Flow Lab

Syllabus / Module 6: Laplace Transforms

Lesson 6.2

Solving Initial Value Problems

Transform the ODE, solve for Y(s), partial fractions, invert. Systems via Laplace.

Written byClaude Sonnet

🌊
Why this matters for simulation: Linearized aircraft dynamics are analyzed in the s-domain.

By the end of this lesson you can…

  • Run the pipeline transform, algebra, partial fractions, inverse on a second-order IVP with forcing, and read off $Y(s) = \dfrac{(as+b)y(0) + ay'(0)}{P(s)} + \dfrac{G(s)}{P(s)}$.
  • Split a solution into its zero-input response (initial data only) and zero-state response (forcing only), and say which term of $Y$ each comes from.
  • Invert $Y(s)$ for distinct real poles, a repeated pole (a factor $t$, via the shift theorem or $\mathcal L\{tf\} = -F'$) and a complex pair (complete the square, shift the sine and cosine).
  • Read a pole plot: pole location gives the mode, so a real pole is an exponential, a pair off the axis is an oscillation, and a double pole carries a factor $t$.
  • Solve a $2\times2$ first-order system by Laplace algebra and show that its poles are the eigenvalues.
  • Compare Laplace with the characteristic-equation and undetermined-coefficients method, and say when each is the better tool.
On this page
·
  1. The pipeline, with forcing included
  2. What the split is good for
  3. Repeated poles and the factor ttt
  4. Complex poles: completing the square
  5. Pole location is mode
  6. First-order systems by the same algebra
  7. Free sanity checks: initial and final values
  8. Laplace against the characteristic equation
  9. Computing it
  10. Concept checks
  11. Problems
  12. Key Takeaways

In 6.1 The Laplace Transform we built the table and the derivative rule: L{y′′}=s2Y−sy(0)−y′(0)\mathcal L\{y''\} = s^2Y - sy(0) - y'(0). This lesson cashes them in. A simulation engineer meets IVPs with forcing all the time (a spring-mass damper under a push, a control surface under a command), and the method of 5.2 Constant Coefficients & the Characteristic Equation handles them in three stages: solve the homogeneous equation, guess a particular solution, then fix constants from the initial data. Laplace does everything in one stroke. Transform the equation, solve for Y(s)Y(s) by algebra, split it into table entries, invert. The initial data and the forcing both sit inside YY from the start. By the end you will see why 5.6 Forced Oscillations & Resonance had to treat resonance as a special case, and why here it arrives for free as a double pole.

