MHB Integral equation Fredholm series

Click For Summary
The discussion centers on the application of the Fredholm series to the integral equation defined by f(x) and the calculations for D(λ) and D(x,y;λ). The orders of λ were calculated, revealing that terms of order 3 and 4 are zero, leading to D(λ) being expressed as 1 - λ/2 + λ²/90. The derived expression for D(x,y;λ) shows repeated zeros after order λ² and λ³, resulting in a specific form for f(x). The discrepancy between the results obtained from the Fredholm method and the sum of separable kernels raises questions about potential errors in the calculations.
Dustinsfl
Messages
2,217
Reaction score
5
Given
\[
f(x) = 1 + \lambda\int_0^1(xy + x^3y^2)f(y)dy.
\]
To use the Fredholm series, we need to find \(D(\lambda)\) which is \(1 - \text{order }\lambda + \cdots\). I have already calculated the orders and of order 3 and 4 the terms are 0 so the terms greater than 3 are all zero since we have repeated zeros.
\[
\text{order }\lambda:\quad\int_0^1K(y,y)dy = \frac{1}{2}
\]
and
\[
\text{order }\lambda^2:\quad\iint_0^1
\begin{vmatrix}
K(y,y) & K(y,y')\\
K(y',y) & K(y',y')
\end{vmatrix}
dydy' = \frac{1}{90}
\]
Therefore,
\[
D(\lambda) = 1 - \frac{\lambda}{2} + \frac{\lambda^2}{90}.
\]
Next we need to find \(\mathcal{D}(x,y;\lambda)\). In the Taylor series, we obtain repeated zeros after order lambda squared and lambda cubed. Again, repeated zeros so
\[
\mathcal{D}(x,y;\lambda) = K(x,y) - \lambda\int_0^1
\begin{vmatrix}
K(x,y) & K(x,z)\\
K(z,y) & K(z,z)
\end{vmatrix}dz =
-\lambda \left(\frac{x^3 y^2}{3}-\frac{x^3 y}{4}-\frac{x y^2}{5}+\frac{x y}{6}\right)+x^3 y^2+x y
\]
I get
\[
f(x) = 1 + \frac{\lambda}{D(\lambda)}\int_0^1\mathcal{D}(x,y;\lambda)dy
= \text{see second post}
\]
 
Last edited:
Physics news on Phys.org
The solution I get for the Fredholm method is
\[
f(x) = 1 - 3x\lambda\frac{\lambda - 30}{180 + \lambda(2\lambda - 45)} +
5x^3\lambda\frac{\lambda + 24}{360 + 2\lambda(2\lambda - 45)},
\]
and when I use the sum of separable kernels, I get
\[
f(x) = 1 - 3x\lambda\frac{\lambda - 30}{180 + \lambda(\lambda - 90)} +
5x^3\lambda\frac{\lambda + 24}{360 + 2\lambda(\lambda - 90)},
\]
The math for the Fredholm method is above and the math for the separable kernels is in http://mathhelpboards.com/differential-equations-17/integral-equations-7278.html second post. They should yield the same result but aren't. They are just off a little. Which one is wrong? Where is there a mistake?
 

Similar threads

  • · Replies 1 ·
Replies
1
Views
3K
  • · Replies 13 ·
Replies
13
Views
2K
  • · Replies 1 ·
Replies
1
Views
407
  • · Replies 4 ·
Replies
4
Views
2K
  • · Replies 7 ·
Replies
7
Views
3K
  • · Replies 11 ·
Replies
11
Views
4K
  • · Replies 3 ·
Replies
3
Views
5K
  • · Replies 65 ·
3
Replies
65
Views
7K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 4 ·
Replies
4
Views
3K