Nth order linear ode, why do we have n general solutions?

  • Context: Undergrad 
  • Thread starter Thread starter popopopd
  • Start date Start date
  • Tags Tags
    General Linear Ode
Click For Summary

Discussion Overview

The discussion revolves around the existence and uniqueness of solutions to nth order linear ordinary differential equations (ODEs). Participants explore why there are n general solutions and the implications of this in terms of vector spaces and linear combinations of solutions.

Discussion Character

  • Exploratory
  • Technical explanation
  • Mathematical reasoning

Main Points Raised

  • One participant expresses confusion about why there are n general solutions for nth order linear ODEs.
  • Another participant asks about the significance of the characteristic equation related to the ODE.
  • Several participants explain that the set of solutions to linear homogeneous equations forms a vector space with dimension n, implying the existence of n independent solutions.
  • Participants provide examples of initial value problems to illustrate the uniqueness of solutions corresponding to different initial conditions.
  • It is noted that any solution can be expressed as a linear combination of the n independent solutions, with specific initial values leading to unique functions.
  • One participant mentions using Picard's iteration to demonstrate the existence and uniqueness of solutions, suggesting that without n initial conditions, the ODE will have n constants, reflecting the dimensionality of the solution space.
  • There is a suggestion that function spaces are vector spaces, which relates to the dimensionality of the solutions.

Areas of Agreement / Disagreement

Participants generally agree on the concept that the solutions form a vector space with dimension n and that there are n independent solutions. However, there is some uncertainty regarding the implications of Picard's iteration and the nature of initial conditions.

Contextual Notes

Some discussions involve assumptions about the nature of initial conditions and the dependence on definitions of vector spaces and solutions, which remain unresolved.

popopopd
Messages
12
Reaction score
0
hi, I looked up the existence and uniqueness of nth order linear ode and I grasped the idea of them, but still kind of confused why we get n numbers of general solutions.
 
Physics news on Phys.org
A basic property of linear homogeneous equations is that the set of solutions forms a vector space. That is, any linear combination of solutions, a_1y_1+ a_2y_2 is again a solution. One can show that, for an nth order homogeneous differential equation, this vector space has dimension n. That is, there exist n independent solutions such that any solution can be written in terms of those n solutions.

To prove that, consider the differential equation a_n(x)y^{(n)}(x)+ a_{n-1}y^{(n-1)}(x)+ \cdot\cdot\cdot+ a_2y''(x)+ a_1y'(x)+ a_0y(x)= 0.
with the following initial values:

I) y(0)= 1, y'(0)= y''(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
By the fundamental "existence and uniqueness" theorem for initial value problems, there exist a unique function, y_0(x), satisfying the differential equation and those initial conditions.

II) y(0)= 0, y'(0)= 1, y''(0)= y'''(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
Again, there exist a unique solution, y_1(x), satisfying the differential equation and those initial conditions.

III) y(0)= y'(0)= 0, y''(0)= 1, y'''(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
Again, there exist a unique solution, y_2(x), satisfying the differential equation and those initial conditions.

Continue that, shifting the "= 1" through the derivatives until we get to
X) y(0)= y'(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0, y^{(n-1)}(0)= 1.

First, any solution to the differential equation can written as a linear combination of those n solutions:
Suppose y(x) is a solution to the differential equation. Let A_0= y(0), A_1= y'(0), etc. until A_{n-1}= y^{(n-1)}(0).
Then y(x)= A_0y_0(x)+ A_1y_1(x)+ \cdot\cdot\cdot+ A_{n-1}y_{n-1}(x). That can be shown by evaluating both sides, and their derivatives, at x= 0.