▶ Video The pipeline on one IVP, the three kinds of poles and their time terms, the zero-input and zero-state split, a 2x2 system, and Laplace against the characteristic-equation methodcaptions only
Transcript
  1. Last lesson gave us the table and the derivative rule. Now we cash them in: transform, solve by algebra, split into partial fractions, invert. Data and forcing are inside Y from the start.
  2. Transforming gives P of s times Y equals G of s plus a s plus b times y of 0, plus a y prime of 0. Dividing by P, the initial-data piece is the zero-input response and the forcing piece is the zero-state response. Add them.
  3. Take y double prime plus 3 y prime plus 2 y equals 4, y of 0 equal 1, y prime of 0 zero. Solving gives s squared plus 3 s plus 4, over s times s plus 1 times s plus 2. Poles: 0 from the forcing, minus 1 and minus 2 from P.
  4. Cover up each factor: the rest gives 2 at s equals 0, minus 2 at minus 1, and 1 at minus 2. Each pole gives one term, in the same color on the pole plot: 2, minus 2 e to the minus t, e to the minus 2 t. Check: y of 0 is 1, y prime of 0 is 0.
  5. Zero-input response: 2 e to the minus t minus e to the minus 2 t, 0.60042 at t equals 1. Zero-state response: 2 minus 4 e to the minus t plus 2 e to the minus 2 t, 0.79915. Together, 1.39958. Only the second has the forcing's pole at 0, and its term 2 is where it settles.
  6. Now y double prime plus 2 y prime plus y equals 1, y of 0 equal 2. P is s plus 1 squared: a double pole. Minus F prime multiplies f by t, so 1 over s plus 1 squared inverts to t e to the minus t. At t equals 1: 1, 0.36788 and 0.36788, so 1.73576.
  7. Now y double prime plus 2 y prime plus 5 y equals 5, y of 0 zero, y prime of 0 equal 2. P is s plus 1 squared plus 4. Complete the square, writing s as s plus 1 minus 1: y is 1 minus e to the minus t cosine 2 t plus half e to the minus t sine 2 t. Poles at minus 1 plus or minus 2 i; y of 1 is 1.32035.
  8. One dictionary covers every example. A pole at sigma plus i omega, multiplicity m, gives t to the k, e to the sigma t, times cosine or sine of omega t. Sigma sets growth or decay, omega the frequency, multiplicity the power of t.
  9. Resonance: y double prime plus 4 y equals cosine 2 t. Y is s over s squared plus 4, squared: the forcing's poles at plus and minus 2 i land on the system's own, so each is double. So y is one quarter t sine 2 t: the factor t grows the envelope. At t equals 1, 0.22732.
  10. Forcing on a pole: the right side is e to the minus t. Its pole at minus 1 is a system pole, so it is double, and y is t minus 1 times e to the minus t, plus e to the minus 2 t. At t equals 1 the first two cancel: e to the minus 2, 0.13534.
  11. For a system, s I minus A, times X, equals x of 0. The determinant is s plus 1 times s plus 3, so X 1 and X 2 are halves over those poles. That determinant is the characteristic polynomial: the poles are the eigenvalues, minus 1 and minus 3. At t equals 1, 0.208833 and 0.159046.
  12. Example A the other way: roots minus 1 and minus 2, a constant guess of 2, and the data give c 1 equals minus 2, c 2 equals 1. Laplace wins when data and forcing are the whole problem; the characteristic equation wins on homogeneous ones. Neither handles tan t or nonlinear equations.
  13. On all five problems, the closed form and RK45 agree. Poles say which modes appear; the numerator says how much of each. Next: step functions, for forcing that switches on and off.

The pipeline, with forcing included

Take ay′′+by′+cy=g(t)ay'' + by' + cy = g(t) with data y(0)y(0), y′(0)y'(0), and let G(s)=L{g}G(s) = \mathcal L\{g\} and P(s)=as2+bs+cP(s) = as^2 + bs + c. Transforming each term with the derivative rule gives P(s) Y(s)=G(s)+(as+b) y(0)+a y′(0)P(s)\,Y(s) = G(s) + (as + b)\,y(0) + a\,y'(0), so

Y(s)=(as+b) y(0)+a y′(0)P(s)⏟zero-input+G(s)P(s)⏟zero-state.Y(s) = \underbrace{\frac{(as+b)\,y(0) + a\,y'(0)}{P(s)}}_{\text{zero-input}} + \underbrace{\frac{G(s)}{P(s)}}_{\text{zero-state}} .

The first piece is what the system does with g=0g = 0 and only its initial state to drive it (the zero-input response). The second is what the forcing does starting from rest, y(0)=y′(0)=0y(0) = y'(0) = 0 (the zero-state response). Linearity means the true answer is their sum, and the algebra separates them without being asked. The factor 1/P(s)1/P(s) multiplying the input is the transfer function, which 6.5 Convolution & Transfer Functions studies in earnest.

