Non - Singleton Error Message

I'M TRYING TO RUN THE BELOW CODE,

aa = input('Please Input Lift Curve Slope in rad >');

T1 = (TTheta - RTheta)/((D/2)-Hr);
AR = D/Cr;
K = 1/(pi*e*AR);

WB = waitbar(0,'Please Wait.. Calculation In Progress');

for i = 1:3

      waitbar(i/MaxV,WB)

      for j = 1:2
          omega(j) = 2*pi*j;

          a = 0.2;
          finish = 0;
          add = 1;
          while finish == 0

              for k = 1:size(r1,2)
                  % Local Quantities Calculation
                  Loc(k) = r1(k);
                  VT(j,k) = omega(j)*Loc(k);
                  W2(i,j,k) = sqrt((i*(1+a))^2 + omega(j)^2*Loc(k)^2);
                  Phi(i,j,k) = atan((i*(1+a))/VT(j,k));
                  Alpha(i,j,k) = RTheta + T1*Loc(k) - rad2deg(Phi(i,j,k));

                  % Local Coefficient Calculations
                  CL(i,j,k) = aa*deg2rad(Alpha(i,j,k));
                  CD(i,j,k) = Cdo + K*CL(i,j,k)^2;

                  % Local Force Calculations
                  Lift(i,j,k) = 0.5*ro*W2(i,j,k)^2*Cr*aa*deg2rad(Alpha(i,j,k));
                  Drag(i,j,k) = 0.5*ro*W2(i,j,k)^2*Cr*CD(i,j,k);
                  Dt(i,j,k) = (Lift(i,j,k)*cos(Phi(i,j,k)))-(Drag(i,j,k)*sin(Phi(i,j,k)));
                  Dq(i,j,k) = (Lift(i,j,k)*sin(Phi(i,j,k)))+(Drag(i,j,k)*cos(Phi(i,j,k)));
                  Dpr(i,j,k) = Dq(i,j,k)*VT(j,k);

                  % Local Efficiency Calculation
                  eff(i,j,k) = Dt(i,j,k)*(i*(1 + a))/Dpr(i,j,k);
              end

              % Efficiency calculation
              Thrust = sum(Dt,3)*rStep;
              da = Thrust/(2*ro*Ar*i^2);
              anew = (a + da)/2;

              if (a - anew)<1e-10
                  finish = 1;
              end

              a = anew;
              add = add + 1;

              if add > 500
                  finish = 1;
              end
          end
          Torque = sum(Dq,3)*rStep;
          Pr = sum(Dpr,3)*rStep;
          Eff = Thrust.*repmat((MinV:size(Thrust,1))',1,size(Thrust,2))./Pr;
          AdvRatio(i,j) = i/(j*D);
      end
  end

BUT UNFORTUNATELY EVERY TIME I RUN THIS CODE I GET THIS (see below) ERROR MESSAGE,

?? Assignment has more non-singleton rhs dimensions than non-singleton
subscripts

Error in ==> BEMMT at 62
                W2(i,j,k) = sqrt((i*(1+a))^2 + omega(j)^2*Loc(k)^2);

COULD SOMEONE CORRECT THIS FOR ME.

FURTHERMORE, ABOVE CODE IS A MODIFICATION OF THE BELOW CODE, WHERE I HAVE NESTED A WHILE LOOP IN A FOR LOOP.

THIS CODE WORKS FINE BUT WHEN I USE THE WHILE COMMAND (code at the top - modified code with while loop) I GET THE ABOVE ERROR.

aa = input('Please Input Lift Curve Slope in rad >');

T1 = (TTheta - RTheta)/((D/2)-Hr);
AR = D/Cr;
K = 1/(pi*e*AR);

WB = waitbar(0,'Please Wait.. Calculation In Progress');

for i = MinV:2:MaxV
    waitbar(i/MaxV,WB)

      for j = 1:4:40
          omega(j) = 2*pi*j;

          for k = 1:size(r1,2)
              % Local Quantities Calculation
              Loc(k)=r1(k);
              VT(k) = omega(j)*Loc(k);
              W2(i,j,k) = sqrt(i^2 + omega(j)^2*Loc(k)^2);
              Phi(i,j,k) = atan(i/(omega(j)*Loc(k)));
              Alpha(i,j,k) = RTheta + T1*Loc(k) - rad2deg(Phi(i,j,k));

              % Local Coefficient Calculations
              CL(i,j,k) = aa*deg2rad(Alpha(i,j,k));
              CD(i,j,k) = Cdo + K*CL(i,j,k)^2;

              % Local Force Calculations
              Lift(i,j,k) = 0.5*ro*W2(i,j,k)^2*Cr*aa*deg2rad(Alpha(i,j,k));
              Drag(i,j,k) = 0.5*ro*W2(i,j,k)^2*Cr*CD(i,j,k);
              Dt(i,j,k) = (Lift(i,j,k)*cos(Phi(i,j,k)))-(Drag(i,j,k)*sin(Phi(i,j,k)));
              Dq(i,j,k) = (Lift(i,j,k)*sin(Phi(i,j,k)))+(Drag(i,j,k)*cos(Phi(i,j,k)));
              Dpr(i,j,k) = Dq(i,j,k)*omega(j)*Loc(k);

              % Local Efficiency Calculation
              eff(i,j,k) = Dt(i,j,k)*i/Dpr(i,j,k);
          end
          % Efficiency calculation
          Thrust = sum(Dt,3)*rStep;
          Torque = sum(Dq,3)*rStep;
          Pr = sum(Dpr,3)*rStep;
          Eff = Thrust.*repmat((MinV:size(Thrust,1))',1,size(Thrust,2))./Pr; 

          Adv.Ratio(i,j) = i/(j*D);
      end
  end

[Andrew Newell]

It is difficult to reproduce the error in your code because it is chock full of undefined variables. It would help if you could rewrite it so that it can be run after clearing the workspace.

Answer 1

[Walter Roberson]

You have

              Thrust = sum(Dt,3)*rStep;
              da = Thrust/(2*ro*Ar*i^2);
              anew = (a + da)/2;

From this we deduce that Thrust is going to be a scalar for the first iteration of i and j, a row vector for the first iteration of i and further iterations of j, and a 2D matrix for all subsequent iterations of i. The sum is going to involve positions that get default initialized to 0 when i is greater than 1, as Dt is growing as i and j increase so Dt(2,1:4:40,:) are going to exist and be summed in before they are set to anything other than the 0 they get assigned when a matrix implicitly grows in size.

With Dt being a vector or matrix, da is going to be a vector or matrix, so anew will be as well. And you set a=anew so a will become a vector or matrix. And then you are in trouble with the "if" statement and in the assignment that is blowing up.

I note that "a" not being a scalar was exactly the problem I suggested in your previous version of the question, which you have now deleted. (That was the one where the source line reported as being in error did not match any line of code in what you had posted.)

Answer 2

[Pathiraja]

Yes I had to delete it as it had only the first code not the modified one.

So 'a' should be a scalar not a row vector? Could you be able to suggest a way to get this code right. Apologise first of all I'm new to Matlab.

[Walter Roberson]

Please follow up on what I wrote about Dt growing as you execute, and about you making calculations on locations you have not initialized. When you have your code fixed so that you are always working with the same subset of information, it should be much clearer as to what you want "a" to mean and the sizes that it makes sense for "a" to be.