Badly conditioned equations - how can I centre and scale properly

26 Ansichten (letzte 30 Tage)
After writing my own best fit line. I can see that the results are pretty great, however the equation is apparantly badly scaled.
%
alumina_610 = [
25 90
1200 120
1300 150
1350 310
1400 450
1500 790 ];
%%
% % Nextel 610
figure1 = figure;
%
hold on
%
x = (alumina_610(:,1));
y = (alumina_610(:,2));
[xData, yData] = prepareCurveData( x, y );
% Set up fittype and options.
ft = fittype( {'x^12', '100'}, 'independent', 'x', 'dependent', 'y', 'coefficients', {'a', 'b'} );
%
[fitresult, gof] = fit( xData, yData, ft );
Warning: Equation is badly conditioned. Remove repeated data points or try centering and scaling.
% plot fit:
x_fit = linspace(1,1600,100);
y_fit = fitresult(x_fit);
plot(x_fit,y_fit,':r')
% Label axes
xlabel( 'x', 'Interpreter','latex' );
ylabel( 'y', 'Interpreter','latex' );
plot(alumina_610(:,1),alumina_610(:,2),'ro', ...
'MarkerFaceColor','r')
ylim([0 900])
ylabel('Grainsize [nm]','Color',[ 0 0 0 ])
xlim([0 1600])
%
xlabel('Temperature$^{o}$C', 'Interpreter','latex')
% % %
%
How do I correctly scale a best fit line?
Thanks in advance

Akzeptierte Antwort

Torsten
Torsten am 21 Nov. 2023
Bearbeitet: Torsten am 21 Nov. 2023
Your problem is linear, but fitting data with an independent coordinate up to 1500 by a polynomial of degree 12 is nonsense.
alumina_610 = [
25 90
1200 120
1300 150
1350 310
1400 450
1500 790 ];
M = [alumina_610(:,1).^12 1e38*ones(size(alumina_610,1),1)];
rank(M)
ans = 2
b = alumina_610(:,2);
p = M\b
p = 2×1
1.0e-35 * 0.5620 0.0792
hold on
plot(alumina_610(:,1),alumina_610(:,2),'o')
x = linspace(alumina_610(1,1),alumina_610(end,1),100);
plot(x,p(1)*x.^12+p(2)*1e38)
hold off

Weitere Antworten (0)

Kategorien

Mehr zu Linear and Nonlinear Regression finden Sie in Help Center und File Exchange

Produkte


Version

R2022b

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by