How to find parameters of the following coefficient based on optimization tool?
    7 visualizaciones (últimos 30 días)
  
       Mostrar comentarios más antiguos
    
    Parham Babakhani Dehkordi
 el 11 de Abr. de 2016
  
    
    
    
    
    Respondida: Parham Babakhani Dehkordi
 el 12 de Abr. de 2016
            Hi everyone, I have the following input and function, my purpose is to calculate the value of a,m and n which are scalars using curve fitting in matlab. I have:
ew=[0.679 0.613 0.613 0.501 0.497 0.50 0.389 0.386 0.386 0.391 0.390 0.288 0.288 0.288 0.289 0.289]; Hw=[0.757 0.669 0.71 0.577 0.57 0.60 0.50 0.55 0.52 0.50 0.476 0.353 0.388 0.379 0.431 0.371]; k=[0.217 0.217 0.217 0.163 0.163 0.163 0.131 0.131 0.131 0.131 0.131 0.109 0.109 0.109 0.109 0.109];
Hw=[ew*(1+0.36(1-ew))]*C; C=exp[-a*(k^m)*(1-ew)^n];
any help would be appreciated.
1 comentario
  Alan Weiss
    
      
 el 11 de Abr. de 2016
				Please format your question using the {} Code button so that we might read it more easily.
Alan Weiss
MATLAB mathematical toolbox documentation
Respuestas (2)
  Roger Stafford
      
      
 el 12 de Abr. de 2016
        Define the following sum of squares of differences:
   ew=[0.679 0.613 0.613 0.501 0.497 0.50  0.389 0.386 0.386 0.391 0.390 0.288 0.288 0.288 0.289 0.289]; 
   Hw=[0.757 0.669 0.71  0.577 0.57  0.60  0.50  0.55  0.52  0.50  0.476 0.353 0.388 0.379 0.431 0.371]; 
   k= [0.217 0.217 0.217 0.163 0.163 0.163 0.131 0.131 0.131 0.131 0.131 0.109 0.109 0.109 0.109 0.109];
   function s = myfun(a,m,n)
     s = sum((Hw-ew.*(1+0.36(1-ew)).*exp[-a*(k.^m).*(1-ew).^n)).^2);
   return
and then use that subfunction for optimization of a, m, and n using 'fminsearch' or 'fminunc'. Their descriptions can be found at:
   http://www.mathworks.com/help/matlab/ref/fminsearch.html
   http://www.mathworks.com/help/optim/ug/fminunc.html
0 comentarios
Ver también
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!