Further, that set of n solutions are independent. To see that suppose that, for some numbers, A_0, A_1, \cdot\cdot\cdot, A_{n-1}, A_0y_0(x)+ A_1y_1(x)+ \cdot\cdot\cdot+ A_{n-1}y_{n-1}(x)= 0- that is, is equal to 0 for all x. Taking x= 0 we must have A_0(1)+ A_1(0)+ \cdot\cdot\cdot+ A_{n-1}(0)= A_0= 0. Since that linear combination is 0 for all x, it is a constant and its derivative is 0 for all x. That is A_0y_0'(x)+ A_1y_1'(0)+ \cdot\cdot\cdot+ A_{n-1}y'(0)= 0 for all x. Set x= 0 to see that A_1= 0[/tex]. Continue taking derivatives and setting x= 0 to see that each A_i, in turn, is equal to 0.
 
HallsofIvy said:
A basic property of linear homogeneous equations is that the set of solutions forms a vector space. That is, any linear combination of solutions, a_1y_1+ a_2y_2 is again a solution. One can show that, for an nth order homogeneous differential equation, this vector space has dimension n. That is, there exist n independent solutions such that any solution can be written in terms of those n solutions.

To prove that, consider the differential equation a_n(x)y^{(n)}(x)+ a_{n-1}y^{(n-1)}(x)+ \cdot\cdot\cdot+ a_2y''(x)+ a_1y'(x)+ a_0y(x)= 0.
with the following initial values:

I) y(0)= 1, y'(0)= y''(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
By the fundamental "existence and uniqueness" theorem for initial value problems, there exist a unique function, y_0(x), satisfying the differential equation and those initial conditions.

II) y(0)= 0, y'(0)= 1, y''(0)= y'''(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
Again, there exist a unique solution, y_1(x), satisfying the differential equation and those initial conditions.

III) y(0)= y'(0)= 0, y''(0)= 1, y'''(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
Again, there exist a unique solution, y_2(x), satisfying the differential equation and those initial conditions.

Continue that, shifting the "= 1" through the derivatives until we get to
X) y(0)= y'(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0, y^{(n-1)}(0)= 1.

First, any solution to the differential equation can written as a linear combination of those n solutions:
Suppose y(x) is a solution to the differential equation. Let A_0= y(0), A_1= y'(0), etc. until A_{n-1}= y^{(n-1)}(0).
Then y(x)= A_0y_0(x)+ A_1y_1(x)+ \cdot\cdot\cdot+ A_{n-1}y_{n-1}(x). That can be shown by evaluating both sides, and their derivatives, at x= 0.

Further, that set of n solutions are independent. To see that suppose that, for some numbers, A_0, A_1, \cdot\cdot\cdot, A_{n-1}, A_0y_0(x)+ A_1y_1(x)+ \cdot\cdot\cdot+ A_{n-1}y_{n-1}(x)= 0- that is, is equal to 0 for all x. Taking x= 0 we must have A_0(1)+ A_1(0)+ \cdot\cdot\cdot+ A_{n-1}(0)= A_0= 0. Since that linear combination is 0 for all x, it is a constant and its derivative is 0 for all x. That is A_0y_0'(x)+ A_1y_1'(0)+ \cdot\cdot\cdot+ A_{n-1}y'(0)= 0 for all x. Set x= 0 to see that A_1= 0[/tex]. Continue taking derivatives and setting x= 0 to see that each A_i, in turn, is equal to 0.
<br /> Thanks! it helped a lot!
 
HallsofIvy said:
A basic property of linear homogeneous equations is that the set of solutions forms a vector space. That is, any linear combination of solutions, a_1y_1+ a_2y_2 is again a solution. One can show that, for an nth order homogeneous differential equation, this vector space has dimension n. That is, there exist n independent solutions such that any solution can be written in terms of those n solutions.

To prove that, consider the differential equation a_n(x)y^{(n)}(x)+ a_{n-1}y^{(n-1)}(x)+ \cdot\cdot\cdot+ a_2y&#039;&#039;(x)+ a_1y&#039;(x)+ a_0y(x)= 0.
with the following initial values:

I) y(0)= 1, y&#039;(0)= y&#039;&#039;(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
By the fundamental "existence and uniqueness" theorem for initial value problems, there exist a unique function, y_0(x), satisfying the differential equation and those initial conditions.

