Uniqueness of a nonlinear ODE












5












$begingroup$


I'm currently looking through qualifying exam questions and one has had me stumped for a few weeks. We are asked to determine whether or not there exists a unique solution of this differential equation in a neighborhood of $x = 0$. begin{equation} y'' + frac{yy'}{x^4} + y^2 = 0 , text{ } y(0)=y'(0)=0 end{equation} I believe that the solution is unique with $y = 0$ as the only solution. I have managed to prove that a non-trivial solution must satisfy $y leq 0$, but I cannot deal with $y < 0$ case.



Edit: Thank you to Ingix for noticing a mistake in my proof, which I edited below.



Indeed, assume there exists $f(x)$ that solves the above ODE such that $f(x) > 0$ on say $(0,varepsilon_1]$, where $varepsilon_1$ is positive. Then by MVT, we see begin{equation} f'(xi) = frac{f(varepsilon_1) - f(0)}{varepsilon_1-0} > 0, xi in (0, varepsilon_1) end{equation} In particular by the continuity of $text{ }f'$, there's a neighborhood $(delta_{1}, delta_{2}] subset (0,varepsilon_1]$ such that $text{ } f' > 0$ where $delta_{1} = sup_{x in [0,varepsilon_1]}$ such that $text{ }f'(x)=0$.



Then by MVT again begin{equation} frac{f'(delta_2)-f'(delta_1)}{delta_2 - delta_1} > 0 end{equation} Therefore, there's a neighborhood $[eta_1, eta_2] subset (delta_1, delta_2)$ such that $f,f',f'' > 0$. Therefore the ODE cannot be satisfied.



I am unsure how to proceed for the case $y < 0$ because the above argument would not give a contradiction. I have tried rewriting the ODE into weak form, using $u = y^2 Rightarrow u' = 2yy'$, and I do not see any obvious transformations such as equidimensional in $x/y$, scale invariant, or autonomous to simplify the ODE. Any hints on how to proceed will be very helpful!










share|cite|improve this question











$endgroup$








  • 1




    $begingroup$
    I'm not sure that your proof $y le 0$ is entirely correct. Continuity of $f'$ just makes sure that there is a neighborhood of $xi$ where $f'$ is positive. It doesn't make sure that neighborhood extends down to $0$ as the lower limit. If could be $f'=x^2sin(frac1x)$ or something similar.
    $endgroup$
    – Ingix
    Dec 19 '18 at 14:46








  • 2




    $begingroup$
    Exploring the power series of a hypothetical solution finds $y(x)=x^3v(x^5)$, $v(0)=-2$, $v'(0)=-frac{2}{17}$ which reduces the singularity to second degree in $t=x^5$, $$25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0.$$
    $endgroup$
    – LutzL
    Dec 19 '18 at 16:22












  • $begingroup$
    Thank you @Ingix, I fixed the mistake now.
    $endgroup$
    – Story123
    Dec 19 '18 at 17:37












  • $begingroup$
    @LutzL wouldn't the power series start out with a $x^2$ term? In addition why did we ignore the $x^4$ term?
    $endgroup$
    – Story123
    Dec 19 '18 at 17:38
















5












$begingroup$


I'm currently looking through qualifying exam questions and one has had me stumped for a few weeks. We are asked to determine whether or not there exists a unique solution of this differential equation in a neighborhood of $x = 0$. begin{equation} y'' + frac{yy'}{x^4} + y^2 = 0 , text{ } y(0)=y'(0)=0 end{equation} I believe that the solution is unique with $y = 0$ as the only solution. I have managed to prove that a non-trivial solution must satisfy $y leq 0$, but I cannot deal with $y < 0$ case.



Edit: Thank you to Ingix for noticing a mistake in my proof, which I edited below.



Indeed, assume there exists $f(x)$ that solves the above ODE such that $f(x) > 0$ on say $(0,varepsilon_1]$, where $varepsilon_1$ is positive. Then by MVT, we see begin{equation} f'(xi) = frac{f(varepsilon_1) - f(0)}{varepsilon_1-0} > 0, xi in (0, varepsilon_1) end{equation} In particular by the continuity of $text{ }f'$, there's a neighborhood $(delta_{1}, delta_{2}] subset (0,varepsilon_1]$ such that $text{ } f' > 0$ where $delta_{1} = sup_{x in [0,varepsilon_1]}$ such that $text{ }f'(x)=0$.



Then by MVT again begin{equation} frac{f'(delta_2)-f'(delta_1)}{delta_2 - delta_1} > 0 end{equation} Therefore, there's a neighborhood $[eta_1, eta_2] subset (delta_1, delta_2)$ such that $f,f',f'' > 0$. Therefore the ODE cannot be satisfied.



