Explore BrainMass
Share

Solve Finite Difference Equation

This content was STOLEN from BrainMass.com - View the original, and get the already-completed solution here!

See attached file for full problem description with equation.

---
Find analytically the solution of this difference equation with the given initial values:

Without computing the solution recursively, predict whether such a computation would be stable.

(Note: A numerical process is unstable if small errors made at one stage of the process are magnified in subsequent stages and seriously degrade the accuracy of the overall calculation.)
---

© BrainMass Inc. brainmass.com October 16, 2018, 5:23 pm ad1c9bdddf
https://brainmass.com/math/graphs-and-functions/solve-finite-difference-equation-52144

Attachments

Solution Preview

Please see the attached file.

Difference equation:
$$
x_{n+1} = -0.2 x_n + 0.99 x_n. eqno(1)
$$
Starting conditions:
$$
x_0 = 1, hskip 1cm x_1=0.9. eqno(2)
$$
This is a linear homogeneous equation with constant coefficients. Therefore its solutions are a linear combination of basic solutions of form
$$
B_n = cq^n,eqno(3)
$$
where $c$ and $q$ are constants.
Substituting the basic form (3) into equation ...

Solution Summary

The expert solves finite differences in equations. Analytical solutions of the difference equations with the given initial values are provided.

$2.19
Similar Posting

Differential Equations: Solution to Heat Equation

Consider the heat equation

delta(u)/delta(t) = (delta^2)(u)/delta(x^2)

Show that if u(c, t) = (t^alpha)psi(E) where E = x/sqrt(t) and alpha is a constant, then psi(E) satisfies the ordinary differential equation

alpha(psi) = 1/2 E(psi) = psi, where ' = d/dE

is independent of t only if alpha = - 1/2. Further, show that if alpha - 1/2 then

C - 1/2 E(psi) = psi

where C is an arbitrary constant. From this last ordinary differential equation, and assuming C = 0, deduce that

u(x, t) = A/sqrt(t) e^-x^3/At

is a solution of the heat equation (here A is an arbitrary constant).

Show that as t tends to zero from above,

lim(1/sqrt(t) * e^(-x^2/4t)) = 0 for x =/ 0

and that for all t > 0

The integral of 1/sqrt(t) * e^-x^3/4t dx = B

where B is a finite constant

Given that the integral e^-x^2 dx = sqrt(pi), find B. What physical and/or probabilistic interpretation might one give to u(x,t)?

View Full Posting Details