II) y(0)= 0, y&#039;(0)= 1, y&#039;&#039;(0)= y&#039;&#039;&#039;(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
Again, there exist a unique solution, y_1(x), satisfying the differential equation and those initial conditions.

III) y(0)= y&#039;(0)= 0, y&#039;&#039;(0)= 1, y&#039;&#039;&#039;(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0
Again, there exist a unique solution, y_2(x), satisfying the differential equation and those initial conditions.

Continue that, shifting the "= 1" through the derivatives until we get to
X) y(0)= y&#039;(0)= \cdot\cdot\cdot= y^{(n-1)}(0)= 0, y^{(n-1)}(0)= 1.

First, any solution to the differential equation can written as a linear combination of those n solutions:
Suppose y(x) is a solution to the differential equation. Let A_0= y(0), A_1= y&#039;(0), etc. until A_{n-1}= y^{(n-1)}(0).
Then y(x)= A_0y_0(x)+ A_1y_1(x)+ \cdot\cdot\cdot+ A_{n-1}y_{n-1}(x). That can be shown by evaluating both sides, and their derivatives, at x= 0.

Further, that set of n solutions are independent. To see that suppose that, for some numbers, A_0, A_1, \cdot\cdot\cdot, A_{n-1}, A_0y_0(x)+ A_1y_1(x)+ \cdot\cdot\cdot+ A_{n-1}y_{n-1}(x)= 0- that is, is equal to 0 for all x. Taking x= 0 we must have A_0(1)+ A_1(0)+ \cdot\cdot\cdot+ A_{n-1}(0)= A_0= 0. Since that linear combination is 0 for all x, it is a constant and its derivative is 0 for all x. That is A_0y_0&#039;(x)+ A_1y_1&#039;(0)+ \cdot\cdot\cdot+ A_{n-1}y&#039;(0)= 0 for all x. Set x= 0 to see that A_1= 0[/tex]. Continue taking derivatives and setting x= 0 to see that each A_i, in turn, is equal to 0.
<script class="js-extraPhrases" type="application/json"> { "lightbox_close": "Close", "lightbox_next": "Next", "lightbox_previous": "Previous", "lightbox_error": "The requested content cannot be loaded. Please try again later.", "lightbox_start_slideshow": "Start slideshow", "lightbox_stop_slideshow": "Stop slideshow", "lightbox_full_screen": "Full screen", "lightbox_thumbnails": "Thumbnails", "lightbox_download": "Download", "lightbox_share": "Share", "lightbox_zoom": "Zoom", "lightbox_new_window": "New window", "lightbox_toggle_sidebar": "Toggle sidebar" } </script> <div class="bbImageWrapper js-lbImage" title="diff14.png" data-src="https://www.physicsforums.com/attachments/diff14-png.179218/" data-lb-sidebar-href="" data-lb-caption-extra-html="" data-single-image="1"> <img src="https://www.physicsforums.com/attachments/diff14-png.179218/" data-url="" class="bbImage" data-zoom-target="1" style="" alt="diff14.png" title="diff14.png" width="414" height="54" loading="lazy" decoding="async" /> </div> (5)<br /> <div class="bbImageWrapper js-lbImage" title="diff27.png" data-src="https://www.physicsforums.com/attachments/diff27-png.179219/" data-lb-sidebar-href="" data-lb-caption-extra-html="" data-single-image="1"> <img src="https://www.physicsforums.com/attachments/diff27-png.179219/" data-url="" class="bbImage" data-zoom-target="1" style="" alt="diff27.png" title="diff27.png" width="527" height="184" loading="lazy" decoding="async" /> </div> (17)<br /> <br /> - using picard&#039;s iteration in vector form, to prove nth order linear ODE&#039;s existence &amp; uniqueness.