I am unsure how to proceed for the case $y < 0$ because the above argument would not give a contradiction. I have tried rewriting the ODE into weak form, using $u = y^2 Rightarrow u' = 2yy'$, and I do not see any obvious transformations such as equidimensional in $x/y$, scale invariant, or autonomous to simplify the ODE. Any hints on how to proceed will be very helpful!










share|cite|improve this question











$endgroup$








  • 1




    $begingroup$
    I'm not sure that your proof $y le 0$ is entirely correct. Continuity of $f'$ just makes sure that there is a neighborhood of $xi$ where $f'$ is positive. It doesn't make sure that neighborhood extends down to $0$ as the lower limit. If could be $f'=x^2sin(frac1x)$ or something similar.
    $endgroup$
    – Ingix
    Dec 19 '18 at 14:46








  • 2




    $begingroup$
    Exploring the power series of a hypothetical solution finds $y(x)=x^3v(x^5)$, $v(0)=-2$, $v'(0)=-frac{2}{17}$ which reduces the singularity to second degree in $t=x^5$, $$25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0.$$
    $endgroup$
    – LutzL
    Dec 19 '18 at 16:22












  • $begingroup$
    Thank you @Ingix, I fixed the mistake now.
    $endgroup$
    – Story123
    Dec 19 '18 at 17:37












  • $begingroup$
    @LutzL wouldn't the power series start out with a $x^2$ term? In addition why did we ignore the $x^4$ term?
    $endgroup$
    – Story123
    Dec 19 '18 at 17:38














5












5








5


1



$begingroup$


I'm currently looking through qualifying exam questions and one has had me stumped for a few weeks. We are asked to determine whether or not there exists a unique solution of this differential equation in a neighborhood of $x = 0$. begin{equation} y'' + frac{yy'}{x^4} + y^2 = 0 , text{ } y(0)=y'(0)=0 end{equation} I believe that the solution is unique with $y = 0$ as the only solution. I have managed to prove that a non-trivial solution must satisfy $y leq 0$, but I cannot deal with $y < 0$ case.



Edit: Thank you to Ingix for noticing a mistake in my proof, which I edited below.



Indeed, assume there exists $f(x)$ that solves the above ODE such that $f(x) > 0$ on say $(0,varepsilon_1]$, where $varepsilon_1$ is positive. Then by MVT, we see begin{equation} f'(xi) = frac{f(varepsilon_1) - f(0)}{varepsilon_1-0} > 0, xi in (0, varepsilon_1) end{equation} In particular by the continuity of $text{ }f'$, there's a neighborhood $(delta_{1}, delta_{2}] subset (0,varepsilon_1]$ such that $text{ } f' > 0$ where $delta_{1} = sup_{x in [0,varepsilon_1]}$ such that $text{ }f'(x)=0$.



Then by MVT again begin{equation} frac{f'(delta_2)-f'(delta_1)}{delta_2 - delta_1} > 0 end{equation} Therefore, there's a neighborhood $[eta_1, eta_2] subset (delta_1, delta_2)$ such that $f,f',f'' > 0$. Therefore the ODE cannot be satisfied.



I am unsure how to proceed for the case $y < 0$ because the above argument would not give a contradiction. I have tried rewriting the ODE into weak form, using $u = y^2 Rightarrow u' = 2yy'$, and I do not see any obvious transformations such as equidimensional in $x/y$, scale invariant, or autonomous to simplify the ODE. Any hints on how to proceed will be very helpful!










share|cite|improve this question











$endgroup$




I'm currently looking through qualifying exam questions and one has had me stumped for a few weeks. We are asked to determine whether or not there exists a unique solution of this differential equation in a neighborhood of $x = 0$. begin{equation} y'' + frac{yy'}{x^4} + y^2 = 0 , text{ } y(0)=y'(0)=0 end{equation} I believe that the solution is unique with $y = 0$ as the only solution. I have managed to prove that a non-trivial solution must satisfy $y leq 0$, but I cannot deal with $y < 0$ case.



Edit: Thank you to Ingix for noticing a mistake in my proof, which I edited below.



Indeed, assume there exists $f(x)$ that solves the above ODE such that $f(x) > 0$ on say $(0,varepsilon_1]$, where $varepsilon_1$ is positive. Then by MVT, we see begin{equation} f'(xi) = frac{f(varepsilon_1) - f(0)}{varepsilon_1-0} > 0, xi in (0, varepsilon_1) end{equation} In particular by the continuity of $text{ }f'$, there's a neighborhood $(delta_{1}, delta_{2}] subset (0,varepsilon_1]$ such that $text{ } f' > 0$ where $delta_{1} = sup_{x in [0,varepsilon_1]}$ such that $text{ }f'(x)=0$.



Then by MVT again begin{equation} frac{f'(delta_2)-f'(delta_1)}{delta_2 - delta_1} > 0 end{equation} Therefore, there's a neighborhood $[eta_1, eta_2] subset (delta_1, delta_2)$ such that $f,f',f'' > 0$. Therefore the ODE cannot be satisfied.



