Useful Formulae I
1. sin(A ± B) = sin A cos B ± cos A sin B; cos(A ± B) = cos A cos B ∓ sin A sin B;
2 sin A cos B = sin(A + B) + sin(A − B); 2 sin A sin B = cos(A − B) − cos(A + B);
2 cos A cos B = cos(A − B) + cos(A + B).
2. (f · g)0 = f 0 · g + f · g 0Z, and (f · g)(2) = f (2) 0 0 (2)
Z · g + 2f · g + f · g .
Integration by parts: u · d(v) = u · v − v · d(u)
Z Z Z
dt dt 1 t dt t
3. = ln t + C; 2 2
= tan−1 + C; √ = sin−1 + C for any a > 0.
t a +t a a 2
a −t2 a
ebt (b cos (at) + a sin(at) ) ebt (−a cos (at) + b sin(at) )
Z Z
bt
4. e cos(at) dt = +C and ebt sin(at) dt = + C.
b2 + a2 b2 + a2
5. (Separable ODE) An ODE Z written N (x)
Z dx = M (y) dy is separable, and its general solution is given by implicitly
by a level curve defined by M (y) dy = N (x) dx + C, where C is an arbitrary constant.
6. (Exact ODE) The ODE M (x, y) dx + N (x, y) dy = 0 is exact, if My (x, y) = Nx (x, y), and its solution is given
implicitly by a level f (x, y) = C, where f (x, y) is a potential function of the vector field M (x, y)i + N (x, y)j.
7. (First order linear ODE) The integrating factor of y 0 (x) + p(x)y(x) = q(x) is exp( p(t) dt).
R
8. (Bernoulli equation) If y = y(x) is a solution of the Bernoulli equation y 0 (x) + p(x)y(x) = q(x)y(x)n , where n 6= 0
or 1, then u(x) = y(x)1−n is a solution of the following equation u0 (x) + (1 − n)p(x)u(x) = (1 − n)q(x).
9. (Characteristic Roots of constant Coefficient Linear ODE) The characteristic equation of second order linear
differential equation y 00 (x) + ay 0 (x) + by(x) = 0 ♥ with constant coefficients a and b is given by λ2 + aλ + b = 0, and
the general solution yg (x) of ♥ is given by
• yg (x) = Aeλ1 x + Beλ2 x , if λ1 and λ2 are the distinct roots of the characteristic equation;
• yg (x) = eαx ( A cos(βx) + B sin(βx) ), if α ± iβ are the complex conjugate roots of the characteristic equation;
• yg (x) = Axeλx + Beλx , if the characteristic equation has repeated real roots λ = λ1 = λ2 .
10. (Characteristic Equation of Euler-Cauchy Equation) The characteristic equation of second order linear Euler-
Cauchy differential equation x2 y 00 (x) + axy 0 (x) + by(x) = 0 ♥ on (0, +∞) with constant a and b is given by m(m −
1) + am + b = 0. In this case, the general solution of differential equations ♥ is given by
• yg (x) = Axm1 + Bxm2 , if m1 and m2 are the distinct real roots of the characteristic equation;
• yg (x) = Axm + Bxm ln(x), if the characteristic equation has repeated roots m1 = m2 = m;
√
• yg (x) = xµ ( A cos(ln x)+B sin(ln x) ), if µ±ν −1 are the complex conjugate roots of the characteristic equation.
11. Theorem (Reduction of Order for homogeneous equation). Let y1 (x) be a solution of the 2nd order homoge-
neous linear ordinary differential equation y 00 (x) + p(x)y 0 (x) + q(x)y(x) = 0 ♥ on some interval I such that y1 (x) 6= 0
R e− R p(x) dx
on I. Then y2 (x) = y1 (x) y 2 (x)
dx is another solution of ♥ such that y1 (x) and y2 (x) are linearly independent.
1
12. (Wronskian) The Wronskian of two functions f (x) and g(x) on an interval I is defined by W (x) = W (f (x), g(x)) =
f (x)g 0 (x) − f 0 (x)g(x) for all x ∈ I.
13. (Variation of Parameters) Suppose that y1 (x) and y2 (x) are two particular solutions of a second order homogeneous
linear differential equation y 00 (x) + p(x)y 0 (x) + q(x)y(x) = 0 such that their Wronskian W (x) = W ( y1 (x), y2 (x) ) is
non-zero everywhere, then a particular solution yp (x) of the non-homogeneous linear differential equation y 00 (x) +
p(x)y 0 (x) + q(x)y(x) = r(x) is given by yp (x) = u1 (x)y1 (x) + u2 (x)y2 (x), where u1 (x) and u2 (x) are unknown
functions such that their derivatives are determined by the following system of linear equations:
y1 (x)u01 (x) + y2 (x)u02 (x) = 0;
y10 (x)u01 (x) + y20 (x)u02 (x)= r(x).