Example A (distinct real poles). y′′+3y′+2y=4y'' + 3y' + 2y = 4, y(0)=1y(0) = 1, y′(0)=0y'(0) = 0. Here G=4/sG = 4/s, so

  1. Transform: (s2+3s+2)Y−(s+3)=4/s(s^2 + 3s + 2)Y - (s + 3) = 4/s.
  2. Solve: Y=s+3(s+1)(s+2)+4s(s+1)(s+2)=s2+3s+4s(s+1)(s+2)Y = \dfrac{s+3}{(s+1)(s+2)} + \dfrac{4}{s(s+1)(s+2)} = \dfrac{s^2 + 3s + 4}{s(s+1)(s+2)}.
  3. Partial fractions, by cover-up at each pole: at s=0s = 0 the rest gives 4/(1⋅2)=24/(1\cdot2) = 2; at s=−1s = -1, (1−3+4)/((−1)(1))=−2(1 - 3 + 4)/((-1)(1)) = -2; at s=−2s = -2, (4−6+4)/((−2)(−1))=1(4 - 6 + 4)/((-2)(-1)) = 1. So Y=2s−2s+1+1s+2Y = \dfrac2s - \dfrac{2}{s+1} + \dfrac{1}{s+2}.
  4. Invert term by term: y(t)=2−2e−t+e−2ty(t) = 2 - 2e^{-t} + e^{-2t}.

Check the data: y(0)=2−2+1=1y(0) = 2 - 2 + 1 = 1 and y′(0)=2−2=0y'(0) = 2 - 2 = 0. The split is also visible in the algebra: the first fraction inverts to 2e−t−e−2t2e^{-t} - e^{-2t} (zero-input) and the second to 2−4e−t+2e−2t2 - 4e^{-t} + 2e^{-2t} (zero-state).

The lab runs the pipeline for five IVPs, one for each situation in this lesson. For each it assembles Y(s)Y(s) from the coefficients, lists the partial fractions and time terms, draws the poles of YY in the ss-plane in the same colour as the time terms they produce, and compares the closed form against an RK4 solution of the ODE itself. Press Next step to walk the five stages.

◆ Interactive Step through the pipeline, then connect each pole to its dashed time term

Loading simulation…

For Example A, set the probe time t∗=1t^* = 1. The readout gives y(1)=1.39958y(1) = 1.39958, made of 0.600420.60042 (zero-input) plus 0.799150.79915 (zero-state), which add to 1.399581.39958 before rounding. The cross-check at s0=1s_0 = 1 gives Y(1)=4/3=1.33333Y(1) = 4/3 = 1.33333 from the rational function and from the three time terms, and 1.333341.33334 from integrating the RK4 curve against e−te^{-t}, the definition from the previous lesson. So the algebra, the inverse and the definition agree, and the closed form sits within 3×10−113\times10^{-11} of RK4.

What the split is good for

Tick the lab’s split box and the two RK4 curves appear: the zero-input response starts at y(0)y(0) and the zero-state response starts at 00. They answer different questions. Double the forcing and only the zero-state part doubles; change the initial data and only the zero-input part moves. The zero-input part has only the poles that are roots of PP; the zero-state part has those and also the forcing’s own poles, because it carries the extra denominator of G(s)G(s). That is why the transient modes, set by PP, are the same whatever you push the system with, while the forced modes, set by G(s)G(s), show up as extra poles. In Example A the pole at s=0s = 0 comes entirely from the constant forcing, and its term 22 is the level the solution settles at, while the poles at −1-1 and −2-2 belong to PP and die out. A time stepper hands you one curve; the transform hands you the decomposition.

Repeated poles and the factor tt

Example B. y′′+2y′+y=1y'' + 2y' + y = 1, y(0)=2y(0) = 2, y′(0)=0y'(0) = 0. Now P(s)=(s+1)2P(s) = (s+1)^2 and

Y=2(s+2)(s+1)2+1s(s+1)2=2s2+4s+1s(s+1)2=1s+1s+1+1(s+1)2.Y = \frac{2(s+2)}{(s+1)^2} + \frac{1}{s(s+1)^2} = \frac{2s^2 + 4s + 1}{s(s+1)^2} = \frac1s + \frac{1}{s+1} + \frac{1}{(s+1)^2} .

The cover-up rule finds A=1A = 1 at s=0s = 0 and the coefficient of 1/(s+1)21/(s+1)^2 (11) at the double pole; the remaining coefficient follows from matching the large-ss behaviour, A+B=2A + B = 2, so B=1B = 1.