I am unsure how to proceed for the case $y < 0$ because the above argument would not give a contradiction. I have tried rewriting the ODE into weak form, using $u = y^2 Rightarrow u' = 2yy'$, and I do not see any obvious transformations such as equidimensional in $x/y$, scale invariant, or autonomous to simplify the ODE. Any hints on how to proceed will be very helpful!







real-analysis ordinary-differential-equations






share|cite|improve this question















share|cite|improve this question













share|cite|improve this question




share|cite|improve this question








edited Dec 19 '18 at 17:36







Story123

















asked Dec 19 '18 at 9:55









Story123Story123

23217




23217








  • 1




    $begingroup$
    I'm not sure that your proof $y le 0$ is entirely correct. Continuity of $f'$ just makes sure that there is a neighborhood of $xi$ where $f'$ is positive. It doesn't make sure that neighborhood extends down to $0$ as the lower limit. If could be $f'=x^2sin(frac1x)$ or something similar.
    $endgroup$
    – Ingix
    Dec 19 '18 at 14:46








  • 2




    $begingroup$
    Exploring the power series of a hypothetical solution finds $y(x)=x^3v(x^5)$, $v(0)=-2$, $v'(0)=-frac{2}{17}$ which reduces the singularity to second degree in $t=x^5$, $$25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0.$$
    $endgroup$
    – LutzL
    Dec 19 '18 at 16:22












  • $begingroup$
    Thank you @Ingix, I fixed the mistake now.
    $endgroup$
    – Story123
    Dec 19 '18 at 17:37












  • $begingroup$
    @LutzL wouldn't the power series start out with a $x^2$ term? In addition why did we ignore the $x^4$ term?
    $endgroup$
    – Story123
    Dec 19 '18 at 17:38














  • 1




    $begingroup$
    I'm not sure that your proof $y le 0$ is entirely correct. Continuity of $f'$ just makes sure that there is a neighborhood of $xi$ where $f'$ is positive. It doesn't make sure that neighborhood extends down to $0$ as the lower limit. If could be $f'=x^2sin(frac1x)$ or something similar.
    $endgroup$
    – Ingix
    Dec 19 '18 at 14:46








  • 2




    $begingroup$
    Exploring the power series of a hypothetical solution finds $y(x)=x^3v(x^5)$, $v(0)=-2$, $v'(0)=-frac{2}{17}$ which reduces the singularity to second degree in $t=x^5$, $$25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0.$$
    $endgroup$
    – LutzL
    Dec 19 '18 at 16:22












  • $begingroup$
    Thank you @Ingix, I fixed the mistake now.
    $endgroup$
    – Story123
    Dec 19 '18 at 17:37












  • $begingroup$
    @LutzL wouldn't the power series start out with a $x^2$ term? In addition why did we ignore the $x^4$ term?
    $endgroup$
    – Story123
    Dec 19 '18 at 17:38








1




1




$begingroup$
I'm not sure that your proof $y le 0$ is entirely correct. Continuity of $f'$ just makes sure that there is a neighborhood of $xi$ where $f'$ is positive. It doesn't make sure that neighborhood extends down to $0$ as the lower limit. If could be $f'=x^2sin(frac1x)$ or something similar.
$endgroup$
– Ingix
Dec 19 '18 at 14:46






$begingroup$
I'm not sure that your proof $y le 0$ is entirely correct. Continuity of $f'$ just makes sure that there is a neighborhood of $xi$ where $f'$ is positive. It doesn't make sure that neighborhood extends down to $0$ as the lower limit. If could be $f'=x^2sin(frac1x)$ or something similar.
$endgroup$
– Ingix
Dec 19 '18 at 14:46






2




2




$begingroup$
Exploring the power series of a hypothetical solution finds $y(x)=x^3v(x^5)$, $v(0)=-2$, $v'(0)=-frac{2}{17}$ which reduces the singularity to second degree in $t=x^5$, $$25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0.$$
$endgroup$
– LutzL
Dec 19 '18 at 16:22






$begingroup$
Exploring the power series of a hypothetical solution finds $y(x)=x^3v(x^5)$, $v(0)=-2$, $v'(0)=-frac{2}{17}$ which reduces the singularity to second degree in $t=x^5$, $$25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0.$$
$endgroup$
– LutzL
Dec 19 '18 at 16:22














$begingroup$
Thank you @Ingix, I fixed the mistake now.
$endgroup$
– Story123
Dec 19 '18 at 17:37






$begingroup$
Thank you @Ingix, I fixed the mistake now.
$endgroup$
– Story123
Dec 19 '18 at 17:37














$begingroup$
@LutzL wouldn't the power series start out with a $x^2$ term? In addition why did we ignore the $x^4$ term?
$endgroup$
– Story123
Dec 19 '18 at 17:38




