1MathematicalPreliminaries.ppt

0 Preface Purpose: To give an introduction to modern approximation techniques its realization; To explain how, why ,and when (3w)they can be expected to work; To provide a form basis for future study of numerical analysis and scientific computing. Software: Matlab; Mathematica 1 Mathematics Preliminaries 1.1 Review of Calculus 1.2 Round off Errors and Computer Arithmetic 1.3 Algorithms and Convergence 1.4 Numerical Soft ware 1.1 Review of Calculus Limit of a sequence; Continuity; Differentiable, Derivative; Integral; Rolle’s Theorem. 1.1 Review of Calculus (Taylor’s Theorem) 1.1 Review of Calculus 1.2 Resources Classics of Error Round off Errors and Computer Arithmetic The arithmetic performed by a computer is different from the arithmetic in our algebra and calculus course, here is numerical method. Round off Errors(舍入) occurs because the arithmetic performed in a machine involves numbers with only a finite number of digits. The mantissa(尾数)is chopped (or round up/down)---- ceiling/ floor ; Underflow, Overflow Definition1.15 If p* is an approximation to p ,the absolute error is | p- p* |, the relative error is | p- p* | / | p| provided that p ≠ 0. The absolute error can be misleading and the relative error is more meaningful since the relative error takes into consideration the size of the value. 1.3 Algorithms and Convergence 1.4 Numerical Software Matlab; Mathematica; Maple etc. 1.5 函数计算的误差估计 误差危害及其防止 避免两个相近的数相减；绝对值小的数作除数 办法：改变计算公式；增加字长。 注意运算次序，防止大数”吃掉”小数，应按绝对值由小到大的次序运算。 误差危害及其防止 计算要讲效率，减少误差产生的机会 误差危害及其防止 误差危害及其防止 注意初始误差的恶性传播 病态问题和条件数 病态问题和条件数 Ch2 Solutions of Equations in One Variable 2.1 The Bisection Methods Algorithm of Bisection Convergence Theorem Suppose that f∈C[a,b] and f(a)?f(b)0,then the Bisection methods generates a sequence pn approximating a zero p of f with: | pn -p|≤ (b-a)/2n Proof: | bn - an |=(b-a)/2n-1 ， pn , p ∈ (an , bn) 2.2 Fixed-Point Iteration (Example) Theorem