Solution of
with
y(0)=1.
We seek a solution of form![]()
Logan 1.4 on page 50 (1st edition).
Here is the proposed series solution:
![]()
![]()
![]()
![]()
Here is the equation with the proposed perturbation solution:
![]()
![y _ 0^'[t] + y _ 1^'[t] ϵ + y _ 2^'[t] ϵ^2 + O[ϵ]^3 == (1 + y _ 0[t]^2) + (y _ 0[t]^2 + 2 y _ 0[t] y _ 1[t]) ϵ + (2 y _ 0[t] y _ 1[t] + y _ 1[t]^2 + 2 y _ 0[t] y _ 2[t]) ϵ^2 + O[ϵ]^3](HTMLFiles/index_9.gif)
With the O[ϵ] symbol, the LogicalExpand does a powerful step:
![]()
![]()
Here is another way to look at it:
![]()
|
|
|
|
|
|
Here is the ODE for
:
![]()
![]()
...and it's solution:
![]()
![]()
![]()
Here is the ODE for
, making use of the known solution for
:
![]()
![]()
Here is the solution for
:
![]()
![]()
![]()
![]()
![]()
![]()
![]()
![-(-1 - 2 t + Cos[2 t])^2/(4 (-1 + Sin[2 t])^2) - ((-1 - 2 t + Cos[2 t]) Tan[π/4 + t])/(-1 + Sin[2 t]) - 2 Tan[π/4 + t] y _ 2[t] + y _ 2^'[t] == 0](HTMLFiles/index_34.gif)
![]()
![{{y _ 2[t] -> 1/8 (Cos[t] - Sin[t])^(-1 - (2 Cos[2 t])/((Cos[t] - Sin[t])^3 (Cos[t] + Sin[t])) + Sin[4 t]/((Cos[t] - Sin[t])^3 (Cos[t] + Sin[t]))) (-3 Cos[t] - 2 t Cos[t] + 4 t^2 Cos[t] + 3 Cos[t] Cos[2 t] + 5 Sin[t] + 10 t Sin[t] + 4 t^2 Sin[t] - 3 Cos[2 t] Sin[t])}}](HTMLFiles/index_36.gif)
Now convert these replacement rules into approximate solutions, of increasing O[ϵ],
that we can plot:
![]()
![]()
![]()
![]()
![]()
![Tan[π/4 + t] + ((-1 - 2 t + Cos[2 t]) ϵ)/(2 (-1 + Sin[2 t])) + 1/8 (Cos[t] - Sin[t])^(-1 - (2 Cos[2 t])/((Cos[t] - Sin[t])^3 (Cos[t] + Sin[t])) + Sin[4 t]/((Cos[t] - Sin[t])^3 (Cos[t] + Sin[t]))) (-3 Cos[t] - 2 t Cos[t] + 4 t^2 Cos[t] + 3 Cos[t] Cos[2 t] + 5 Sin[t] + 10 t Sin[t] + 4 t^2 Sin[t] - 3 Cos[2 t] Sin[t]) ϵ^2 + O[ϵ]^3](HTMLFiles/index_42.gif)
![]()
![]()
![]()
![]()
![]()
![1/8 ϵ^2 (Cos[t] - Sin[t])^(-1 - (2 Cos[2 t])/((Cos[t] - Sin[t])^3 (Cos[t] + Sin[t])) + Sin[4 t]/((Cos[t] - Sin[t])^3 (Cos[t] + Sin[t]))) (-3 Cos[t] - 2 t Cos[t] + 4 t^2 Cos[t] + 3 Cos[t] Cos[2 t] + 5 Sin[t] + 10 t Sin[t] + 4 t^2 Sin[t] - 3 Cos[2 t] Sin[t]) + (ϵ (-1 - 2 t + Cos[2 t]))/(2 (-1 + Sin[2 t])) + Tan[π/4 + t]](HTMLFiles/index_48.gif)
Test these functions:
![]()
![]()
Lucky for us, we can find the exact solution:
![]()
![]()
![]()
and here it is as a convenient function:
![]()
![]()
Let's test it:
![]()
![]()
![]()
Here we plot the exact solution (green), the
(red), ![]()
(blue), and
(black).
![Plot[{yexact[t, ep], ya0[t], ya1[t, ep], ya2[t, ep]}, {t, 0, .7}, PlotStyle -> {RGBColor[0, 1, 0], RGBColor[1, 0, 0], RGBColor[0, 0, 1], RGBColor[0, 0, 0]}]](HTMLFiles/index_63.gif)
![[Graphics:HTMLFiles/index_64.gif]](HTMLFiles/index_64.gif)
![]()
Converted by Mathematica (March 3, 2003)