Mr. Tenenbaum's and Prof. Mattuck's advice not working (ODE)

Click For Summary
The discussion focuses on solving the differential equation y'' + y = 4x sin x using complexification. The complementary function is derived from the characteristic equation, yielding y_c = c_1 e^{ix} + c_2 e^{-ix}. The particular solution is proposed as y_p = (Ax^2 + Bx + C)e^{ix}, leading to a system of equations for coefficients A and B. A participant expresses uncertainty about the correctness of the second derivative calculation of (x^2 e^{ix}), suggesting that errors may lie there. The thread emphasizes the need for careful verification of the derivatives to resolve the discrepancies in the solution.
Hall
Messages
351
Reaction score
87
Homework Statement
$$
y'' + y = 4x \sin x
$$
Relevant Equations
Case 2: Q(x) contains a term which, ignoring constant coefficients, is ##x^k## times a term ##u(x)## of ##y_c##, where ##k## is zero or a positive integer. In this case a particular solution ##y_p## of ##(21.1)## will be a linear combination of ##x^{k+1} u(x)## and all its linearly independent derivatives.
All right, we got
$$
y'' + y = 4x \sin x
$$

We are doing the Complexification
$$
\tilde{y''} + \tilde{y} = 4x e^{ix}
$$
Complementary function:
$$
\begin{align*}
\textrm{characteristic equation =}\\
m^2 + 1 = 0 \\
m = \pm i \\
\tilde{y_c} = c_1 e^{ix} + c_2 e^{-ix} \\
\end{align*}
$$

Q(x), that is RHS of the original non-homogeneous equation, is xeix (ignoring constant coefficients), therefore,
$$
\begin{align*}
\tilde{y_p} = (Ax^2 + Bx + C) e^{ix} \\
\tilde{y_p }'' + \tilde{y_ p} = 2 A e^{ix} + 2 i (Ax+B) e^{ix} \\
4x = 2A + 2iAx + 2B \\
\implies A = -2 i , ~B = 2 i \\
\end{align*}
$$

Hence,
$$
\begin{align*}
\tilde{y_p} = (-2 i x^2 + 2i x + C) e^{ix} \\
\textrm{We can leave that C because it is already there in}~y_c \\
Im(\tilde{y_p})= y_p = -2 x^2 \cos x + 2x \cos x \\
\end{align*}
$$

The answer that I have got is wrong, but I don't know where the mistake lies.
 
Physics news on Phys.org
I don't think you calculated (x^2e^{ix})'' correctly. Please show your working for this.
 
pasmith said:
I don't think you calculated (x^2e^{ix})'' correctly. Please show your working for this.
$$
\begin{align*}
(x^2 e^{ix})' = 2xe^{ix} + \iota x^2 e^{ix} \\
(x^2 e^{ix})'' = 2 e^{ix} + \iota ~ 2x e^{ix} + \iota 2 x e^{ix} - x^2e^{ix} \\
(x^2 e^{ix})'' = 2 e^{ix} + \iota~ 4x e^{ix} - x^2e^{ix}
\end{align*}
$$
A linear combination of ##x^2 e^{ix}##, ##2x e^{ix} + \iota x^2 e^{ix}##, and ##2 e^{ix} + \iota~4x e^{ix} -x^2e^{ix}## would give us something like this:
$$
\begin{align*}
A' x^2 e^{ix} + 2B' x e^{ix} + \iota ~ B' x^2 e^{ix} + 2C' e^{ix} + \iota~ 4C' x e^{ix} - C' x^2 e^{ix} \\
(A'+ \iota~ B' -C') x^2 e^{ix} + (2B' + \iota 4C') x e^{ix} + 2C' e^{ix} \\
A x^2 e^{ix} + B x e^{ix} + C e^{ix} \\
(Ax^2 + Bx + C)e^{ix} \\
\end{align*}
$$
 
If y_p = Ax^2 e^{ix} + Bxe^{ix} then <br /> y&#039;&#039;_p + y_p = A(2 + 4ix)e^{ix} + B(2i)e^{ix}. Setting that equal to 4xe^{ix} gives <br /> \begin{split}<br /> 2A + 2iB &amp;= 0 \\<br /> 4iA &amp;= 4.\end{split}
 
First, I tried to show that ##f_n## converges uniformly on ##[0,2\pi]##, which is true since ##f_n \rightarrow 0## for ##n \rightarrow \infty## and ##\sigma_n=\mathrm{sup}\left| \frac{\sin\left(\frac{n^2}{n+\frac 15}x\right)}{n^{x^2-3x+3}} \right| \leq \frac{1}{|n^{x^2-3x+3}|} \leq \frac{1}{n^{\frac 34}}\rightarrow 0##. I can't use neither Leibnitz's test nor Abel's test. For Dirichlet's test I would need to show, that ##\sin\left(\frac{n^2}{n+\frac 15}x \right)## has partialy bounded sums...