear model valid only if sin φ ≈φ Must consider nonlinear model.1 meter in 0.05φ0 2 so that 0. friction etc.05 Unrealistic answer.1 seconds from steady state? Linear model (step response with φ = φ0 ) gives t2 x(t) ≈ 10 φ0 ≈ 0.

2E1262 2006 2 minute exercise: Find a simple system x = f (x. u) that is stable ˙ for a small input step but unstable for large input steps. Lecture 1 12 .

2E1262 2006 Stability Can Depend on Reference Signal Example: Control system with valve characteristic f (u) Process Valve Motor y r 1 1 = u2 Σ s (s+1)2 −1 Simulink block diagram: 1 s u2 1 (s+1)(s+1) y −1 Lecture 1 13 .

2E1262 2006 S TEP R ESPONSES 0.72 Output y 2 0 0 10 Output y 5 0 0 5 10 15 Time t 20 25 30 5 10 15 Time t 20 25 30 Stability depends on amplitude of the reference signal! (The linearized gain of the valve increases with increasing amplitude) Lecture 1 14 .2 0 0 4 5 10 15 Time t 20 25 30 r = 1.2 Output y 0.4 r = 0.68 r = 1.

2E1262 2006 Stable Periodic Solutions Example: Position control of motor with back-lash 0 Constant Sum 5 P−controller 1 5s 2+s Motor Backlash y −1 Motor: G(s) = 1 s(1+5s) Controller: K=5 Lecture 1 15 .

2E1262 2006 Back-lash induces an oscillation Frequency and amplitude independent of initial conditions: 0.5 Output y 0 −0.5 0 0.5 Output y 0 −0.5 Output y 0 −0.5 0 0.5 0 10 20 Time t 30 40 50 10 20 Time t 30 40 50 10 20 Time t 30 40 50 How predict and avoid oscillations? Lecture 1 16 .

2E1262 2006 Automatic Tuning of PID Controllers Relay induces a desired oscillation whose frequency and amplitude are used to choose PID parameters PID Σ Relay A T u Process y −1 1 y u 0 −1 0 5 10 Time Lecture 1 17 .

2E1262 2006 Harmonic Distortion Example: Sinusoidal response of saturation a sin t Saturation y(t) = ∞ k=1 Ak sin(kt) Amplitude y 10 0 a=1 0 2 1 0 10 −2 −1 −2 0 2 1 0 1 2 3 Time t 4 5 10 Frequency (Hz) Amplitude y 10 0 a=2 0 10 −2 −1 −2 0 1 2 3 Time t 4 5 10 Frequency (Hz) Lecture 1 18 .

switched electronics.2E1262 2006 Example: Electrical power distribution Nonlinearities such as rectifiers. and transformers give rise to harmonic distortion Total Harmonic Distortion = ∞ k=2 Energy in tone k Energy in tone 1 Example: Electrical amplifiers Effective amplifiers work in nonlinear region Introduces spectrum leakage. which is a problem in cellular systems Trade-off between effectivity and linearity Lecture 1 19 .

5 0 1 5 10 Time t 15 20 25 30 a sin ωt 0.2E1262 2006 Subharmonics Example: Duffing’s equation y + y ¨ ˙ 0.5 −1 0 5 10 15 20 25 30 Time t Lecture 1 20 .5 + y − y 3 = a sin(ωt) 0 y −0.5 0 −0.

gives x(t) = exp(At)x0 ˙ Lecture 1 21 . x(0) = x0 . x(0) = x0 ˙ over an interval [0.2E1262 2006 Nonlinear Differential Equations Definition: A solution to x(t) = f (x(t)). T ] → Rn such that (1) • When does there exists a solution? • When is the solution unique? Example: x = Ax. (1) : [0. T ] is a C1 function x is fulfilled.

2E1262 2006 Existence Problems Example: The differential equation x ˙ = x2 . 1 − x0 t 1 Solution not defined for tf = x0 Solution interval depends on initial condition! dx Recall the trick: x = x ⇒ ˙ = dt 2 x −1 −1 x0 Integrate ⇒ − = t ⇒ x(t) = x(t) x(0) 1 − x0 t 2 Lecture 1 22 . x(0) = x0 1 0≤t< x0 x0 has solution x(t) = .

