Math, asked by sumitverma0108, 1 year ago

The initial value problem dy/dx = x² + y² , y(0) = 0 has a unique solution in some interval of the form - h < x < h. is it true or false? justify please

Answers

Answered by kvnmurty
7
Yes. It is true.

See the diagram. That is the graph of  y(x)  for which:
          y '(x) = x² + y²,      y(x=0) = 0

Solution:
   y(x) = x³/3 + x⁷/(3²*7) + 2 x¹¹/(3²*7*11*3)
               + 13 x¹⁵/(3²*7*11*3*15*49) + 166 x¹⁹ /(3²*7*11*3*19*15*49*3)+..

We will find the solution by iteration methods.  We cannot express the solution to the given ODE in terms of known elementary functions. 
 
  One form of exact solution uses Bessel functions of first kind J_α (x) and second kind Y_α (x) with the order α = 1/4 and argument of x^2/2.  The Bessel function uses Gamma function with α = 1/4.

 An exact solution for the given problem may be given as a polynomial in terms of exp(x^3/3) and x, under Integral sign.

We can use Taylor's series expansion method or Picardi's method to find the solution as a polynomial in x , by using iterations.
===========

y ' = f(x,y) = x² + y² ,   y(0) = 0

y_{k+1}(x)=y(0)+\int \limits_{0}^{x} {f(t,y_k)} \, dt=0+\int\limits_{0}^{x} {(t^2+y_k^2)} \, dt\\\\So\ Initial\ solution: \ y_1(x)=\frac{x^3}{3}\\\\y_2(x)=\frac{x^3}{3}+\int\limits_{0}^{x} {(t^3/3)^2} \, dt=\frac{x^3}{3}+\frac{x^7}{3^2*7}\\\\y_3(x)=\frac{x^3}{3}+\int\limits_{0}^{x} {(\frac{t^3}{3}+\frac{t^7}{3^2*7})^2} \, dt=\frac{x^3}{3}+\frac{x^7}{3^2*7}+\frac{2x^{11}}{3^2*7*11*3}+\frac{x^{15}}{63^2*15}\\\\y_4(x)=\frac{x^3}{3}+\int \limits_{0}^{x} {(y_3(t))^2} \, dt\\\\

Thus
y₄(x) = x³/3 + x⁷/(3²*7) + 2 x¹¹/(3²*7*11*3)
               + 13 x¹⁵/(3²*7*11*3*15*49) + 166 x¹⁹ /(3²*7*11*3*19*15*49*3)+..

So we see that we have a unique solution for  - h < x < h  in the neighbourhood of zero.
Attachments:

kvnmurty: click on the red heart thanks button above pls
kvnmurty: @sumitverma respond if that is good for u
sumitverma0108: @kvnmurty it's very lengthy buddy..this question is only of 2 marks....please make an easy solution ( 2 marks solution ) of this..
kvnmurty: OKAY. We see that y' = x^2+y^2 is defined, continuous and is unique over -h<x <h;; y(0) = 0... From the definition of derivative: For small h, y(x+h) = y(x) + h * y'(h/2) ;; So y(h)=y(0)+h*[(h/2)^2 + {y(h/2)}^2 ]....y(h)=h^3/2 + h* {y(h/2)}^2
kvnmurty: So y(h) is unique and is well defined for small h in the neighbourhood of x=0.. Further y', y", y''', y'''' etc are all defined in the neighbourhood of x = 0...thus in the interval of -h < x < h, for small h.... y is uniquely defined... There is no singularity.
sumitverma0108: yes...it's okay..thankyou
Similar questions