Useful Formulae II
Z +∞
1. Definition of Laplace transform. L [f (t)] or F (s) = e−st f (t) dt.
0
n! Γ(α + 1)
2. L [tn ] = for non-negative integer n; L [tα ] = for any α ≥ 0.
sn+1 sα+1
s a
3. L [cos(at)] = and L [sin(at)] = .
s2 + a2 s2
+ a2
e + e−at eat − e−at
at
1 s a
4. L [e ] =
at
; L [cosh(at)] = L = 2 ; L [sinh(at)] = L = 2 .
s−a 2 s − a2 2 s − a2
5. L [y 0 (t)] = sL [y(t)] − y(0) L [y 00 (t)] = s2 L [y(t)] − y 0 (0) − sy(0).
and
6. First Shifting Theorem. L eat f (t) = F (s − a), where F (s) = L [f (t)].
7. Second Shifting Theorem. L [u(t − a)f (t − a)] = e−as F (s), where F (s) = L [f (t)],
u(t − a) is the unit step function at t = a, i.e. u(t − a) = 0 if t < a and u(t − a) = 1 if t ≥ a.
e−as
In particular, L [u(t − a)] = .
s
8. L [δ(t − a)] = e−as , where δ(t − a) is the Dirac delta function at t = a.
Z t
9. L [(f ∗ g)(t)] = L [f ]L [g], where (f ∗ g)(t) = f (τ )g(t − τ ) dτ.
0
10. L [t · f (t)] = where F (s) = L [f (t)].
−F 0 (s),
Z +∞
f (t)
11. L = F (u) du, where F (s) = L [f (t)].
t s
1 s
12. L [f (ct)] = F ( ), where F (s) = L [f (t)].
c c
13. Definition of Inverse Laplace Transform. f (t) = L −1 [F (s)], where F (s) = L [f (t)].
14. Theorem. Laplace Transform is one-to-one.
More precisely, let f, g be two continuous functions on [0, +∞) with exponential growth.
If L [f (t)] = L [g(t)], then f (t) = g(t) on [0, +∞).
tn
−1 1 −1 s −1 a
15. L = ; L = cos(at); L = sin(at), where a 6= 0.
sn+1 n! s 2 + a2 s 2 + a2
eat + e−at eat − e−at
−1 1 −1 s −1 a
16. L =e ; L
at
= cosh(at) = ; L = sinh(at) = .
s−a s 2 − a2 2 s 2 − a2 2
17. (2nd Shifting Theorem) L −1 e−as F (s) = u(t − a)f (t − a), where F (s) = L [f (t)].
Z t
−1
18. L [F (s)G(s)] = f (τ )g(t − τ ) dτ, where f (t) = L −1 [F (s)] and g(t) = L −1 [G(s)] .
0
19. L −1 [F (s − a)] = eat f (t), where f (t) = L −1 [F (s)] .
−1 −1 dF
0
20. L F (s) = L = −t · f (t), where f (t) = L −1 [F (s)] .
ds
Z t
F (s) h i
21. L −1 = f (τ ) dτ, where f (t) = L −1 F (s) .
s 0
Useful Formulae III
1. The Fourier coefficients of 2L-periodic function f (x) on [−L, L], i.e. f (x + L) = f (x) for all x ∈ R, are given
Z L Z L Z L
1 1 nπ 1 nπ
a0 = f (x) dx, an = f (x) cos x dx, bn = f (x) sin x dx, n ≥ 1.
2L −L L −L L L −L L
∞
X nπx nπx
The Fourier series of the function f (x) is a0 + an cos( ) + bn sin( ), we sometimes write
L L
n=1
∞
X nπx nπx
f (x) ∼ a0 + an cos( ) + bn sin( ).
L L
n=1
Z L Z L ∞
1 2 2
X
2. (Parseval identity) If f (x) is 2L-periodic with f (x) dx < +∞, then f (x) dx = 2a20 + (a2n + b2n ).
−L L −L n=1
Z L
1 inπx
3. The complex Fourier coefficients of 2L-periodic function f (x) on [−L, L] are given cn = f (x)e− L dx for any
2L −L
∞ ∞
X inπx X inπx
integer n. The complex Fourier series of f is cn e− L , in this case, we write f (x) ∼ cn e− L .
n=−∞ n=−∞
Z L Z L ∞
1 X
In addition, if f (x)2 dx < +∞, then f (x)2 dx = |cn |2 .
−L L −L n=−∞
4. Theorem. (Dirichlet condition convergence criteria of real/complex Fourier series) Let f be piecewise
continuous real-valued function defined on R with period 2L, i.e. f (x + 2L) = f (x) for all x ∈ R. Suppose that the
left hand derivative and right-hand derivatives of f exist at any points in R, then the Fourier series
∞
a0 X 1 − −
+ (an cos nx + bn sin nx) of f at x0 converges to (f (x+ +
0 ) + f (x0 ), where f (x0 ) and f (x0 ) are the limits of f
2 2
n=1
as x approaches to x0 from the right and from the left respectively.
Z ∞ Z ∞
f (x) cos(ωx) f (x) sin(ωx)
5. The Fourier integrals of function f (x) on R are given A(ω) = dx, B(ω) = dx.
−∞ π −∞ π
Z +∞ h i
In this case, we write f (x) ∼ A(ω) cos(ωx) + B(ω) sin(ωx) dω.
0
The last term is the Fourier integral representation of the function f on R.
1 −
In general, the Fourier integral representation converges to (f (x+0 ) + f (x0 ) under the Dirichlet condition on f.
2
Z ∞
1
ˆ
6. Fourier transform of function f (x) defined on R is given by f (ω) = √ f (x)e−iωx dx,
Z ∞ 2π −∞
ˆ 1 ˆ iωx
and the inverse Fourier transform of f is given by √ f (ω)e dω.
2π −∞
Remark. In general, the value of inverse Fourier transform of fˆ at x may not be related to the f (x). However, we
have the following
7. Theorem. (DirichletZ condition convergence criteria of the Fourier transform) If f is absolutely integrable
+∞
on the real line R i.e. |f (x)| dx < +∞, and f is piecewise smooth on each interval [−L, L] for all L > 0, then
−∞
Z +∞
1
(i) the Fourier transform fˆ(ω) of f is the function defined by fˆ(ω) = √ f (x)e−iωx dx;
2π −∞
+ −
f (x ) + f (x ) 1
Z +∞
(ii) the inverse Fourier transform =√ fˆ(ω)eiωx dω.
2 2π −∞
