demonewtonsys.m

数值方法和MATLAB实现与应用.zip

function x = demoNewtonSys(maxit,x0)
% demoNewtonSys   Solve a 2-by-2 nonlinear system by Newton's method
%                 The system is
%                             1.4*x1 - x2 = 0.6
%                      x1^2 - 1.6*x1 - x2 = 4.6
%
% Synopsis:  x = demoNewtonSys
%            x = demoNewtonSys(maxit)
%            x = demoNewtonSys(maxit,x0)
%
% Input:   maxit = (optional) max number of iterations.  Default: maxit = 5
%          x0 = (optional) initial guess at solution.  Default:  x0 = [0; 0]
%
% Output:  x = estimate of solution after maxit iterations

if nargin<1,  maxit=5;          end
if nargin<2,  x0 = zeros(2,1);  end

% --- Coefficients for the case of two distinct solutions
alpha =  1.4;  bbeta = -0.6;  sigma = -1.6;  tau = -4.6;

x = x0;   f = zeros(size(x));

fprintf('\n   k       x(1)        x(2)      norm(f)      norm(dx)\n');
for k = 1:maxit
   f(1) = alpha*x(1) - x(2) + bbeta;
   f(2) = x(1)^2 + sigma*x(1) - x(2) + tau;
   J = [ alpha  -1; (2*x(1)+sigma)  -1 ];
   dx = -J\f;
   fprintf('%4d   %9.5f   %9.5f   %10.2e   %10.2e\n',...
            k-1,x(1),x(2),norm(f),norm(dx));
   x = x + dx;
end
fprintf('%4d   %9.5f   %9.5f\n',k,x(1),x(2));