<br /> <br /> ex<br /> <div class="bbImageWrapper js-lbImage" title="diff31.png" data-src="https://www.physicsforums.com/attachments/diff31-png.179220/" data-lb-sidebar-href="" data-lb-caption-extra-html="" data-single-image="1"> <img src="https://www.physicsforums.com/attachments/diff31-png.179220/" data-url="" class="bbImage" data-zoom-target="1" style="" alt="diff31.png" title="diff31.png" width="360" height="79" loading="lazy" decoding="async" /> </div> (21)<br /> <br /> <div class="bbImageWrapper js-lbImage" title="diff32.png" data-src="https://www.physicsforums.com/attachments/diff32-png.179221/" data-lb-sidebar-href="" data-lb-caption-extra-html="" data-single-image="1"> <img src="https://www.physicsforums.com/attachments/diff32-png.179221/" data-url="" class="bbImage" data-zoom-target="1" style="" alt="diff32.png" title="diff32.png" width="425" height="443" loading="lazy" decoding="async" /> </div> (22)<br /> <br /> (<a href="http://ghebook.blogspot.ca/2011/10/differential-equation.html" target="_blank" class="link link--external" rel="nofollow ugc noopener">http://ghebook.blogspot.ca/2011/10/differential-equation.html</a>)Hi, I actually did picard&#039;s iteration and found that without n initial conditions, nth order linear ODE will have n number of constants as we assume initial conditions are some arbitrary constants.<br /> <br /> since function spaces are vector spaces, solutions span n dimensional vector space (not very sure of this)<br /> <br /> If we do picard&#039;s iteration,<br /> <br /> y<sup>(n-1)</sup>=y<sub>0</sub><sup>(n-1)</sup>+∫y<sup>(n)</sup>dx<br /> y<sup>(n-2)</sup>=y<sub>0</sub><sup>(n-2)</sup>+∫y<sup>(n-1)</sup>dx<br /> =y<sup>(n-2)</sup>=y<sub>0</sub><sup>(n-2)</sup>+∫y<sub>0</sub><sup>(n-1)</sup>+∫y<sup>(n)</sup>dx<br /> .<br /> .<br /> .<br /> iteration goes on and on until the error is sufficiently decreased.if we assume each initial conditions are some constants, we will eventually sort out the solution function y w.r.t constants after sufficient number of iteration is done. Then it will look like,<br /> <br /> y (c<sub>0</sub> c<sub>1</sub> c<sub>2</sub> - - - - - - )[y1 y2 y3 y4 y5 - - - - - - ] &lt;-- (should be vertical)which is in the form of y = c1y1+c2y2+c3y3 . . .<br /> <br /> and so on<br /> is this correct?<br /> If not, how can we show that solution space has n number of basis?---------------------------------------------------------------------------------------------------------------------------<br /> <br /> Also, I have two questions about Strum Liouville 2nd order ODE.1. if we look at the Strum-Liouville 2nd order ODEs, there is an eigenvalue term within the equation. it seems like we are adding one more constant in the equation, which imposes a restriction to find a solution (n+1 constants with n conditions).<br /> <br /> [m(x)y&#039;]+[λr(x)-q(x)]y<br /> =m(x)[y&#039;&#039;+P(x)y&#039;+Q(x)y]<br /> =m(x)[y&#039;&#039;+P(x)y&#039;+(λr(x)-q(x))]<br /> =0<br /> <br /> ∴ [y&#039;&#039;+P(x)y&#039;+(λr(x)-q(x))]=0<br /> <br /> if we do Picard&#039;s iteration, then we have one more constant λ along with n constants..2. I don&#039;t understand how eigenvalue directly influence the solutions of 2nd order ODEThanks always, Your answers help me a lot!
 
Last edited:

Similar threads

  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
5K
  • · Replies 2 ·
Replies
2
Views
2K
  • · Replies 3 ·
Replies
3
Views
3K
  • · Replies 1 ·
Replies
1
Views
2K
  • · Replies 2 ·
Replies
2
Views
3K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 2 ·
Replies
2
Views
2K
Replies
3
Views
3K