What's the Mistake in Transforming the Differential Equation Bases?

  • Thread starter Thread starter kasse
  • Start date Start date
kasse
Messages
383
Reaction score
1

Homework Statement



Find a differential eq. y''+ay'+by=0 with this basis: exp(-1-i)x , exp(-1+i)x

2. The attempt at a solution

I see from this that the solutions of the characteristic eq. is two complex conjugates, and that a^2-4b=-4.

Substitution gives me the eq. -2i-a+ai+b=0, and combining these two gives me

a=(1/2) v (1/2)-i

Then I calcutale b when a=(1/2)

b=(17/16)

So that the eq. is

y''+0.5y'+(17/16)y=0

Looks quite nice, but when I transform the bases (exp(-x)*sinx and exp(-x)*cosx) and substitute, it proves not to be correct.

What's my mistake?
 
Physics news on Phys.org
kasse said:

Homework Statement



Find a differential eq. y''+ay'+by=0 with this basis: exp(-1-i)x , exp(-1+i)x

2. The attempt at a solution

I see from this that the solutions of the characteristic eq. is two complex conjugates, and that a^2-4b=-4.

Can you explain how you got the above equation?
 
learningphysics said:
Can you explain how you got the above equation?

The roots of the characteristic eq. are 0.5(-a (+/-) sqrt(a^2-4c))

Since the two solutions are complex numbers, a^2-4c must be negative, more precisely -4, since the imaginary part of the complex solutions is (+/-) 1.
 
kasse said:
The roots of the characteristic eq. are 0.5(-a (+/-) sqrt(a^2-4c))

Since the two solutions are complex numbers, a^2-4c must be negative, more precisely -4, since the imaginary part of the complex solutions is (+/-) 1.

I see... you must have made a mistake when you solved the two equations... I'd just take the sum and product of the roots... then you can write out the equation:

s^2 - qs + p... where q is the sum of the roots and p is the product...

then you get a = -q and b = p for the differential equation

or you can just multiply out:

(s - (-1 - i))(s - (-1 + i))
 
Ah, so I use the superposition principle and substitute the sum of Y1 and Y2 into the original eq?
 
No, I'm not using the superposition principle here. Since you know that the basis is exp(-1-i)x , exp(-1+i)x... you immediately know that the characteristic equation has roots (-1-i) and (-1+i).

or another way to think about it... you know that the characterstic equation is:
s^2 + as + b

This means that any function of the form e^sx that satisfies the equation must have the property that s^2 + as + b =0... ie s=(-1-i) must satisfy this equation, and s = (-1+i) must satisfy this equation... ie these are the two roots of the equation... Once you know the roots of a degree-2 equation r1 and r2... then you immediately know the form of the equation... s^2 - (r1+r2)s + r1r2... which you can also get by (s-r1)(s-r2)

You can also plug in -1-i into the equation and -1+i into the equation, then get 2 equations in 2 unknowns (a and b) and solve... but this is a little more roundabout way to do the problem I think...
 
Prove $$\int\limits_0^{\sqrt2/4}\frac{1}{\sqrt{x-x^2}}\arcsin\sqrt{\frac{(x-1)\left(x-1+x\sqrt{9-16x}\right)}{1-2x}} \, \mathrm dx = \frac{\pi^2}{8}.$$ Let $$I = \int\limits_0^{\sqrt 2 / 4}\frac{1}{\sqrt{x-x^2}}\arcsin\sqrt{\frac{(x-1)\left(x-1+x\sqrt{9-16x}\right)}{1-2x}} \, \mathrm dx. \tag{1}$$ The representation integral of ##\arcsin## is $$\arcsin u = \int\limits_{0}^{1} \frac{\mathrm dt}{\sqrt{1-t^2}}, \qquad 0 \leqslant u \leqslant 1.$$ Plugging identity above into ##(1)## with ##u...
Back
Top