How to solve x*a^x=b with math reasoning?

  • Thread starter Thread starter raul_l
  • Start date Start date
Click For Summary

Homework Help Overview

The discussion revolves around solving the equation x*a^x=b, where a and b are considered unknown constants. Participants explore the nature of the problem and the methods applicable for finding solutions.

Discussion Character

  • Exploratory, Mathematical reasoning, Problem interpretation

Approaches and Questions Raised

  • Some participants note the general impossibility of finding an exact solution and suggest numerical methods for approximation. Others discuss the application of Newton's method, including the selection of an initial guess and the iterative process involved. Additionally, one participant proposes a transformation of the equation and examines the implications of the constants a and b being positive.

Discussion Status

The conversation includes various approaches to the problem, with some participants sharing insights on numerical methods and derivatives. There is acknowledgment of the complexity of the problem, particularly when considering different ranges for the constant a. While some guidance has been offered regarding methods, there is no explicit consensus on a singular approach.

Contextual Notes

Participants emphasize the need for a and b to be greater than zero and discuss the implications of different values for a on the existence of roots. The discussion reflects on the challenges posed by the problem's structure and the necessity of numerical methods in certain cases.

raul_l
Messages
105
Reaction score
0
x*a^x=b

how to solve this?
 
Physics news on Phys.org
You can't, in general, solve it exactly. Use a numerical method to approximate the roots.
 
This is a good candidate for Newton's method.
You first choose an x0 arbitrarily. You can graph it first, then choose an x0 wisely so it's near the roots.
Then use the formula:
x_{n + 1} = x_n - \frac{f(x_n)}{f'(x_n)}, and let n increase without bound to obtain the solution, i.e:
x = \lim_{n \rightarrow \infty} x_n.
Can you get this? :)
 
Yes.
And it works! :)
I didn't know about Newton's method before.
 
with math reasoning

Suppose a and b are unknown constants
Let y=xa^x and hence y=b
Take a ln of bothe sides leading to lny=lnx+xlna
We understand that x,a,b must be > 0

Taking a derivative of y gives us
\frac{dy}{dx}=(\frac{1}{x}+lna)xa^x
Now we find that x=0 (omitted), \frac{-1}{lna}

Next, we draw a table to check signs of \frac{dy}{dx}, but before that we check a
1. if 0<a<1
Look at the table and mark for sign (+/-), then check for y to compare with y=b (a.straight.line), which means you need to reason the value of b for where the root(s) exist.
2. if a>1
Do the same to find out root domain

Now things become easier when you know concrete constant a, b. just put them inthere to find a root. This way looks crary though :biggrin: but sovable domain can be understood
 
Last edited:

Similar threads

Replies
18
Views
3K
Replies
3
Views
3K
  • · Replies 5 ·
Replies
5
Views
2K
  • · Replies 25 ·
Replies
25
Views
3K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 19 ·
Replies
19
Views
3K
Replies
6
Views
3K
  • · Replies 10 ·
Replies
10
Views
2K
  • · Replies 7 ·
Replies
7
Views
2K
  • · Replies 17 ·
Replies
17
Views
3K