Skip to main content

New answers tagged

1 vote

Approximating solution to vector recurrence relation with element-wise exponential

This turned out to be quite long and complicated, maybe someone can come up with a shorter solution. You have the system of ODEs: $$\frac{d\vec{x}(t)}{dt} = A \exp(-\gamma \odot \vec{x(t)}) $$ ...
orchi_d's user avatar
  • 301
1 vote

Can we convert the following integral equation to a differential equation:$h(r)= \int_0^\infty\frac{f(x)}{e^{r x} + 1} dx$?

Alternatively, you can consider a little bit more general problem by introducing an auxiliary variable and defining the following function : $$ g(r,s) := \int_0^\infty \frac{f(sx)}{e^{rx}+1} \,\mathrm{...
Abezhiko's user avatar
  • 10.3k
1 vote

Global existence of an ODE

To guarantee the existence of a global solution (you can't in general), you need to look at the forward dynamics $u'=f(u)$ and the reverse dynamics $u'=-f(u)$. Let $f(u) = u-u^3$, note that $f(u) = 0 $...
copper.hat's user avatar
  • 174k
1 vote

Finding singular solution to a Lagrange equation

A differential equation of type $$y = x\varphi \left( {y'} \right) + \psi \left( {y'} \right),$$ where φ (y' ) and ψ (y' ) are known functions differentiable on a certain interval, is called the ...
Prasoon Pandey's user avatar
2 votes

Stochastic differential equations with only time integrals

OK - time for an update. Thanks to everyone who commented. As pointed out by some commentators, these types of differential equations are called Random (Ordinary) Differential Equations. They have ...
rufus_lawrence's user avatar
0 votes
Accepted

Change of variables by scaling.

You need to take the transformation rule into account. Because of the scaling $\mathrm{d}{z}\not = \mathrm{d}r$ but rather $\mathrm{d}z=\sqrt{\lambda}~\mathrm{d}r$ and if I did not mix up my ...
Jonas Lenz's user avatar
  • 1,414
1 vote
Accepted

How would I prove or disprove this statement?

There is nothing that prevents this equation or its explicit form $$ y^{(n)}=\frac{\sum_{k=0}^{n-1}y^{(k)}}{\prod_{k=0}^{n-1}y^{(k)}-1} $$ from having a local, unique and analytic solution. Obviously ...
Lutz Lehmann's user avatar
3 votes
Accepted

The space of solutions for an ODE on an interval containing a regular singular point

Try $x^2 y''(x) - 6 x y'(x) + 12 y(x) = 0$. $y(x) = a x^3 + b x^4$ are solutions, and since all of these have $y''(0) = 0$ you can have a solution with different $a$'s and $b$'s on $(0,+\infty)$ and ...
Robert Israel's user avatar
1 vote
Accepted

Asymptotic stability and Lyapunov functions

That is not possible because on any closed annulus $\overline B_{\epsilon_1}(x_0)\setminus B_{\epsilon_2}(x_0)$ with $\epsilon_2<\epsilon_1$, $\Lambda'(t)\le -\alpha$ for some $\alpha>0$ (here $...
GReyes's user avatar
  • 17k
1 vote
Accepted

Global existence of an ODE

Since $f(u)=u-u^3$ is locally Lipschitz, the IVP $u'=f(u),u(t_0)=u_0$ has a unique solution $u=u(t)$ in $(t_0,t_0+\delta)$ for any $u_0\in \mathbb R$ for small $\delta>0$. Extend $u=u(t)$ to the ...
xpaul's user avatar
  • 45.4k
4 votes
Accepted

Does Solution to Banach Space ODE $\dot{x} = Ax$ with Non-Positive Spectrum Converges to Kernel?

No. Consider the case $X = \mathbb{R}^2$ and the system $$ \begin{aligned} x_1' &= x_2 \\ x_2' &= 0. \end{aligned} $$ The kernel of the operator $A = \begin{pmatrix}0 & 1 \\ 0 & 0 \end{...
whpowell96's user avatar
  • 6,099
2 votes
Accepted

Can we convert the following integral equation to a differential equation:$h(r)= \int_0^\infty\frac{f(x)}{e^{r x} + 1} dx$?

You can go part of the way if you substitute $x$ with $u=rx$. You then have $x=u/r$ and $dx=\frac{1}{r}du$. If $r>0$ $h(r)=\frac{1}{r}\int_{0}^{\infty}\frac{f(\frac{u}{r})}{e^u+1}du \\ \Rightarrow ...
Baenazril's user avatar
1 vote

Maximal interval of existence of a first order IVP

Since $f$ is bounded, say $|f(s)|\le M$ and therefore $|y'(x)|\le M$, you know that any local solution issued from $(0,0)$ will be contained in the sector $|y|\le Mx$. Therefore you can extend the ...
GReyes's user avatar
  • 17k
0 votes
Accepted

Differential inequation $f'(x) > f(x)/x$

"Is this condition sufficient?" It is almost sufficient. "Can I already conclude? Or am I missing something?" You are missing negative numbers. Consider this Image : We do not ...
Prem's user avatar
  • 12.3k
1 vote

A brick sliding in an horizontal plane after an initial push (under Coulomb's dry friction) - closed form solutions validation?

There is nothing particularly obstructive about the use of $\mathrm{sgn}$ in the DE. The DE remains locally Lipschitz almost everywhere. To start, the DE in question, $$x'' = -k\,g\,\mathrm{sgn}(x')$$ ...
Rollen S. D'Souza's user avatar
2 votes
Accepted

Using the Laplace transform, solve the system of linear differential equations with constant coefficients

You can rewrite the Laplace transforms of the two ODEs as \begin{align} Y(s)+Z(s)&=\frac{1}{s^3}+\frac{3}{s}, \tag{1} \\ s^2Y(s)-Z(s)&=3s-2+\frac{1}{s+1}. \tag{2} \end{align} Adding $(1)$ and $...
Gonçalo's user avatar
  • 11.3k
0 votes
Accepted

What is the definition of a differential equation's general solution?

I like the way you combined the Solutions ! There are ways to combine 2 (or more) Constants into one , though that is not Simple Constant , In other words , I think Constants should be Simple ...
Prem's user avatar
  • 12.3k
2 votes

Solving ODE system with less equations

Yes, in this case of a linear equation you do not need to solve the original equation as prerequisite to solve the sensitivity equations. This is of course different for non-linear equations, as each ...
Lutz Lehmann's user avatar
0 votes

What is the definition of a differential equation's general solution?

An ODE, to be "ordinary", has a domain, an open set in the time-state space, that contains no singular points for the DE. Inside the domain the derivative can be expressed as a continuous ...
Lutz Lehmann's user avatar
2 votes
Accepted

Given is the system of differential equations

a) and b) $H=3x^2+4y^2$ is a solution. So the phase portraits are $ 3x^2+4y^2=C$ with different constant 'C'. c) You solve the eigenvalue of this system,and for a linear system, if all the eigenvalues ...
UnnamedUser's user avatar
4 votes

Find solution of the IVP $ y' = y+ \frac12 |\sin(y^2)|,\,\, x>0,\,\, y(0) = -1$

You have $0 \leqslant y' - y \leqslant \frac{1}{2}$. Multiplying by $e^{-x}$ yields $0 \leqslant y'(x)e^{-x} - y(x)e^{-x} \leqslant \frac{1}{2}e^{-x}$. Integrating on $[0,x]$, after noticing that $y'(...
Didier's user avatar
  • 20k
1 vote

About solution to homogeneous ODE $u'' + u = 0$

Another approach, let $f(x) = u(x)^2+(u'(x))^2$. Then $f(0) = 0$ and $f'(x) = 0$ hence $f(x) = 0$.
copper.hat's user avatar
  • 174k
2 votes

Find solution of the IVP $ y' = y+ \frac12 |\sin(y^2)|,\,\, x>0,\,\, y(0) = -1$

You should check first that the right hand side is locally Lipschitz in $y$. Picard-Lindelöf implies existence and uniqueness. Under these circumstances, solutions to first order autonomous ODEs (like ...
Hyperbolic PDE friend's user avatar
1 vote
Accepted

About solution to homogeneous ODE $u'' + u = 0$

We can check that $u''+u=0$ has the Solution $u=c_1e^{-x}+c_2e^{+x}$ : $u'=-c_1e^{-x}+c_2e^{+x}$ $u''=c_1e^{-x}+c_2e^{+x}$ [[ Solving Characteristic Polynomial $D^2+1=0$ , we get $D=+1,D=-1$ , hence ...
Prem's user avatar
  • 12.3k
3 votes

Why are some differential equations unsolvable?

The simple pendulum is also unsolvable analytically, unless you allow an Elliptic Integral, a special function. This is in direct relation to Liouville's theorem on analytical integration. ODEs are as ...
Yves Daoust's user avatar
1 vote

Global existence of solution of ODE - Gray-Scott model

Even easier/more intuitive and with some qualitative remarks on the behaviour of solutions. It can be seen that the positive cone $K_+:=[0,+\infty)\times [0,+\infty)$ is forward invariant (if we start ...
MatteoDR's user avatar
  • 101
2 votes

Poincaré-Bendixson applied to region enclosed by homoclinic orbit

I think your assumption is generally wrong. Consider the following system of differential equations: $x'=x^2-y^2,\; y'=2xy.$ The origin is the only equilibrium position. Note that the equation of the ...
Á. Nágel's user avatar
1 vote

Solving Laplace equation

In order to use the method of separation of variables, we need homogeneous boundary conditions. Defining $v:=u-1$, we can rewrite the problem as \begin{align} v_{xx}+v_{yy}&=0, \tag{1} \\ v(x,0)&...
Gonçalo's user avatar
  • 11.3k
5 votes
Accepted

Global existence of solution of ODE - Gray-Scott model

Assume by contradiction, there exists a local in time solution $(x,y)$, which explodes in finite time $T$ (this is the opposite of having a global solution by the explosion in finite time theorem). ...
alex440's user avatar
  • 556
4 votes

What does ordinary differential equation mean, based on the word itself?

Elaborating the "WHY" : Etymology/History of the term can be traced back by the HSM Post which user Moo commented. The semantic meaning can be considered like this : We can have Linear ...
Prem's user avatar
  • 12.3k
0 votes

Green function of a differential equation

It is to be noted that the operator $\Lambda = \partial_{yy} \circ L$ generates a PDE by acting on a bivariate function $u(x,y)$. In consequence, the associated fundamental solution $G(x,y;s,t)$ will ...
Abezhiko's user avatar
  • 10.3k
2 votes

Using the Laplace transform, solve the system of linear differential equations with constant coefficients

Now you face the following linear system : $$ \begin{bmatrix} s & s \\ s^2 & -1 \end{bmatrix} \begin{bmatrix} Y(s) \\ Z(s) \end{bmatrix} = \begin{bmatrix} 3+\frac{1}{s^2} \\ 3s-2+\frac{1}{s+1} ...
Abezhiko's user avatar
  • 10.3k
4 votes

Find $\sum_{n=1}^{\infty}\left(n\sin\left(\frac{\pi n}{2}\right)\left(e^x-1-\frac{x}{1!}-\frac{x^2}{2!}-\cdots-\frac{x^n}{n!}\right)\right)$

Another way to solve the problem is to use: How to integrate $ \int x^n e^x dx$? or the regularized gamma function to find: $$\sum_{j=n+1}^\infty\frac{x^j}{j!}=\frac{e^x}{n!}\int_0^x e^{-t}t^ndt$$ ...
Тyma Gaidash's user avatar
5 votes
Accepted

Find $\sum_{n=1}^{\infty}\left(n\sin\left(\frac{\pi n}{2}\right)\left(e^x-1-\frac{x}{1!}-\frac{x^2}{2!}-\cdots-\frac{x^n}{n!}\right)\right)$

Some thoughts. (We need to prove that the function is differentiable etc.) We have $$A'(x) = A(x) + \sum_{n=1}^{\infty}\left(n\sin\left(\frac{\pi n}{2}\right)\cdot \frac{x^n}{n!}\right) = A(x) + x\cos ...
River Li's user avatar
  • 40.3k
2 votes
Accepted

Showing $M = i\sum\limits_{n=-\infty}^{\infty} u_{n}u_{n+1}^{\ast} - u_{n}^{\ast}u_{n+1}$ is conserved

Your last equation has a few mistakes. Since $i \dot{u}_{n}^* = (-i\dot{u}_n)^*=(u_{n+1}^*+u_{n-1}^*)(1+|u_{n}|^2)$, it should be \begin{align} \frac{dM}{dt}&= \sum\limits_{n=-\infty}^{\infty}[-\,(...
Gonçalo's user avatar
  • 11.3k
3 votes

Using the Laplace transform, solve the system of linear differential equations with constant coefficients

Hint: From your first equation $sY(s) + sZ(s) -3 = 1/s^2$, solve for $Z(s)$ in terms of $s$ and $Y(s)$. Then, plug this into the second equation $s^2Y(s) - 3s + 2 - Z(s) = \frac 1 {(s+1)}$ and solve ...
user1149748's user avatar
2 votes
Accepted

How to solve linear differential equation with polynomial coefficients?

As this equation has an essential singularity at $x=0$ it might help to shift that singularity to infinity. Thus insert $y(x)=u(t)=u(1/x)$, $$ y'(x)=-\frac1{x^2}u'(1/x)\\ y''(x)=\frac1{x^4}u''(1/x)+\...
Lutz Lehmann's user avatar
1 vote

How to linearize first order ODE

Assuming you need to linearize $$ f' + g(f) = 0 $$ around $f_0(x)$ we have $$ f' + \left(\frac{\partial g}{\partial f}\right)_{f=f_0}(f-f_0) + O(|f-f_0|^2)=0 $$ so we follow with $$ f' + \left(\frac{...
Cesareo's user avatar
  • 34.2k
1 vote
Accepted

Solutions to $y'=y$

This approach gives you all the solutions: First, note that $e^x$ is a solution. Suppose that $f$ is any other solution and consider $g(x)=f(x)e^{-x}$. Then $g'(x)=f'(x)e^{-x}-f(x)e^{-x} = 0$, so $g(x)...
jjagmath's user avatar
  • 18.9k
2 votes
Accepted

How to linearize first order ODE

\begin{equation} \frac{df}{dx} + \frac{f^2}{a} + \frac{f}{2a} - \frac{3}{2a} =0\quad \text{Condition : }f(a)=1 \end{equation} This is a Riccati ODE which can solved easily for the general solution. ...
JJacquelin's user avatar
  • 66.8k
2 votes

Numerical methods derivation

In one-step or Runge-Kutta methods you would explore their properties by inserting abstract Taylor series. This includes computing compositions of power series with symbolic variables as coefficients, ...
Lutz Lehmann's user avatar
1 vote

How to solve linear differential equation with polynomial coefficients?

Writing the ode as (for non zero $x$) \begin{equation} x^{2}y^{\prime\prime}-\frac{y}{x^{2}}=0\tag{1}% \end{equation} Comparing the above to the generalized form of the Bessel ode given by Bowman (...
Nasser's user avatar
  • 2,180
5 votes
Accepted

Periodic solutions of differential equation

The constant $y=0$ is technically a periodic solution, but I'm assuming you're excluding that from consideration. Suppose $y$ is a periodic solution of your equation that is not identically $0$, and ...
Robert Israel's user avatar
1 vote

How is this calculation of solution to damped oscillator done in MIT OCW "Vibrations and Waves" course?

There is a very silly oversight in the OP that was pointed out by user Semiclassical. Let me show what that oversight was. We posited a trial solution $z(t)=e^{i\alpha t}$. We found two roots $$\alpha=...
xoux's user avatar
  • 5,021
1 vote

$\frac{\text {d}y}{\text {d}x} = e^y$ general solution $y = -\ln(-x+C)$ or $y = -\ln|-x+C|$?

Checking the given answer: \begin{align*} \frac{\mathrm{d}}{\mathrm{d}x} \left( -\ln |-x-C| \right) &= \frac{-1}{|-x-C|}\frac{\mathrm{d}}{\mathrm{d}x}\left( |-x-C| \right) \\ &= \frac{...
Eric Towers's user avatar
  • 67.8k
2 votes
Accepted

$\frac{\text {d}y}{\text {d}x} = e^y$ general solution $y = -\ln(-x+C)$ or $y = -\ln|-x+C|$?

Your answer is right and the given answer is wrong. One can check for example that $y=-\ln|{-}x|$ is not a solution to the differential equation when $x>0$ (where $y=-\ln x$).
Greg Martin's user avatar
  • 82.2k
-2 votes

How to solve linear differential equation with polynomial coefficients?

In TeX With the differential equations for the chain rule $$\frac{d}{dx}= \frac{dt}{dx}\ \frac{d}{dt} = \frac{1}{\frac{dx}{dt}} \ \frac{d}{dt} $$ $$x^4 y''(x) \ - \ y(x) \quad \rightarrow \...
Roland F's user avatar
  • 3,234
2 votes

Non-trivial solution to $x'(t)=2\cdot t\cdot x+t^2\cdot y, y'(t)=t^3\cdot x+4\cdot t\cdot y$ via by hand possible?

We face a system characterized by the following matrix : $$ \begin{cases} \dot{x} = -2tx + t^2y \\ \dot{y} = t^3x + 4ty \end{cases} \quad\Longrightarrow\quad A = \begin{pmatrix} -2t & t^...
Abezhiko's user avatar
  • 10.3k
1 vote

Existence and uniqueness of homoclinic orbit

H(x,y) in the solution is not a first integral. It is easy to check that the trajectories along H are not constant.
ceg69y's user avatar
  • 11
1 vote
Accepted

Determining $t(x)$ from $\frac{dx}{dt}$?

Numerically, the calculation of $x(t)$ is a table with two rows, $t$, and $x$, and it simultaneously calculates $t(x)$: calculating $t(x)$ from is simply a matter of reading the table inversely. ...
TheAlertGerbil's user avatar

Top 50 recent answers are included