$begingroup$
@LutzL wouldn't the power series start out with a $x^2$ term? In addition why did we ignore the $x^4$ term?
$endgroup$
– Story123
Dec 19 '18 at 17:38










1 Answer
1






active

oldest

votes


















2












$begingroup$

Obviously, $y(x)=0$ is a solution. Using the supplied informations, $y=x^au(x)$, $age 2$, $u(0)ne 0$, if a non-zero solution exists. Insert this and its derivatives to find
$$
[x^au''+2ax^{a-1}u'+a(a-1)x^{a-2}u]+frac{[x^au][x^au'+ax^{a-1}u]}{x^4}+x^{2a}u^2=0\~\
[x^2u''+2axu'+a(a-1)u]+[u][x^{a-2}u'+ax^{a-3}u]+x^{a+2}u^2=0\~\
$$

To get a non-trivial value for $u(0)$, at least two terms need to remain in this equation when setting $x=0$. This only happens for $a=3$ with the remaining equation $$6u(0)+3u(0)^2=0,$$ so that the non-trivial solution is $u(0)=-2$.



Thus for $y(x)=x^3u(x)$ the equation is
$$
x^2u''+(6+u)xu'+6u+(3+x^5)u^2=0
$$



Now one can solve for the power series coefficients of $u$ using your preferred CAS that can do power series over polynomial rings. The CAS Magma is used in the code below:



A<a>:=FunctionField(Rationals());
PS<x>:=PowerSeriesRing(A);
Pol<z>:=PolynomialRing(Rationals());
Diff:=func<w|Degree(w) lt 1 select 0 else &+[ Coefficient(w,k)*k*x^(k-1) : k in [1..Degree(w)] ]>;
prec := 40;
u := -2+0*x;
for m in [1..prec] do
ua := u+a*x^m+O(x^(m+1));
Dua := Diff(ua); D2ua:=Diff(Dua);
eqn := x^2*D2ua + (6+ua)*x*Dua + 6*ua + (3+x^5)*ua^2;
eqn;
u := u + Roots(Pol!Coefficient(eqn,m))[1][1]*x^m;
m,u;
end for;


with the result
$$
u(x)=-2 - frac{2}{17}x^5 - frac{42}{8959}x^{10}
- frac{841}{5025999}x^{15} - frac{1134494}{200418411457}x^{20}
- frac{659042819}{3546804627554529}x^{25}
- frac{60555294564493}{10115926601559333469596}x^{30}
- frac{21682330793829581}{113844637973948738866833384}x^{35}
- frac{8763668817047648604028}{1458948251016373153173581517077169}x^{40}+...
$$

This strongly suggests that the solution has the general form $y(x)=x^3v(x^5)$. Inserting again
begin{align}
y'(x)&=5x^7v'(x^5)+3x^2v(x^5)text{ and }\ y''(x)&=25x^{11}v''(x^5)+50x^6v'(x^5)+6xv(x^5)
end{align}

results in
$$
[25x^{10}v''(x^5)+50x^5v'(x^5)+6v(x^5)]+v(x^5)[5x^5v'(x^5)+3v(x^5)]+x^5v(x^5)^2=0
$$

and with $t=x^5$
$$
25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0,
$$

which looks a little easier from a singularity point-of-view.





One could also start the power series computation directly with $y$ with the same method,



y := 0*x;
for m in [1..prec] do
ya := y+a*x^m+O(x^(m+1));
Dya := Diff(ya); D2ya:=Diff(Dya);
eqn := x^4*D2ya + ya*Dya + x^4*ya^2;
m,eqn;
n := Valuation(eqn);
q := Pol!Coefficient(eqn,n);
if Degree(q) gt 0 then
rts := Roots(q); "a in ",[ rr[1]: rr in rts];
r := rts[1][1];
if r eq 0 and #rts gt 1 then r := rts[2][1]; end if;
y := y + Roots(q)[1][1]*x^m;
end if;
m,y;
end for;


which returns the same



-2*x^3 - 2/17*x^8 - 42/8959*x^13 - 841/5025999*x^18 - ...


only with more logistical effort.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thank you so much, this was very helpful. But I was wondering if there were any analytic methods for solving the above ODE since it was given on a qual exam with closed everything.
    $endgroup$
    – Story123
    Dec 19 '18 at 18:58











Your Answer





StackExchange.ifUsing("editor", function () {
return StackExchange.using("mathjaxEditing", function () {
StackExchange.MarkdownEditor.creationCallbacks.add(function (editor, postfix) {
StackExchange.mathjaxEditing.prepareWmdForMathJax(editor, postfix, [["$", "$"], ["\\(","\\)"]]);
});
});
}, "mathjax-editing");

StackExchange.ready(function() {
var channelOptions = {
tags: "".split(" "),
id: "69"
};
initTagRenderer("".split(" "), "".split(" "), channelOptions);