The new entry is L−1{1/(s+1)2}\mathcal L^{-1}\{1/(s+1)^2\}. Two routes reach it. The first shift theorem applied to L{t}=1/s2\mathcal L\{t\} = 1/s^2 gives L{te−t}=1/(s+1)2\mathcal L\{te^{-t}\} = 1/(s+1)^2. The second differentiates the definition with respect to ss:

F′(s)=dds∫0∞f(t)e−stdt=−∫0∞tf(t) e−stdt,soL{tf(t)}=−F′(s).F'(s) = \frac{d}{ds}\int_0^\infty f(t)e^{-st}dt = -\int_0^\infty t f(t)\,e^{-st}dt, \qquad\text{so}\qquad \mathcal L\{t f(t)\} = -F'(s).

This second route multiplies by tt for any ff, not only for exponentials. Either way y(t)=1+e−t+te−ty(t) = 1 + e^{-t} + te^{-t}, and at t∗=1t^* = 1 the lab’s three terms read 11, 0.367880.36788 and 0.367880.36788, summing to y(1)=1.73576y(1) = 1.73576 (zero-input 1.471521.47152 plus zero-state 0.264240.26424). The mechanism is the same as in the characteristic-equation rule: a pole counted twice produces a term multiplied by one extra power of tt.

Complex poles: completing the square

Example C. y′′+2y′+5y=5y'' + 2y' + 5y = 5, y(0)=0y(0) = 0, y′(0)=2y'(0) = 2. Here P(s)=s2+2s+5=(s+1)2+4P(s) = s^2 + 2s + 5 = (s+1)^2 + 4 has no real roots, and

Y=2(s+1)2+4+5s ((s+1)2+4)=2s+5s ((s+1)2+4).Y = \frac{2}{(s+1)^2 + 4} + \frac{5}{s\,\bigl((s+1)^2+4\bigr)} = \frac{2s + 5}{s\,\bigl((s+1)^2+4\bigr)} .

Split off the real pole with the cover-up rule, A=5/5=1A = 5/5 = 1, then write the remainder over the quadratic: 2s+5=(s2+2s+5)+(Bs+C) s2s + 5 = (s^2 + 2s + 5) + (Bs + C)\,s. Matching s2s^2 gives B=−1B = -1 and matching ss gives C=0C = 0, so

Y=1s−s(s+1)2+4=1s−s+1(s+1)2+22+1(s+1)2+22.Y = \frac1s - \frac{s}{(s+1)^2+4} = \frac1s - \frac{s+1}{(s+1)^2+2^2} + \frac{1}{(s+1)^2+2^2} .

The trick is to rewrite the numerator ss as (s+1)−1(s+1) - 1 so that each piece matches a shifted table entry: L{e−tcos⁡2t}=(s+1)/((s+1)2+4)\mathcal L\{e^{-t}\cos2t\} = (s+1)/((s+1)^2+4) and L{e−tsin⁡2t}=2/((s+1)2+4)\mathcal L\{e^{-t}\sin2t\} = 2/((s+1)^2+4), which is why the last fraction needs a factor 12\tfrac12 out front. Hence

y(t)=1−e−tcos⁡2t+12e−tsin⁡2t.y(t) = 1 - e^{-t}\cos2t + \tfrac12e^{-t}\sin2t .

Data check: y(0)=1−1=0y(0) = 1 - 1 = 0. For the slope, ddt(−e−tcos⁡2t)=1\frac{d}{dt}(-e^{-t}\cos2t) = 1 at 00 and ddt(12e−tsin⁡2t)=1\frac{d}{dt}(\tfrac12e^{-t}\sin2t) = 1 at 00, total 22. The lab at t∗=1t^* = 1 gives y(1)=1.32035y(1) = 1.32035, of which the zero-input part is e−1sin⁡2=0.33451e^{-1}\sin2 = 0.33451 and the zero-state part 0.985840.98584.