2E1262 2006 Finite Escape Time Simulation for various initial conditions x0 Finite escape time of dx/dt = x2 5 4.5 0 0 1 2 3 4 5 Time t Lecture 1 23 .5 3 x(t) 2.5 2 1.5 1 0.5 4 3.

has many solutions: (t − C)2 /4 t > C 0 t≤C x(t) = 2 1. x(0) = 0.2E1262 2006 Uniqueness Problems Example: x= ˙ √ x.5 1 x(t) 0.5 0 −0.5 −1 0 1 2 Time t 3 4 5 Lecture 1 24 .

i. the water tank lab process with √ x = − x. Hint: Reverse time s = T − t ⇒ ds = −dt and thus dx dx =− ds dt Lecture 1 25 .2E1262 2006 Physical Interpretation Consider the reverse example..e. It is then impossible to know at what time t < T the level was x(t) = x0 > 0. ˙ x(T ) = 0 where x is the water level.

f (x) − f (y) ≤ L x − y Euclidean norm is given by Slope L x 2 = x2 + · · · + x 2 1 n Lecture 1 26 . y ∈ Br (x0 ) = {z ∈ Rn : z − x0 < r}.2E1262 2006 Lipschitz Continuity Definition: f : Rn → Rn is Lipschitz continuous if there exists L. r > 0 such that for all x.

2E1262 2006 Local Existence and Uniqueness Theorem: If f is Lipschitz continuous. δ]. Based on the contraction mapping theorem Remarks • δ = δ(r.1. tank example) • f being C 1 implies Lipschitz continuity (L = maxx∈Br (x0 ) f (x)) Lecture 1 27 . L) • f being C 0 is not sufficient (cf. Proof: See Khalil. Appendix C.. ˙ x(0) = x0 has a unique solution in Br (x0 ) over [0. then there exists δ > 0 such that x(t) = f (x(t)).

x. ˙ y = h(x) y = h(x) x = f (x) + g(x)u. u. output y General: Explicit: Affine in u: Linear: f (x.) = 0 ˙ ˙ ˙ x = f (x. y. input u. . . ˙ x = Ax + Bu. y.2E1262 2006 State-Space Models State x. . ˙ y = Cx Lecture 1 28 . u. u).

t) ˙ is always possible to transform to an autonomous system by introducing xn+1 = t: x = f (x.2E1262 2006 Transformation to Autonomous System A nonautonomous system x = f (x. xn+1 ) ˙ xn+1 = 1 ˙ Lecture 1 29 .

..2E1262 2006 Transformation to First-Order System Given a differential equation in y with highest derivative express the equation in x Example: Pendulum = y dy dt . dtn ¨ ˙ M R2 θ + k θ + M gR sin θ = 0 ˙ x= θ θ T gives x1 = x2 ˙ k g x2 = − ˙ x − sin x1 2 2 MR R Lecture 1 30 . dn−1 y dtn−1 T dn y .

y ∗ = Cx∗ Often the equilibrium is defined only through the state x∗ Lecture 1 31 . 0. u∗ . u∗ ). y ∗ = h(x∗ ) 0 = Ax∗ + Bu∗ .2E1262 2006 Equilibria Definition: A point (x∗ . if a solution starting in (x∗ . y ∗ . . 0. y ∗ ) is an equilibrium. Corresponds to putting all derivatives to zero: General: Explicit: Affine in u: Linear: f (x∗ . . . u∗ . u∗ . y ∗ = h(x∗ ) 0 = f (x∗ ) + g(x∗ )u∗ .) = 0 0 = f (x∗ . y ∗ ) stays there forever.

2E1262 2006 Multiple Equilibria Example: Pendulum ¨ ˙ M R2 θ + k θ + M gR sin θ = 0 ¨ ˙ θ = θ = 0 gives sin θ = 0 and thus θ∗ = kπ Alternatively in first-order form: x1 = x2 ˙ k g x2 = − ˙ x − sin x1 2 2 MR R x1 = x2 = 0 gives x∗ = 0 and sin(x∗ ) = 0 ˙ ˙ 2 1 Lecture 1 32 .

2E1262 2006 Some Static and Dynamic Nonlinearities |u| Abs u e Math Function Saturation Sign Dead Zone Look−Up Table Relay Backlash Coulomb & Viscous Friction Lecture 1 33 .

a system that limit the rate of change of a signal) using one of the previous models − ? 1 s Hint: Try a saturation! Lecture 1 34 .e.2E1262 2006 2 minute exercise: Construct a rate limiter (i..

2E1262 2006 Next Lecture • Simulation in Matlab • Linearization Lecture 1 35 .

Sign up to vote on this title
UsefulNot useful