StackExchange.using("externalEditor", function() {
// Have to fire editor after snippets, if snippets enabled
if (StackExchange.settings.snippets.snippetsEnabled) {
StackExchange.using("snippets", function() {
createEditor();
});
}
else {
createEditor();
}
});

function createEditor() {
StackExchange.prepareEditor({
heartbeatType: 'answer',
autoActivateHeartbeat: false,
convertImagesToLinks: true,
noModals: true,
showLowRepImageUploadWarning: true,
reputationToPostImages: 10,
bindNavPrevention: true,
postfix: "",
imageUploader: {
brandingHtml: "Powered by u003ca class="icon-imgur-white" href="https://imgur.com/"u003eu003c/au003e",
contentPolicyHtml: "User contributions licensed under u003ca href="https://creativecommons.org/licenses/by-sa/3.0/"u003ecc by-sa 3.0 with attribution requiredu003c/au003e u003ca href="https://stackoverflow.com/legal/content-policy"u003e(content policy)u003c/au003e",
allowUrls: true
},
noCode: true, onDemand: true,
discardSelector: ".discard-answer"
,immediatelyShowMarkdownHelp:true
});


}
});














draft saved

draft discarded


















StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3046227%2funiqueness-of-a-nonlinear-ode%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown

























1 Answer
1






active

oldest

votes








1 Answer
1






active

oldest

votes









active

oldest

votes






active

oldest

votes









2












$begingroup$

Obviously, $y(x)=0$ is a solution. Using the supplied informations, $y=x^au(x)$, $age 2$, $u(0)ne 0$, if a non-zero solution exists. Insert this and its derivatives to find
$$
[x^au''+2ax^{a-1}u'+a(a-1)x^{a-2}u]+frac{[x^au][x^au'+ax^{a-1}u]}{x^4}+x^{2a}u^2=0\~\
[x^2u''+2axu'+a(a-1)u]+[u][x^{a-2}u'+ax^{a-3}u]+x^{a+2}u^2=0\~\
$$

To get a non-trivial value for $u(0)$, at least two terms need to remain in this equation when setting $x=0$. This only happens for $a=3$ with the remaining equation $$6u(0)+3u(0)^2=0,$$ so that the non-trivial solution is $u(0)=-2$.



Thus for $y(x)=x^3u(x)$ the equation is
$$
x^2u''+(6+u)xu'+6u+(3+x^5)u^2=0
$$



Now one can solve for the power series coefficients of $u$ using your preferred CAS that can do power series over polynomial rings. The CAS Magma is used in the code below:



A<a>:=FunctionField(Rationals());
PS<x>:=PowerSeriesRing(A);
Pol<z>:=PolynomialRing(Rationals());
Diff:=func<w|Degree(w) lt 1 select 0 else &+[ Coefficient(w,k)*k*x^(k-1) : k in [1..Degree(w)] ]>;
prec := 40;
u := -2+0*x;
for m in [1..prec] do
ua := u+a*x^m+O(x^(m+1));
Dua := Diff(ua); D2ua:=Diff(Dua);
eqn := x^2*D2ua + (6+ua)*x*Dua + 6*ua + (3+x^5)*ua^2;
eqn;
u := u + Roots(Pol!Coefficient(eqn,m))[1][1]*x^m;
m,u;
end for;


with the result
$$
u(x)=-2 - frac{2}{17}x^5 - frac{42}{8959}x^{10}
- frac{841}{5025999}x^{15} - frac{1134494}{200418411457}x^{20}
- frac{659042819}{3546804627554529}x^{25}
- frac{60555294564493}{10115926601559333469596}x^{30}
- frac{21682330793829581}{113844637973948738866833384}x^{35}
- frac{8763668817047648604028}{1458948251016373153173581517077169}x^{40}+...
$$

This strongly suggests that the solution has the general form $y(x)=x^3v(x^5)$. Inserting again
begin{align}
y'(x)&=5x^7v'(x^5)+3x^2v(x^5)text{ and }\ y''(x)&=25x^{11}v''(x^5)+50x^6v'(x^5)+6xv(x^5)
end{align}

results in
$$
[25x^{10}v''(x^5)+50x^5v'(x^5)+6v(x^5)]+v(x^5)[5x^5v'(x^5)+3v(x^5)]+x^5v(x^5)^2=0
$$

and with $t=x^5$
$$
25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0,
$$

which looks a little easier from a singularity point-of-view.