Pole location is mode

Every example shows the same dictionary. A pole at p=σ+iωp = \sigma + i\omega of multiplicity mm contributes terms tkeσtcos⁡ωtt^{k}e^{\sigma t}\cos\omega t and tkeσtsin⁡ωtt^{k}e^{\sigma t}\sin\omega t for k=0,…,m−1k = 0,\dots,m-1 (for real pp, just tkeptt^ke^{pt}). The real part σ\sigma sets growth or decay, the imaginary part ω\omega sets the oscillation frequency, and the multiplicity sets the power of tt. A pole in the left half-plane decays, one on the imaginary axis neither decays nor grows, and one in the right half-plane grows exponentially, because eσte^{\sigma t} does. The pole plot in the lab shows the first three: in Example C the pair at −1±2i-1 \pm 2i sits left of the dashed axis, matching the envelope e−te^{-t} and the frequency 22.

Two more presets use the same machinery on cases where the characteristic-equation method needs a special rule.

Resonance (preset 4). y′′+4y=cos⁡2ty'' + 4y = \cos2t from rest. Then Y=s/(s2+4)2Y = s/(s^2+4)^2: the forcing’s poles ±2i\pm2i land exactly on the system’s own poles, so each is double. Using L{tf}=−F′\mathcal L\{tf\} = -F' with f=sin⁡2tf = \sin 2t, F=2/(s2+4)F = 2/(s^2+4),

L{tsin⁡2t}=−dds2s2+4=4s(s2+4)2,soy(t)=14 tsin⁡2t.\mathcal L\{t\sin2t\} = -\frac{d}{ds}\frac{2}{s^2+4} = \frac{4s}{(s^2+4)^2}, \qquad\text{so}\qquad y(t) = \tfrac14\,t\sin2t .

This is the growing oscillation of 5.6 Forced Oscillations & Resonance: the envelope t/4t/4 is the factor tt from the double pole. No guessing rule is needed. At t∗=1t^* = 1 the lab gives y(1)=sin⁡(2)/4=0.22732y(1) = \sin(2)/4 = 0.22732.

Forcing on a pole (preset 5). y′′+3y′+2y=e−ty'' + 3y' + 2y = e^{-t} from rest. The forcing pole at −1-1 coincides with a system pole, so Y=1/((s+1)2(s+2))=−1s+1+1(s+1)2+1s+2Y = 1/((s+1)^2(s+2)) = -\frac{1}{s+1} + \frac{1}{(s+1)^2} + \frac{1}{s+2} and y=(t−1)e−t+e−2ty = (t - 1)e^{-t} + e^{-2t}. At t∗=1t^* = 1 the first two terms cancel and y(1)=e−2=0.13534y(1) = e^{-2} = 0.13534, which the lab prints.

First-order systems by the same algebra

A system x′=Ax\mathbf x' = A\mathbf x transforms the same way. With X(s)=L{x}\mathbf X(s) = \mathcal L\{\mathbf x\} the derivative rule gives sX−x(0)=AXs\mathbf X - \mathbf x(0) = A\mathbf X, so (sI−A)X=x(0)(sI - A)\mathbf X = \mathbf x(0), and the unknown is found by solving a small linear system whose entries are polynomials in ss. (A second-order ODE is the special case with x1=yx_1 = y and x2=y′x_2 = y'.)

Take A=(−211−2)A = \begin{pmatrix}-2 & 1\\ 1 & -2\end{pmatrix} and x(0)=(1,0)\mathbf x(0) = (1, 0). Then sI−A=(s+2−1−1s+2)sI - A = \begin{pmatrix}s+2 & -1\\ -1 & s+2\end{pmatrix} with determinant (s+2)2−1=(s+1)(s+3)(s+2)^2 - 1 = (s+1)(s+3). For a 2×22\times2 matrix the inverse is the swapped-and-negated matrix over the determinant, so

X=1(s+1)(s+3)(s+211s+2)(10),X1=12s+1+12s+3,X2=12s+1−12s+3.\mathbf X = \frac{1}{(s+1)(s+3)}\begin{pmatrix}s+2 & 1\\ 1 & s+2\end{pmatrix}\begin{pmatrix}1\\0\end{pmatrix}, \qquad X_1 = \frac{\tfrac12}{s+1} + \frac{\tfrac12}{s+3},\quad X_2 = \frac{\tfrac12}{s+1} - \frac{\tfrac12}{s+3} .

