Learning Numerical Analysis for Physics Problems

In summary, you should have a background in differential equations, multivariable calculus, and linear algebra. The first and last are obviously prerequisites to numerical methods for differential equations.
  • #1
pc2-brazil
205
3
Good afternoon,

I'm willing to study (teach myself) http://en.wikipedia.org/wiki/Numerical_analysis" , which I find very interesting, and which seems to be very useful for solving Physics problems (specifically, the motion of celestial bodies).
I have a background in Calculus (basically Limits, Derivatives and Integrals), but all my knowledge is self-taught.
My plan is to understand Numerical Analysis, specially Numerical Integration and Numerical Ordinary Differential Equations.
My question is: what background in Mathematics should I have in order to start doing this? In what sequence does it normally appear in courses?
 
Last edited by a moderator:
Physics news on Phys.org
  • #2
Well, usually you have already studied differential equations (DE's), multivariable calculus, and linear algebra. The first and last are obviously prerequisites to numerical methods for DE's.

Start your home-study program with numerical differentiation, since you know some calculus. I recommend Hamming's book "Numerical Methods for Scientists and Engineers" because it is clearly written, slow-paced, and application-oriented rather than math oriented. Integration is presented towards the end, so you may need to flip back to pick up the required introductory material.
 
  • #3
marcusl said:
Well, usually you have already studied differential equations (DE's), multivariable calculus, and linear algebra. The first and last are obviously prerequisites to numerical methods for DE's.

Start your home-study program with numerical differentiation, since you know some calculus. I recommend Hamming's book "Numerical Methods for Scientists and Engineers" because it is clearly written, slow-paced, and application-oriented rather than math oriented. Integration is presented towards the end, so you may need to flip back to pick up the required introductory material.

Thank you for your answer.
I found one exemplar of this book online, in a library here in Brazil, for a very cheap price. But it's a 1962 edition. Is it good?
Another question: What is the relation between Numerical Analysis and Perturbation Theory? Is the latter a subset of the former?
 

FAQ: Learning Numerical Analysis for Physics Problems

What is numerical analysis?

Numerical analysis is a branch of mathematics that deals with developing and using numerical methods to solve problems in various fields, including physics. It involves using algorithms and computer programs to approximate solutions to mathematical problems.

Why is numerical analysis important in physics?

Numerical analysis is important in physics because it allows us to solve complex mathematical equations and models that cannot be solved analytically. It also helps us to validate and verify theoretical predictions by providing numerical solutions that can be compared to experimental data.

What are some common numerical methods used in physics?

Some common numerical methods used in physics include the finite difference method, finite element method, and Monte Carlo method. These methods involve dividing a problem into smaller, more manageable parts and using iterative calculations to approximate the solution.

How can I learn numerical analysis for physics problems?

The best way to learn numerical analysis for physics problems is through practice and hands-on experience. Start by learning the basic principles and concepts, then move on to implementing numerical methods in computer programs. You can also take courses or read books on the subject to deepen your understanding.

What are some challenges of using numerical analysis in physics?

Some challenges of using numerical analysis in physics include round-off errors, stability issues, and the need for a large amount of computational resources. It is also important to choose the appropriate numerical method for a specific problem, as some methods may be more accurate or efficient than others.

Back
Top