One could also start the power series computation directly with $y$ with the same method,



y := 0*x;
for m in [1..prec] do
ya := y+a*x^m+O(x^(m+1));
Dya := Diff(ya); D2ya:=Diff(Dya);
eqn := x^4*D2ya + ya*Dya + x^4*ya^2;
m,eqn;
n := Valuation(eqn);
q := Pol!Coefficient(eqn,n);
if Degree(q) gt 0 then
rts := Roots(q); "a in ",[ rr[1]: rr in rts];
r := rts[1][1];
if r eq 0 and #rts gt 1 then r := rts[2][1]; end if;
y := y + Roots(q)[1][1]*x^m;
end if;
m,y;
end for;


which returns the same



-2*x^3 - 2/17*x^8 - 42/8959*x^13 - 841/5025999*x^18 - ...


only with more logistical effort.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thank you so much, this was very helpful. But I was wondering if there were any analytic methods for solving the above ODE since it was given on a qual exam with closed everything.
    $endgroup$
    – Story123
    Dec 19 '18 at 18:58
















2












$begingroup$

Obviously, $y(x)=0$ is a solution. Using the supplied informations, $y=x^au(x)$, $age 2$, $u(0)ne 0$, if a non-zero solution exists. Insert this and its derivatives to find
$$
[x^au''+2ax^{a-1}u'+a(a-1)x^{a-2}u]+frac{[x^au][x^au'+ax^{a-1}u]}{x^4}+x^{2a}u^2=0\~\
[x^2u''+2axu'+a(a-1)u]+[u][x^{a-2}u'+ax^{a-3}u]+x^{a+2}u^2=0\~\
$$

To get a non-trivial value for $u(0)$, at least two terms need to remain in this equation when setting $x=0$. This only happens for $a=3$ with the remaining equation $$6u(0)+3u(0)^2=0,$$ so that the non-trivial solution is $u(0)=-2$.



Thus for $y(x)=x^3u(x)$ the equation is
$$
x^2u''+(6+u)xu'+6u+(3+x^5)u^2=0
$$



Now one can solve for the power series coefficients of $u$ using your preferred CAS that can do power series over polynomial rings. The CAS Magma is used in the code below:



A<a>:=FunctionField(Rationals());
PS<x>:=PowerSeriesRing(A);
Pol<z>:=PolynomialRing(Rationals());
Diff:=func<w|Degree(w) lt 1 select 0 else &+[ Coefficient(w,k)*k*x^(k-1) : k in [1..Degree(w)] ]>;
prec := 40;
u := -2+0*x;
for m in [1..prec] do
ua := u+a*x^m+O(x^(m+1));
Dua := Diff(ua); D2ua:=Diff(Dua);
eqn := x^2*D2ua + (6+ua)*x*Dua + 6*ua + (3+x^5)*ua^2;
eqn;
u := u + Roots(Pol!Coefficient(eqn,m))[1][1]*x^m;
m,u;
end for;


with the result
$$
u(x)=-2 - frac{2}{17}x^5 - frac{42}{8959}x^{10}
- frac{841}{5025999}x^{15} - frac{1134494}{200418411457}x^{20}
- frac{659042819}{3546804627554529}x^{25}
- frac{60555294564493}{10115926601559333469596}x^{30}
- frac{21682330793829581}{113844637973948738866833384}x^{35}
- frac{8763668817047648604028}{1458948251016373153173581517077169}x^{40}+...
$$

This strongly suggests that the solution has the general form $y(x)=x^3v(x^5)$. Inserting again
begin{align}
y'(x)&=5x^7v'(x^5)+3x^2v(x^5)text{ and }\ y''(x)&=25x^{11}v''(x^5)+50x^6v'(x^5)+6xv(x^5)
end{align}

results in
$$
[25x^{10}v''(x^5)+50x^5v'(x^5)+6v(x^5)]+v(x^5)[5x^5v'(x^5)+3v(x^5)]+x^5v(x^5)^2=0
$$

and with $t=x^5$
$$
25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0,
$$

which looks a little easier from a singularity point-of-view.





One could also start the power series computation directly with $y$ with the same method,



y := 0*x;
for m in [1..prec] do
ya := y+a*x^m+O(x^(m+1));
Dya := Diff(ya); D2ya:=Diff(Dya);
eqn := x^4*D2ya + ya*Dya + x^4*ya^2;
m,eqn;
n := Valuation(eqn);
q := Pol!Coefficient(eqn,n);
if Degree(q) gt 0 then
rts := Roots(q); "a in ",[ rr[1]: rr in rts];
r := rts[1][1];
if r eq 0 and #rts gt 1 then r := rts[2][1]; end if;
y := y + Roots(q)[1][1]*x^m;
end if;
m,y;
end for;


which returns the same



-2*x^3 - 2/17*x^8 - 42/8959*x^13 - 841/5025999*x^18 - ...


only with more logistical effort.






share|cite|improve this answer











$endgroup$













  • $begingroup$
    Thank you so much, this was very helpful. But I was wondering if there were any analytic methods for solving the above ODE since it was given on a qual exam with closed everything.
    $endgroup$
    – Story123
    Dec 19 '18 at 18:58














2












2








2





$begingroup$

Obviously, $y(x)=0$ is a solution. Using the supplied informations, $y=x^au(x)$, $age 2$, $u(0)ne 0$, if a non-zero solution exists. Insert this and its derivatives to find
$$
[x^au''+2ax^{a-1}u'+a(a-1)x^{a-2}u]+frac{[x^au][x^au'+ax^{a-1}u]}{x^4}+x^{2a}u^2=0\~\
[x^2u''+2axu'+a(a-1)u]+[u][x^{a-2}u'+ax^{a-3}u]+x^{a+2}u^2=0\~\
$$

To get a non-trivial value for $u(0)$, at least two terms need to remain in this equation when setting $x=0$. This only happens for $a=3$ with the remaining equation $$6u(0)+3u(0)^2=0,$$ so that the non-trivial solution is $u(0)=-2$.