Inverting gives x1=12(e−t+e−3t)x_1 = \tfrac12(e^{-t} + e^{-3t}) and x2=12(e−t−e−3t)x_2 = \tfrac12(e^{-t} - e^{-3t}). The eigenvalue method finds the same thing: the eigenvalues of AA are −1-1 and −3-3 with eigenvectors (1,1)(1,1) and (1,−1)(1,-1), and (1,0)=12(1,1)+12(1,−1)(1,0) = \tfrac12(1,1) + \tfrac12(1,-1) supplies the two weights. The determinant det⁡(sI−A)\det(sI - A) is the characteristic polynomial, so the poles of a system’s transform are its eigenvalues. Laplace did it without ever mentioning an eigenvector, only a 2×22\times2 solve with polynomial entries. At t=1t = 1 the lab cell below prints x1=0.208833x_1 = 0.208833 and x2=0.159046x_2 = 0.159046, and RK45 agrees.

Free sanity checks: initial and final values

Two standard facts about Y(s)Y(s) (stated here without proof, from memory) catch most algebra slips before you invert anything. The initial value theorem says y(0)=lim⁡s→∞sY(s)y(0) = \lim_{s\to\infty} sY(s). For Example A, sY=(s2+3s+4)/((s+1)(s+2))sY = (s^2+3s+4)/((s+1)(s+2)) tends to 11, matching y(0)=1y(0) = 1; for Example B it tends to 22, matching y(0)=2y(0) = 2. The final value theorem says that if every pole of sY(s)sY(s) is strictly in the left half-plane, then y(t)→lim⁡s→0sY(s)y(t) \to \lim_{s\to0} sY(s). For Example A, sY→4/2=2sY \to 4/2 = 2 at s=0s = 0, and the dashed constant term in the lab is exactly 22; for Examples B and C the limits are 11 and 5/5=15/5 = 1, again the constant terms of the lab. The condition on the poles matters: in the resonant preset sY=s2/(s2+4)2sY = s^2/(s^2+4)^2 has poles on the imaginary axis, and y(t)=14tsin⁡2ty(t) = \tfrac14t\sin2t does not settle at all, so the formula would claim a limit of 00 that the solution never reaches.

The other habit worth building is to check the answer against the data every time: evaluate y(0)y(0) and y′(0)y'(0) from your closed form, as every example above does. A forgotten −y(0)-y(0) in the transform of y′y', or a lost factor 12\tfrac12 when matching 2/((s+1)2+4)2/((s+1)^2+4), shows up immediately as a mismatch there.

Laplace against the characteristic equation

Solve Example A the other way. The characteristic roots are −1,−2-1, -2, so yh=c1e−t+c2e−2ty_h = c_1e^{-t} + c_2e^{-2t}. A constant forcing 44 suggests the constant yp=2y_p = 2 (since 2yp=42y_p = 4). Then y(0)=1y(0) = 1 gives 2+c1+c2=12 + c_1 + c_2 = 1 and y′(0)=0y'(0) = 0 gives −c1−2c2=0-c_1 - 2c_2 = 0, so c1=−2c_1 = -2, c2=1c_2 = 1: the same yy. The two methods perform the same linear algebra in different orders. Which should you reach for?

