PowerPoints organized by Dr. Michael R. Gustafson II, Duke University

Size: px

Start display at page:

Download "PowerPoints organized by Dr. Michael R. Gustafson II, Duke University"

Willis Lucas
5 years ago
Views:

1 Part 5 Chapter 19 Numerical Differentiation PowerPoints organized by Dr. Michael R. Gustafson II, Duke University All images copyright The McGraw-Hill Companies, Inc. Permission required for reproduction or display.

2 Chapter Objectives Understanding the application of high-accuracy numerical differentiation formulas for equispaced data. Knowing how to evaluate derivatives for unequally spaced data. Understanding how Richardson extrapolation is applied for numerical differentiation. Recognizing the sensitivity of numerical differentiation to data error. Knowing how to evaluate derivatives in MATLAB with the diff and gradient functions. Knowing how to generate contour plots and vector fields with MATLAB.

3 The problem Velocity of a free-falling bungee jumper as a function of time: gm gc d v( t) tanh t c d m In Chap 17, we used calculus to integrate this equation to determine vertical distance at time t: m gc d z( t) ln cosh t c d m

4 Differentiation Problem The reverse problem is to find the velocity based on the jumper s position as a function of time: vt () Or compute acceleration: at () dv t dt dz() t dt 2 ( ) d z( t) In this case, we could analytically differentiate to obtain: dt gc d m 2 a( t) g sech t 2

5 Why would we want to differentiate numerically? Some functions may be difficult or impossible to differentiate analytically. If we measured the jumper s position at various times during the fall, we d have tabulated discrete values of z(t). This is a very common case for laboratory data where several differential equations may be used as a model describing the collected data over time.

6 Differentiation The mathematical definition of a derivative begins with a difference approximation: and as x is allowed to approach zero, the difference becomes a derivative: y x f x x i f x i x dy dx lim x 0 f x i x f x i x

7 High-Accuracy Differentiation Formulas Taylor series expansion can be used to generate high-accuracy formulas for derivatives by using linear algebra to combine the expansion around several points. Three categories for the formula include forward finite-difference, backward finitedifference, and centered finite-difference.

8 Forward Finite-Difference

9 Backward Finite-Difference

10 Centered Finite-Difference

11 Richardson Extrapolation As with integration, the Richardson extrapolation can be used to combine two lower-accuracy estimates of the derivative to produce a higher-accuracy estimate. For the cases where there are two O(h 2 ) estimates and the interval is halved (h 2 =h 1 /2), an improved O(h 4 ) estimate may be formed using: For the cases where there are two O(h 4 ) estimates and the interval is halved (h 2 =h 1 /2), an improved O(h 6 ) estimate may be formed using: D 4 3 D(h 2) 1 3 D(h 1) For the cases where there are two O(h 6 ) estimates and the interval is halved (h 2 =h 1 /2), an improved O(h 8 ) estimate may be formed using: D D(h 2) 1 15 D(h 1) D D(h 2) 1 63 D(h 1)

12 Example of Richardson Extrapolation f(x) = -0.1x x 3 0.5x x +1.2 Estimate the first derivative at x = 0.5 using step sizes of h 1 =0.5 and h 2 = Solution: The exact solution is Use centered differences: f(1) f(0) ( 1) D(0.5) 1.0 = 9.6% 2(0.5)

13 f(.75) f(.25) ( ) D(0.25) = 2.4% 2(0.25) Use the first Richardson formula to obtain an improved estimate: 4 1 D ( ) ( 1) This estimate is exact in this case because the polynomial is 4 th order, and the error for the Richardson formula has error O(h 4 )

14 Unequally Spaced Data One way to calculated derivatives of unequally spaced data is to determine a polynomial fit and take its derivative at a point. As an example, using a second-order Lagrange polynomial to fit three points and taking its derivative yields: f x f x 0 2x x 1 x 2 2x x 0 x 2 2x x 0 x 1 x 0 x 1 x 0 x 2 f x 1 x 1 x 0 x 1 x 2 f x 2 x 2 x 0 x 2 x 1

Derivatives and Integrals for Data with Errors A shortcoming of numerical differentiation is that it tends to amplify errors in data, whereas integration tends to

15 Derivatives and Integrals for Data with Errors A shortcoming of numerical differentiation is that it tends to amplify errors in data, whereas integration tends to smooth data errors. One approach for taking derivatives of data with errors is to fit a smooth, differentiable function to the data and take the derivative of the function.

16 Numerical Differentiation with MATLAB MATLAB has two built-in functions to help take derivatives, diff and gradient: diff(x) Returns the difference between adjacent elements in x diff(y)./diff(x) Returns the difference between adjacent values in y divided by the corresponding difference in adjacent values of x

17 Numerical Differentiation with MATLAB fx = gradient(f, h) Determines the derivative of the data in f at each of the points. The program uses forward difference for the first point, backward difference for the last point, and centered difference for the interior points. h is the spacing between points; if omitted h=1. The major advantage of gradient over diff is gradient s result is the same size as the original data. Gradient can also be used to find partial derivatives for matrices: [fx, fy] = gradient(f, h)

18 Visualization MATLAB can generate contour plots of functions as well as vector fields. Assuming x and y represent a meshgrid of x and y values and z represents a function of x and y, contour(x, y, z) can be used to generate a contour plot [fx, fy]=gradient(z,h) can be used to generate partial derivatives and quiver(x, y, fx, fy) can be used to generate vector fields

PowerPoints organized by Dr. Michael R. Gustafson II, Duke University

Part 5 Chapter 21 Numerical Differentiation PowerPoints organized by Dr. Michael R. Gustafson II, Duke University 1 All images copyright The McGraw-Hill Companies, Inc. Permission required for reproduction