Thus for $y(x)=x^3u(x)$ the equation is
$$
x^2u''+(6+u)xu'+6u+(3+x^5)u^2=0
$$



Now one can solve for the power series coefficients of $u$ using your preferred CAS that can do power series over polynomial rings. The CAS Magma is used in the code below:



A<a>:=FunctionField(Rationals());
PS<x>:=PowerSeriesRing(A);
Pol<z>:=PolynomialRing(Rationals());
Diff:=func<w|Degree(w) lt 1 select 0 else &+[ Coefficient(w,k)*k*x^(k-1) : k in [1..Degree(w)] ]>;
prec := 40;
u := -2+0*x;
for m in [1..prec] do
ua := u+a*x^m+O(x^(m+1));
Dua := Diff(ua); D2ua:=Diff(Dua);
eqn := x^2*D2ua + (6+ua)*x*Dua + 6*ua + (3+x^5)*ua^2;
eqn;
u := u + Roots(Pol!Coefficient(eqn,m))[1][1]*x^m;
m,u;
end for;


with the result
$$
u(x)=-2 - frac{2}{17}x^5 - frac{42}{8959}x^{10}
- frac{841}{5025999}x^{15} - frac{1134494}{200418411457}x^{20}
- frac{659042819}{3546804627554529}x^{25}
- frac{60555294564493}{10115926601559333469596}x^{30}
- frac{21682330793829581}{113844637973948738866833384}x^{35}
- frac{8763668817047648604028}{1458948251016373153173581517077169}x^{40}+...
$$

This strongly suggests that the solution has the general form $y(x)=x^3v(x^5)$. Inserting again
begin{align}
y'(x)&=5x^7v'(x^5)+3x^2v(x^5)text{ and }\ y''(x)&=25x^{11}v''(x^5)+50x^6v'(x^5)+6xv(x^5)
end{align}

results in
$$
[25x^{10}v''(x^5)+50x^5v'(x^5)+6v(x^5)]+v(x^5)[5x^5v'(x^5)+3v(x^5)]+x^5v(x^5)^2=0
$$

and with $t=x^5$
$$
25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0,
$$

which looks a little easier from a singularity point-of-view.





One could also start the power series computation directly with $y$ with the same method,



y := 0*x;
for m in [1..prec] do
ya := y+a*x^m+O(x^(m+1));
Dya := Diff(ya); D2ya:=Diff(Dya);
eqn := x^4*D2ya + ya*Dya + x^4*ya^2;
m,eqn;
n := Valuation(eqn);
q := Pol!Coefficient(eqn,n);
if Degree(q) gt 0 then
rts := Roots(q); "a in ",[ rr[1]: rr in rts];
r := rts[1][1];
if r eq 0 and #rts gt 1 then r := rts[2][1]; end if;
y := y + Roots(q)[1][1]*x^m;
end if;
m,y;
end for;


which returns the same



-2*x^3 - 2/17*x^8 - 42/8959*x^13 - 841/5025999*x^18 - ...


only with more logistical effort.






share|cite|improve this answer











$endgroup$



Obviously, $y(x)=0$ is a solution. Using the supplied informations, $y=x^au(x)$, $age 2$, $u(0)ne 0$, if a non-zero solution exists. Insert this and its derivatives to find
$$
[x^au''+2ax^{a-1}u'+a(a-1)x^{a-2}u]+frac{[x^au][x^au'+ax^{a-1}u]}{x^4}+x^{2a}u^2=0\~\
[x^2u''+2axu'+a(a-1)u]+[u][x^{a-2}u'+ax^{a-3}u]+x^{a+2}u^2=0\~\
$$

To get a non-trivial value for $u(0)$, at least two terms need to remain in this equation when setting $x=0$. This only happens for $a=3$ with the remaining equation $$6u(0)+3u(0)^2=0,$$ so that the non-trivial solution is $u(0)=-2$.