  • Laplace wins when the data and forcing are the whole problem: initial conditions enter YY directly with no constants to fix afterward; and resonant or coinciding forcing (presets 4 and 5) needs no modified guess, because a repeated pole in YY makes the extra tt automatically. It wins even more when the forcing is switched on and off or is a sum of pieces, which 6.3 Step Functions & Discontinuous Forcing handles with a single new rule.
  • The characteristic equation wins for a homogeneous equation: the roots give yhy_h directly and the constants come from two equations, with no transform needed. It also extends to variable coefficients through series or other methods, where L\mathcal L does not simplify (from memory: a coefficient like tt turns L{ty}\mathcal L\{ty\} into −Y′(s)-Y'(s), so the equation for yy becomes a differential equation for YY in ss, which is usually no easier).
  • Both cost work in different places. Undetermined coefficients costs a guess table and a linear solve for the unknown constants. Laplace costs a partial-fraction expansion, which grows with the degree of the denominator: every pole the forcing adds raises the degree of the denominator of YY by one.
  • Neither handles a forcing that has no rational transform or no neat guess (say tan⁡t\tan t), and neither handles a nonlinear equation. For those, 5.5 Variation of Parameters or a time stepper from 4.3 Runge–Kutta Methods & Butcher Tableaux is the tool.

Computing it

🐍 PythonLab: sympy solves five IVPs by Laplace, a 2x2 system, and the definition check (blocks A to C)
⌘/Ctrl + Enter

Block A prints y(1)=1.399576, 1.735759, 1.320348, 0.227324, 0.135335y(1) = 1.399576,\ 1.735759,\ 1.320348,\ 0.227324,\ 0.135335 both ways, with the closed forms found above. Block B finds the factorisation (s+1)(s+3)(s+1)(s+3), the eigenvalues −1,−3-1, -3 and agreement with RK45. Block C confirms Y(1)=4/3Y(1) = 4/3 against the definition.

Now write the solver yourself. The exercise asks for the three steps in order: build Y(s)Y(s), invert it, and compare against RK45.

🧪 Coding exerciseExercise: solve an IVP by Laplace in sympy and measure the gap to RK45
⌘/Ctrl + Enter

With the cell completed, the Laplace solution and RK45 agree to roughly 10−910^{-9} or better on both test problems.

Concept checks

✅ Concept check

For $y'' + 3y' + 2y = 4$ with $y(0) = 1$, $y'(0) = 0$, what is $Y(1)$? (Give a number.)

✅ Concept check

$Y(s)$ has poles at $-2 \pm 3i$ and nothing else. What does $y(t)$ look like?

✅ Concept check

Why does $Y(s) = s/(s^2+4)^2$ give a response that grows like $t$?

✅ Concept check

An IVP has constant coefficients, nonzero initial data and a forcing that switches on at $t = 5$. Which method needs the fewest special rules?

Problems

✏️ Pen & paperDistinct poles with zero-input and zero-state●○○

Solve y′′+5y′+6y=12y'' + 5y' + 6y = 12, y(0)=0y(0) = 0, y′(0)=1y'(0) = 1 by Laplace. Then write the zero-input and zero-state responses separately.

Show worked solution

Y=1(s+2)(s+3)+12s(s+2)(s+3)=s+12s(s+2)(s+3)Y = \dfrac{1}{(s+2)(s+3)} + \dfrac{12}{s(s+2)(s+3)} = \dfrac{s+12}{s(s+2)(s+3)}. Cover-up: at s=0s = 0, 12/6=212/6 = 2; at s=−2s = -2, 10/((−2)(1))=−510/((-2)(1)) = -5; at s=−3s = -3, 9/((−3)(−1))=39/((-3)(-1)) = 3. So y=2−5e−2t+3e−3ty = 2 - 5e^{-2t} + 3e^{-3t}. Check: y(0)=0y(0) = 0, y′(0)=10−9=1y'(0) = 10 - 9 = 1. The zero-input part is L−1{1/((s+2)(s+3))}=e−2t−e−3t\mathcal L^{-1}\{1/((s+2)(s+3))\} = e^{-2t} - e^{-3t} and the zero-state part is the remainder, 2−6e−2t+4e−3t2 - 6e^{-2t} + 4e^{-3t}.

