This preview shows pages 1–3. Sign up to view the full content.
This preview has intentionally blurred sections. Sign up to view the full version.View Full Document
Unformatted text preview: 14:440:127 Introduction to Computers for Engineers Notes for Lecture 11 Rutgers University, Fall 2008 Instructor- Blase E. Ur Were going to begin with a corrected version of a section from Lecture 10 with 2 stupid and serious mistakes I had left in there, but noticed and corrected as I was giving the lecture: 1 Correction from last lecture: Fitting To More Compli- cated Functions Oftentimes in engineering, youll want to find equations for functions that are not polynomial. However, youll sometimes be able to transform these functions into polynomials: 1.0.1 Power Functions A function such as y = bx m can be called a power function since x is raised to a power. Note that this is a polynomial, but you have no idea what degree it is, so it might be hard to choose which degree polynomial to specify for polyfit . Note that if you plot this function on a log-log plot using the loglog command, the data will appear in a straight line. Now, the key is that if you take the logarithm of both sides of this equation, youll get log ( y ) = log ( bx m ). Using log properties, you can rewrite this as log ( y ) = log ( b )+ log ( x m ), and then rewrite it as log ( y ) = m * log ( x ) + log ( b ). This has the form j = m * h + k if you define j = log ( y ), h = log ( x ), and k = log ( b ). This is perfectly analogous to y = mx + b , which is a linear equation. Now, find the coefficients using polyfit(log(x),log(y),1) , and call these coeffients [m k]. Now, go- ing back to the original equation, you can realize that youve already found m . To find b , you can recall that k = log ( b ), so b = e k . You can now write an equation for your data as y = exp ( coeffs (2)) * x coeffs (1) . Thus, youve used a transformation, fit the log of the power function to a line, and then reversed your transformation to find the original coefficients. 1.0.2 Exponential Functions Lets say youre given an exponential function, such as y = b ( e ) mx . This function would be displayed as a straight line on a semilogy plot. This can be rewritten as log ( y ) = mx + log ( b ). Now, this has the form j = m * x + k if you set j = log ( y ) and k = log ( b ). Note that you do not have to transform x. Now, find the coefficients using polyfit(x,log(y),1) , and call these coeffients [m k]. Again, youve already found m from the original equation, and you can find b by recalling that i = log ( b ), so b = e k . You can now write an equation for your data as y = exp ( coeffs (2)) * ( e ) coeffs (1) * x . 2 Recursion Theres a very interesting technique in computer programming called recursion, in which you write a function that calls itself . Eventually, this string of functions calling itself stops at some base case , and the answers work their way back up. Of course, in order to understand recursion, you must understand recursion. (Thats a joke)....
View Full Document
This note was uploaded on 11/09/2011 for the course ENGINEERIN 127 taught by Professor Finch during the Fall '08 term at Rutgers.
- Fall '08