Thus for $y(x)=x^3u(x)$ the equation is
$$
x^2u''+(6+u)xu'+6u+(3+x^5)u^2=0
$$



Now one can solve for the power series coefficients of $u$ using your preferred CAS that can do power series over polynomial rings. The CAS Magma is used in the code below:



A<a>:=FunctionField(Rationals());
PS<x>:=PowerSeriesRing(A);
Pol<z>:=PolynomialRing(Rationals());
Diff:=func<w|Degree(w) lt 1 select 0 else &+[ Coefficient(w,k)*k*x^(k-1) : k in [1..Degree(w)] ]>;
prec := 40;
u := -2+0*x;
for m in [1..prec] do
ua := u+a*x^m+O(x^(m+1));
Dua := Diff(ua); D2ua:=Diff(Dua);
eqn := x^2*D2ua + (6+ua)*x*Dua + 6*ua + (3+x^5)*ua^2;
eqn;
u := u + Roots(Pol!Coefficient(eqn,m))[1][1]*x^m;
m,u;
end for;


with the result
$$
u(x)=-2 - frac{2}{17}x^5 - frac{42}{8959}x^{10}
- frac{841}{5025999}x^{15} - frac{1134494}{200418411457}x^{20}
- frac{659042819}{3546804627554529}x^{25}
- frac{60555294564493}{10115926601559333469596}x^{30}
- frac{21682330793829581}{113844637973948738866833384}x^{35}
- frac{8763668817047648604028}{1458948251016373153173581517077169}x^{40}+...
$$

This strongly suggests that the solution has the general form $y(x)=x^3v(x^5)$. Inserting again
begin{align}
y'(x)&=5x^7v'(x^5)+3x^2v(x^5)text{ and }\ y''(x)&=25x^{11}v''(x^5)+50x^6v'(x^5)+6xv(x^5)
end{align}

results in
$$
[25x^{10}v''(x^5)+50x^5v'(x^5)+6v(x^5)]+v(x^5)[5x^5v'(x^5)+3v(x^5)]+x^5v(x^5)^2=0
$$

and with $t=x^5$
$$
25t^2v''(t)+5t(10+v(t))v'(t)+6v(t)+(3+t)v(t)^2=0,
$$

which looks a little easier from a singularity point-of-view.





One could also start the power series computation directly with $y$ with the same method,



y := 0*x;
for m in [1..prec] do
ya := y+a*x^m+O(x^(m+1));
Dya := Diff(ya); D2ya:=Diff(Dya);
eqn := x^4*D2ya + ya*Dya + x^4*ya^2;
m,eqn;
n := Valuation(eqn);
q := Pol!Coefficient(eqn,n);
if Degree(q) gt 0 then
rts := Roots(q); "a in ",[ rr[1]: rr in rts];
r := rts[1][1];
if r eq 0 and #rts gt 1 then r := rts[2][1]; end if;
y := y + Roots(q)[1][1]*x^m;
end if;
m,y;
end for;


which returns the same



-2*x^3 - 2/17*x^8 - 42/8959*x^13 - 841/5025999*x^18 - ...


only with more logistical effort.







share|cite|improve this answer














share|cite|improve this answer



share|cite|improve this answer








edited Dec 19 '18 at 18:32

























answered Dec 19 '18 at 17:57









LutzLLutzL

59.2k42056




59.2k42056












  • $begingroup$
    Thank you so much, this was very helpful. But I was wondering if there were any analytic methods for solving the above ODE since it was given on a qual exam with closed everything.
    $endgroup$
    – Story123
    Dec 19 '18 at 18:58


















  • $begingroup$
    Thank you so much, this was very helpful. But I was wondering if there were any analytic methods for solving the above ODE since it was given on a qual exam with closed everything.
    $endgroup$
    – Story123
    Dec 19 '18 at 18:58
















$begingroup$
Thank you so much, this was very helpful. But I was wondering if there were any analytic methods for solving the above ODE since it was given on a qual exam with closed everything.
$endgroup$
– Story123
Dec 19 '18 at 18:58




$begingroup$
Thank you so much, this was very helpful. But I was wondering if there were any analytic methods for solving the above ODE since it was given on a qual exam with closed everything.
$endgroup$
– Story123
Dec 19 '18 at 18:58


















draft saved

draft discarded




















































Thanks for contributing an answer to Mathematics Stack Exchange!


  • Please be sure to answer the question. Provide details and share your research!

But avoid



  • Asking for help, clarification, or responding to other answers.

  • Making statements based on opinion; back them up with references or personal experience.


Use MathJax to format equations. MathJax reference.


To learn more, see our tips on writing great answers.




draft saved


draft discarded














StackExchange.ready(
function () {
StackExchange.openid.initPostLogin('.new-post-login', 'https%3a%2f%2fmath.stackexchange.com%2fquestions%2f3046227%2funiqueness-of-a-nonlinear-ode%23new-answer', 'question_page');
}
);

Post as a guest















Required, but never shown





















































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown

































Required, but never shown














Required, but never shown












Required, but never shown







Required, but never shown







Popular posts from this blog

How do I know what Microsoft account the skydrive app is syncing to?

Grease: Live!

When does type information flow backwards in C++?