problem in symbolic matrix multiplication

Hi, I am performing simple multiplication on two matrices. However getting error while doing so. I am not able to understand what is error and where I am going wrong. Therefore I am not able to correct error. Can someone provide any ideas or insights? here is my code:

clear all;
clc;

syms xlc
%sr=sym([]);
sr=zeros(4,1);

B=[ 0.64*xlc - 1.2,  1.6 - 1.28*xlc, 0.64*xlc - 0.4; 0.64*xlc - 2.8,  4.8 - 1.28*xlc, 0.64*xlc - 2.0; 0.64*xlc - 4.4,  8.0 - 1.28*xlc, 0.64*xlc - 3.6; 0.64*xlc - 6.0, 11.2 - 1.28*xlc, 0.64*xlc - 5.2];

d=1.0e-03 *[0.000000000095833;0.058593750095833;   0.109375000095833;   0.152343750095834;   0.187500000095834;   0.214843750095834;   0.234375000095834; 0.246093750095834; 0.250000000095834];

e_n=[1,2,     3;     3,     4,     5;     5,     6,     7;     7,     8,     9];

   for i=1:1:4
      for j=1:1:3
          I = e_n(i,j);
          sr(i)=sr(i)+(B(i,j)*d(j));
      end
  end

What I am doing: I want to multiply

sr(1)=B(1,1)*d(1)+B(1,2)*d(2)+B(1,3)*d(3)
sr(2)=B(2,1)*d(3)+B(2,2)*d(4)+B(2,3)*d(5)

and so on. But matlab is giving errors. Any help or ideas are welcome. Thanks.

Answer 1

[Walter Roberson]

You initialize sr as empty, so sr(i) does not exist yet to be indexed by "i" for each different "i" value.

sr = sym(zeros(4,1));

Answer 2

[Andrei Bobrov]

syms xlc
B=[ 0.64*xlc - 1.2,  1.6 - 1.28*xlc, 0.64*xlc - 0.4; 0.64*xlc - 2.8,  4.8 - 1.28*xlc, 0.64*xlc - 2.0; 0.64*xlc - 4.4,  8.0 - 1.28*xlc, 0.64*xlc - 3.6; 0.64*xlc - 6.0, 11.2 - 1.28*xlc, 0.64*xlc - 5.2];
d=1.0e-03 *[0.000000000095833;0.058593750095833;   0.109375000095833;   0.152343750095834;   0.187500000095834;   0.214843750095834;   0.234375000095834; 0.246093750095834; 0.250000000095834];

ii = bsxfun(@plus,1:size(B,2),2*(0:size(B,1)-1)');
sr = sum(B.*d(ii),2);

[Andrew]

Thank you very much for your help. However to be honest, I did not understand your code. I am a beginner of matlab and your code, at first glance, seems complicated. This may be the way that professionals or experts use, however I am not one among them. But, I am trying to find out what you have done. Again, thanks!

Answer 3

[Roger Stafford]

It isn't clear whether you are receiving a matlab error message or are getting erroneous results. If it is the latter the line in the inner for-loop should be changed to:

 sr(i)=sr(i)+(B(i,j)*d(I));

[Andrew]

Thanks Roger Stafford for your comment. Even if I change inner loop line to

sr(i)=sr(i)+(B(i,j)*d(I));

there is error. When I just put these lines

sr(i)+(B(i,j)*d(I))

for i=I=1, matlab gives proper result. But as soon as I do

sr(1) = sr(i)+(B(i,j)*d(I))

for i=I=1, matlab gives error.

However, I tried to solve this problem with another way,

clear all;
clc;
syms xlc
sr=sym([]);  %MODIFICATION
%sr=zeros(4,1);
B=[ 0.64*xlc - 1.2,  1.6 - 1.28*xlc, 0.64*xlc - 0.4; 0.64*xlc - 2.8,  4.8 - 1.28*xlc, 0.64*xlc - 2.0; 0.64*xlc - 4.4,  8.0 - 1.28*xlc, 0.64*xlc - 3.6; 0.64*xlc - 6.0, 11.2 - 1.28*xlc, 0.64*xlc - 5.2];
d=1.0e-03 *[0.000000000095833;0.058593750095833;   0.109375000095833;   0.152343750095834;   0.187500000095834;   0.214843750095834;   0.234375000095834; 0.246093750095834; 0.250000000095834];
e_n=[1,2,     3;     3,     4,     5;     5,     6,     7;     7,     8,     9];
   for i=1:1:4
      for j=1:1:3
          I = e_n(i,j);
          sr(i)=sr(i)+(B(i,j)*d(I));
      end
  end

I made sr a symbolic array. And there is another error "Index exceeds matrix dimensions." Can someone comment on this? Thanks.

[Andrew]

Making sr variable symbolic or double does not seem to make difference. Any ideas or suggestions?

[Walter Roberson]

What error message is it that you are getting for the first situation?