✏️ Pen & paperA triple pole●●○

Solve y′′+4y′+4y=e−2ty'' + 4y' + 4y = e^{-2t} with y(0)=1y(0) = 1, y′(0)=0y'(0) = 0, and identify the pole structure of YY.

Show worked solution

Y=s+4(s+2)2+1(s+2)3Y = \dfrac{s+4}{(s+2)^2} + \dfrac{1}{(s+2)^3}. Write s+4=(s+2)+2s + 4 = (s+2) + 2 to get 1s+2+2(s+2)2+1(s+2)3\dfrac{1}{s+2} + \dfrac{2}{(s+2)^2} + \dfrac{1}{(s+2)^3}. By the shift theorem with L{tk}=k!/sk+1\mathcal L\{t^k\} = k!/s^{k+1}, the inverses are e−2te^{-2t}, 2te−2t2te^{-2t} and 12t2e−2t\tfrac12t^2e^{-2t}, so y=e−2t(1+2t+12t2)y = e^{-2t}\bigl(1 + 2t + \tfrac12t^2\bigr). Check: y′(0)=−2+2=0y'(0) = -2 + 2 = 0. The pole −2-2 has multiplicity three: two from the characteristic polynomial and one from the forcing e−2te^{-2t}, which is why the top power is t2t^2.

✏️ Pen & paperResonance through the transform●●●

Solve y′′+y=sin⁡ty'' + y = \sin t with y(0)=y′(0)=0y(0) = y'(0) = 0. Use L{tf}=−F′\mathcal L\{tf\} = -F' with f=cos⁡tf = \cos t.

Show worked solution

Y=1/(s2+1)2Y = 1/(s^2+1)^2. With F=s/(s2+1)F = s/(s^2+1) for cos⁡t\cos t, −F′=(s2−1)/(s2+1)2=L{tcos⁡t}-F' = (s^2-1)/(s^2+1)^2 = \mathcal L\{t\cos t\}. Since 1(s2+1)2=12[1s2+1−s2−1(s2+1)2]\dfrac{1}{(s^2+1)^2} = \tfrac12\left[\dfrac{1}{s^2+1} - \dfrac{s^2-1}{(s^2+1)^2}\right] (the bracket is 2/(s2+1)22/(s^2+1)^2), y=12(sin⁡t−tcos⁡t)y = \tfrac12(\sin t - t\cos t). Check: y(0)=0y(0) = 0 and y′=12 tsin⁡ty' = \tfrac12\,t\sin t, so y′(0)=0y'(0) = 0. The term tcos⁡tt\cos t grows without bound: resonance.

Key Takeaways

  • Transforming ay′′+by′+cy=gay'' + by' + cy = g gives P(s)Y=G+(as+b)y(0)+ay′(0)P(s)Y = G + (as+b)y(0) + ay'(0), so YY is a rational function in which the initial data and the forcing both appear.
  • The two pieces of YY are the zero-input response (initial data, no forcing) and zero-state response (forcing, from rest); their sum is the solution.
  • Partial fractions then inverse: distinct poles give exponentials, a double pole gives a factor tt (shift theorem or L{tf}=−F′\mathcal L\{tf\} = -F'), a complex pair gives eσtcos⁡ωte^{\sigma t}\cos\omega t and eσtsin⁡ωte^{\sigma t}\sin\omega t after completing the square.
  • Pole location is mode: real part is the decay rate, imaginary part the frequency, multiplicity the power of tt. Resonance is a forcing pole landing on a system pole.
  • A first-order system is solved by (sI−A)X=x(0)(sI - A)\mathbf X = \mathbf x(0); its poles are the eigenvalues of AA.
  • Laplace wins when initial data and discontinuous forcing are the problem; the characteristic equation is quicker for homogeneous equations; neither helps with nonlinear or variable-coefficient equations, where a time stepper takes over.
  • Next, 6.3 Step Functions & Discontinuous Forcing adds switched forcing.
Finished this lesson?

Mark it complete to track your progress (saved in this browser). Reading every section does this for you.

Where this leads

These later lessons build directly on this one: