# On the stability of the heat equation with an initial condition

## Abstract

In this paper, we prove the generalized Hyers-Ulam stability of the heat equation with an initial condition

in a class of twice continuously differentiable functions under certain conditions.

## 1 Introduction

Let X be a normed space, and let I be an open interval. If for any function $f:I\to X$ satisfying the differential inequality

$\parallel {a}_{n}\left(x\right){y}^{\left(n\right)}\left(x\right)+{a}_{n-1}\left(x\right){y}^{\left(n-1\right)}\left(x\right)+\cdots +{a}_{1}\left(x\right){y}^{\prime }\left(x\right)+{a}_{0}\left(x\right)y\left(x\right)+h\left(x\right)\parallel \le \epsilon$

for all $x\in I$ and for some $\epsilon \ge 0$ there exists a solution ${f}_{0}:I\to X$ of the differential equation

${a}_{n}\left(x\right){y}^{\left(n\right)}\left(x\right)+{a}_{n-1}\left(x\right){y}^{\left(n-1\right)}\left(x\right)+\cdots +{a}_{1}\left(x\right){y}^{\prime }\left(x\right)+{a}_{0}\left(x\right)y\left(x\right)+h\left(x\right)=0$

such that $\parallel f\left(x\right)-{f}_{0}\left(x\right)\parallel \le K\left(\epsilon \right)$ for any $x\in I$, where $K\left(\epsilon \right)$ is an expression of ε only, then we say that the above differential equation has the Hyers-Ulam stability.

If the above statement is also true when we replace ε and $K\left(\epsilon \right)$ by $\phi \left(x\right)$ and $\mathrm{\Phi }\left(x\right)$, where $\phi ,\mathrm{\Phi }:I\to \left[0,\mathrm{\infty }\right)$ are functions not depending on f and ${f}_{0}$ explicitly, then we say that the corresponding differential equation has the generalized Hyers-Ulam stability. (This type of stability is sometimes called the Hyers-Ulam-Rassias stability.)

We may apply these terminologies to other differential equations and partial differential equations. For more detailed definitions of the Hyers-Ulam stability and the generalized Hyers-Ulam stability, refer to .

Obłoza seems to be the first author who has investigated the Hyers-Ulam stability of linear differential equations (see [9, 10]). Here, we introduce the result of Alsina and Ger (see ): If a differentiable function $f:I\to \mathbb{R}$ is a solution of the differential inequality $|{y}^{\prime }\left(x\right)-y\left(x\right)|\le \epsilon$, where I is an open subinterval of , then there exists a solution ${f}_{0}:I\to \mathbb{R}$ of the differential equation ${y}^{\prime }\left(x\right)=y\left(x\right)$ such that $|f\left(x\right)-{f}_{0}\left(x\right)|\le 3\epsilon$ for any $x\in I$. This result was generalized by Miura et al. (see [11, 12]).

In 2007, Jung and Lee  proved the Hyers-Ulam stability of the first-order linear partial differential equation

$a{u}_{x}\left(x,y\right)+b{u}_{y}\left(x,y\right)+cu\left(x,y\right)+d=0,$

where $a,b\in \mathbb{R}$ and $c,d\in \mathbb{C}$ are constants with $\mathrm{\Re }\left(c\right)\ne 0$. It seems that the first paper dealing with the Hyers-Ulam stability of partial differential equations was written by Prastaro and Rassias . For a recent result on this subject, refer to .

In this paper, using an idea from the papers [16, 17], we investigate the generalized Hyers-Ulam stability of the heat equation with an initial value condition

(1.1)

in the class of radially symmetric functions, where denotes the Laplace operator. The heat equation plays an important role in a number of fields of science. It is strongly related to the Brownian motion in probability theory. The heat equation is also connected with chemical diffusion, and it is sometimes called the diffusion equation.

## 2 Main result

For a given integer $n\ge 2$, ${x}_{i}$ denotes the i th coordinate of any point x in ${\mathbb{R}}^{n}$, i.e., $x=\left({x}_{1},\dots ,{x}_{i},\dots ,{x}_{n}\right)$. We assume that ${t}_{1}$ is a constant with $0<{t}_{1}\le \mathrm{\infty }$, and we define

$T:=\left\{t\in \mathbb{R}\mid 0

Due to an idea from [, Section 2.3.1], we may search for a solution of (1.1) of the form $u\left(x,t\right)=\left(1/{t}^{n/2}\right)v\left(|x|/{t}^{1/2}\right)$ for some function v. Based on this argument, we define

Theorem 2.1 Let ${\phi }_{1},{\phi }_{2}:\left[0,\mathrm{\infty }\right)\to \left[0,\mathrm{\infty }\right)$ and $\psi :T\to \left[0,\mathrm{\infty }\right)$ be functions such that

$D:={\int }_{0}^{\mathrm{\infty }}\frac{{e}^{{u}^{2}/4}}{{u}^{n-1}}{\int }_{u}^{\mathrm{\infty }}{s}^{n-1}{\phi }_{1}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\phantom{\rule{0.2em}{0ex}}du<\mathrm{\infty },$
(2.1)
$c:=\underset{t\in T}{inf}{t}^{n/2+1}\psi \left(t\right)>0.$
(2.2)

Assume that $g:{\mathbb{R}}^{n}\to \mathbb{R}$ is a function with $g\in C\left({\mathbb{R}}^{n}\right)\cap {L}^{\mathrm{\infty }}\left({\mathbb{R}}^{n}\right)$. If a twice continuously differentiable function $u\in U$ satisfies

$\left\{\begin{array}{cc}|\mathrm{△}u\left(x,t\right)-{u}_{t}\left(x,t\right)|\le {\phi }_{1}\left(\frac{|x|}{{t}^{1/2}}\right)\psi \left(t\right)\hfill & \left(\mathit{\text{for all}}\phantom{\rule{0.25em}{0ex}}x\in {\mathbb{R}}^{n}\phantom{\rule{0.25em}{0ex}}\mathit{\text{and}}\phantom{\rule{0.25em}{0ex}}t\in T\right)\hfill \\ |u\left(x,0\right)-g\left(x\right)|\le {\phi }_{2}\left(|x|\right)\hfill & \left(\mathit{\text{for all}}\phantom{\rule{0.25em}{0ex}}x\in {\mathbb{R}}^{n}\right),\hfill \end{array}$
(2.3)

then there exist solutions ${u}_{0},{u}_{1}:{\mathbb{R}}^{n}×T\to \mathbb{R}$ of the heat equation and a real number γ such that

$|u\left(x,t\right)-\gamma {u}_{0}\left(x,t\right)|\le \frac{cD}{{t}^{n/2}}{e}^{-{|x|}^{2}/4t},$
(2.4)
$|{\int }_{{\mathbb{R}}^{n}}u\left(x-y,t\right)g\left(y\right)\phantom{\rule{0.2em}{0ex}}dy-\gamma {u}_{1}\left(x,t\right)|\le {\left(4\pi \right)}^{n/2}cD{\int }_{{\mathbb{R}}^{n}}{u}_{0}\left(x-y,t\right)|g\left(y\right)|\phantom{\rule{0.2em}{0ex}}dy$
(2.5)

for all $x\in {\mathbb{R}}^{n}$ and $t\in T$ with $|x|/{t}^{1/2}>0$, where

${u}_{0}\left(x,t\right):=\frac{1}{{\left(4\pi t\right)}^{n/2}}{e}^{-{|x|}^{2}/4t}\phantom{\rule{1em}{0ex}}\mathit{\text{and}}\phantom{\rule{1em}{0ex}}{u}_{1}\left(x,t\right):={\int }_{{\mathbb{R}}^{n}}{u}_{0}\left(x-y,t\right)g\left(y\right)\phantom{\rule{0.2em}{0ex}}dy.$
(2.6)

Proof Since $u\left(x,t\right)\in U$, there exists a function $w:\left[0,\mathrm{\infty }\right)\to \mathbb{R}$ such that

$u\left(x,t\right)=\frac{1}{{t}^{n/2}}w\left(r\right)$

for any $x\in {\mathbb{R}}^{n}$ and $t\in T$, where we set $r=|x|/{t}^{1/2}$. Using this notation, we calculate ${u}_{t}$ and u:

$\begin{array}{c}{u}_{t}\left(x,t\right)=-\frac{n}{2{t}^{n/2+1}}w\left(r\right)-\frac{1}{2{t}^{n/2+1}}r{w}^{\prime }\left(r\right),\hfill \\ {u}_{{x}_{i}}\left(x,t\right)=\frac{1}{{t}^{n/2+1/2}}{w}^{\prime }\left(r\right)\frac{{x}_{i}}{|x|},\hfill \\ {u}_{{x}_{i}{x}_{i}}\left(x,t\right)=\frac{1}{{t}^{n/2+1/2}}\left(\frac{1}{{t}^{1/2}}{w}^{″}\left(r\right)\frac{{x}_{i}^{2}}{{|x|}^{2}}+{w}^{\prime }\left(r\right)\left(\frac{1}{|x|}-\frac{{x}_{i}^{2}}{{|x|}^{3}}\right)\right).\hfill \end{array}$

So, we have

$\begin{array}{c}\mathrm{△}u\left(x,t\right)-{u}_{t}\left(x,t\right)\hfill \\ \phantom{\rule{1em}{0ex}}=\frac{1}{{t}^{n/2+1}}\left({w}^{″}\left(r\right)+\frac{n-1}{r}{w}^{\prime }\left(r\right)+\frac{r}{2}{w}^{\prime }\left(r\right)+\frac{n}{2}w\left(r\right)\right)\hfill \\ \phantom{\rule{1em}{0ex}}=\frac{1}{{r}^{n-1}}\frac{1}{{t}^{n/2+1}}\left(\left({r}^{n-1}{w}^{″}\left(r\right)+\left(n-1\right){r}^{n-2}{w}^{\prime }\left(r\right)\right)+\frac{1}{2}\left({r}^{n}{w}^{\prime }\left(r\right)+n{r}^{n-1}w\left(r\right)\right)\right)\hfill \\ \phantom{\rule{1em}{0ex}}=\frac{1}{{r}^{n-1}}\frac{1}{{t}^{n/2+1}}{\left({r}^{n-1}{w}^{\prime }\left(r\right)+\frac{{r}^{n}}{2}w\left(r\right)\right)}^{\prime }\hfill \end{array}$

for all $x\in {\mathbb{R}}^{n}$, $t\in T$ with $r>0$. Moreover, from the last equality and (2.3), it follows that

$\begin{array}{rcl}|\mathrm{△}u\left(x,t\right)-{u}_{t}\left(x,t\right)|& =& \frac{1}{{r}^{n-1}}\frac{1}{{t}^{n/2+1}}|{\left({r}^{n-1}{w}^{\prime }\left(r\right)+\frac{{r}^{n}}{2}w\left(r\right)\right)}^{\prime }|\\ \le & {\phi }_{1}\left(r\right)\psi \left(t\right)\end{array}$

or

$|{\left({r}^{n-1}{w}^{\prime }\left(r\right)+\frac{{r}^{n}}{2}w\left(r\right)\right)}^{\prime }|\le {r}^{n-1}{\phi }_{1}\left(r\right){t}^{n/2+1}\psi \left(t\right)$

for all $r>0$ and $t\in T$. In view of (2.2), we have

$-c{r}^{n-1}{\phi }_{1}\left(r\right)\le {\left({r}^{n-1}{w}^{\prime }\left(r\right)+\frac{{r}^{n}}{2}w\left(r\right)\right)}^{\prime }\le c{r}^{n-1}{\phi }_{1}\left(r\right)$

for any $r>0$.

We integrate each term of the last inequality from r to ∞ and take account of the definition of U to get

$-c{\int }_{r}^{\mathrm{\infty }}{s}^{n-1}{\phi }_{1}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\le -{r}^{n-1}{w}^{\prime }\left(r\right)-\frac{{r}^{n}}{2}w\left(r\right)\le c{\int }_{r}^{\mathrm{\infty }}{s}^{n-1}{\phi }_{1}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds$

or

$|{w}^{\prime }\left(r\right)+\frac{r}{2}w\left(r\right)|\le \frac{c}{{r}^{n-1}}{\int }_{r}^{\mathrm{\infty }}{s}^{n-1}{\phi }_{1}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds$

for all $r>0$.

According to [, Theorem 1], together with (2.1), there exists a (unique) constant $\gamma \in \mathbb{R}$ such that

$|w\left(r\right)-\frac{\gamma }{{\left(4\pi \right)}^{n/2}}{e}^{-{r}^{2}/4}|\le c{e}^{-{r}^{2}/4}{\int }_{r}^{\mathrm{\infty }}\frac{{e}^{{u}^{2}/4}}{{u}^{n-1}}{\int }_{u}^{\mathrm{\infty }}{s}^{n-1}{\phi }_{1}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\phantom{\rule{0.2em}{0ex}}du$

for all $r>0$, or equivalently

$|u\left(x,t\right)-\frac{\gamma }{{\left(4\pi t\right)}^{n/2}}{e}^{-{|x|}^{2}/4t}|\le \frac{c}{{t}^{n/2}}{e}^{-{|x|}^{2}/4t}{\int }_{|x|/{t}^{1/2}}^{\mathrm{\infty }}\frac{{e}^{{u}^{2}/4}}{{u}^{n-1}}{\int }_{u}^{\mathrm{\infty }}{s}^{n-1}{\phi }_{1}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\phantom{\rule{0.2em}{0ex}}du$

for all $x\in {\mathbb{R}}^{n}$ and $t\in T$ with $|x|/{t}^{1/2}>0$, which proves the validity of inequality (2.4), and in view of (2.6), it is not difficult to prove that ${u}_{0}\left(x,t\right)$ is a solution of the heat equation, i.e., $\mathrm{△}{u}_{0}\left(x,t\right)-\frac{\partial }{\partial t}{u}_{0}\left(x,t\right)=0$.

If we replace x with $x-y$ and multiply each term by $|g\left(y\right)|$, and if we integrate all terms in the last inequality over ${\mathbb{R}}^{n}$, then we obtain

$\begin{array}{c}-{\left(4\pi \right)}^{n/2}cD{\int }_{{\mathbb{R}}^{n}}{u}_{0}\left(x-y,t\right)|g\left(y\right)|\phantom{\rule{0.2em}{0ex}}dy\hfill \\ \phantom{\rule{1em}{0ex}}\le {\int }_{{\mathbb{R}}^{n}}u\left(x-y,t\right)g\left(y\right)\phantom{\rule{0.2em}{0ex}}dy-\gamma {\int }_{{\mathbb{R}}^{n}}{u}_{0}\left(x-y,t\right)g\left(y\right)\phantom{\rule{0.2em}{0ex}}dy\hfill \\ \phantom{\rule{1em}{0ex}}\le {\left(4\pi \right)}^{n/2}cD{\int }_{{\mathbb{R}}^{n}}{u}_{0}\left(x-y,t\right)|g\left(y\right)|\phantom{\rule{0.2em}{0ex}}dy\hfill \end{array}$

for all $x,y\in {\mathbb{R}}^{n}$ and $t\in T$. If we define ${u}_{1}\left(x,t\right)$ as in (2.6) for any $x\in {\mathbb{R}}^{n}$ and $t\in T$, following the proof of [, Theorem 1 in Section 2.3], we can then easily prove that

$\mathrm{△}{u}_{1}\left(x,t\right)-\frac{\partial }{\partial t}{u}_{1}\left(x,t\right)=0$

for all $x\in {\mathbb{R}}^{n}$ and $t\in T$ with $|x|/{t}^{1/2}>0$, which shows the validity of inequality (2.5). □

Remark 2.2

1. (i)

The linearity of solutions of heat equation (1.1) implies that $\gamma {u}_{0}\left(x,t\right)$ and $\gamma {u}_{1}\left(x,t\right)$ are also solutions of the heat equation.

2. (ii)

As in the proof of [, Theorem 1 in Section 2.3], we can show that

$\underset{x\in {\mathbb{R}}^{n},t>0}{\underset{\left(x,t\right)\to \left({x}_{0},0\right)}{lim}}{u}_{1}\left(x,t\right)=g\left({x}_{0}\right)$

for each ${x}_{0}\in {\mathbb{R}}^{n}$.

3. (iii)

If a function ${\phi }_{1}:\left[0,\mathrm{\infty }\right)\to \left[0,\mathrm{\infty }\right)$ satisfies

${\int }_{u}^{\mathrm{\infty }}{s}^{n-1}{\phi }_{1}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds=O\left({u}^{n}{e}^{-\alpha {u}^{2}}\right)$

for all $u\ge 0$ and for some $\alpha >1/4$, then there exists a positive number A such that

$\begin{array}{rcl}D& =& {\int }_{0}^{\mathrm{\infty }}\frac{{e}^{{u}^{2}/4}}{{u}^{n-1}}{\int }_{u}^{\mathrm{\infty }}{s}^{n-1}{\phi }_{1}\left(s\right)\phantom{\rule{0.2em}{0ex}}ds\phantom{\rule{0.2em}{0ex}}du\\ \le & {\int }_{0}^{\mathrm{\infty }}Au{e}^{\left(1/4-\alpha \right){u}^{2}}\phantom{\rule{0.2em}{0ex}}du\\ =& \frac{2A}{4\alpha -1}<\mathrm{\infty },\end{array}$

i.e., ${\phi }_{1}$ satisfies condition (2.1).

## References

1. Alsina C, Ger R: On some inequalities and stability results related to the exponential function. J. Inequal. Appl. 1998, 2: 373–380.

2. Brillouet-Belluot N, Brzdek J, Cieplinski K: On some recent developments in Ulam’s type stability. Abstr. Appl. Anal. 2012., 2012: Article ID 716936 10.1155/2012/716936

3. Czerwik S: Functional Equations and Inequalities in Several Variables. World Scientific, River Edge; 2002.

4. Hyers DH: On the stability of the linear functional equation. Proc. Natl. Acad. Sci. USA 1941, 27: 222–224. 10.1073/pnas.27.4.222

5. Hyers DH, Isac G, Rassias TM: Stability of Functional Equations in Several Variables. Birkhäuser, Boston; 1998.

6. Jung S-M Springer Optimization and Its Applications 48. In Hyers-Ulam-Rassias Stability of Functional Equations in Nonlinear Analysis. Springer, New York; 2011.

7. Rassias TM: On the stability of the linear mapping in Banach spaces. Proc. Am. Math. Soc. 1978, 72: 297–300. 10.1090/S0002-9939-1978-0507327-1

8. Ulam SM: Problems in Modern Mathematics. Wiley, New York; 1964.

9. Obłoza M: Hyers stability of the linear differential equation. Rocznik Nauk.-Dydakt. Prace Mat. 1993, 13: 259–270.

10. Obłoza M: Connections between Hyers and Lyapunov stability of the ordinary differential equations. Rocznik Nauk.-Dydakt. Prace Mat. 1997, 14: 141–146.

11. Miura T, Jung S-M, Takahasi S-E: Hyers-Ulam-Rassias stability of the Banach space valued linear differential equations ${y}^{\prime }=\lambda y$. J. Korean Math. Soc. 2004, 41: 995–1005. 10.4134/JKMS.2004.41.6.995

12. Takahasi S-E, Miura T, Miyajima S: On the Hyers-Ulam stability of the Banach space-valued differential equation ${y}^{\prime }=\lambda y$. Bull. Korean Math. Soc. 2002, 39: 309–315. 10.4134/BKMS.2002.39.2.309

13. Jung S-M, Lee K-S: Hyers-Ulam stability of first order linear partial differential equations with constant coefficients. Math. Inequal. Appl. 2007, 10: 261–266.

14. Prastaro A, Rassias TM: Ulam stability in geometry of PDE’s. Nonlinear Funct. Anal. Appl. 2003, 8: 259–278.

15. Lungu N, Popa D: Hyers-Ulam stability of a first order partial differential equation. J. Math. Anal. Appl. 2012, 385: 86–91. 10.1016/j.jmaa.2011.06.025

16. Hegyi B, Jung S-M: On the stability of Laplace’s equation. Appl. Math. Lett. 2013, 26: 549–552. 10.1016/j.aml.2012.12.014

17. Hegyi, B, Jung, S-M: On the stability of heat equation. Abstr. Appl. Anal. (in press)

18. Evans LC Graduate Studies in Mathematics 19. In Partial Differential Equations. American Mathematical Society, Providence; 1998.

19. Jung S-M: Hyers-Ulam stability of linear differential equations of first order, II. Appl. Math. Lett. 2006, 19: 854–858. 10.1016/j.aml.2005.11.004

## Acknowledgements

This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education (No. 2013R1A1A2005557).

## Author information

Authors

### Corresponding author

Correspondence to Soon-Mo Jung.

### Competing interests

The author declares that he has no competing interests.

### Authors’ contributions

The author declares that this paper is his original paper.

## Rights and permissions

Reprints and Permissions

Jung, SM. On the stability of the heat equation with an initial condition. J Inequal Appl 2013, 475 (2013). https://doi.org/10.1186/1029-242X-2013-475

• Accepted:

• Published:

• DOI: https://doi.org/10.1186/1029-242X-2013-475

### Keywords

• Differential Equation
• Partial Differential Equation
• Brownian Motion
• Recent Result
• Normed Space 