Warning: Explicit solution could not be found.

>> syms x      
>> total = 0;
>> t2 = 0;
>> t=0;
* >> p = 4.8825*exp(-0.0954*t)+ 12.845*exp(-0.0006211*t);*
>> help = 0;
>> for i = 1:2
*t2 = double(solve('4.8825*exp(-0.0954*t)+ 12.845*exp(-0.0006211*t)-p','t'));*
p = 4.8825*exp(-0.0954*(t2+4.77))+ 12.845*exp(-0.0006211*(t2+4.77));
total = total + double(int(4.8825*(exp(-0.0954*x)+12.845/4.8825*exp(-0.0006211*x)),x,t2,t2+4.77));
t=t+9.4446;
end
Warning: Explicit solution could not be found. 
> In solve at 81

Hei, I got this problem with the above loop. I think the problem occurs when I try to assign a value to p; and then chuck it in solve('...').

I need to solve both these equation simultaniously though, and I have no idea of how to get there otherwise. Anybody care to shed some light on this? All help is greatly appreciated :)

[Bart]

I cried victory too soon :D.

I get this error on the second cycle:

>> syms x

>> total = 0;

>> t=0;

>> t2 = 0;

>> p = 4.8825*exp(-0.0954*t)+ 12.845*exp(-0.0006211*t);

>> for i=1:159

t2 = double(solve(4.8825*exp(-0.0954*t)+ 12.845*exp(-0.0006211*t)-p,'t'));

p = double(4.8825*exp(-0.0954*(t2+4.77))+ 12.845*exp(-0.0006211*(t2+4.77)) + 0.05262*4.4253);

total = total + double(int(4.8825*(exp(-0.0954*x)+12.845/4.8825*exp(-0.0006211*x)),x,t2,t2+4.6059));

end

??? Error using ==> solve>getEqns at 178

' 1.590127e+000 ' is not a valid expression or equation.

Error in ==> solve at 67

[eqns,vars] = getEqns(varargin{:});

How do I use sprintf then in this case?

Answer 1

[Walter Roberson]

Instead of

t2 = double(solve('4.8825*exp(-0.0954*t)+ 12.845*exp(-0.0006211*t)-p','t'));

use

t2 = double(solve(4.8825*exp(-0.0954*t)+ 12.845*exp(-0.0006211*t)-p,'t'));

Otherwise it will not pick up the value of p from the MATLAB variable you assigned earlier.

[Bart]

I cried victory too soon :D.

I get this error on the second cycle:

>> syms x

>> total = 0;

>> t=0;

>> t2 = 0;

>> p = 4.8825*exp(-0.0954*t)+ 12.845*exp(-0.0006211*t);

>> for i=1:2

t2 = double(solve(4.8825*exp(-0.0954*t)+ 12.845*exp(-0.0006211*t)-p,'t'));

p = double(4.8825*exp(-0.0954*(t2+4.77))+ 12.845*exp(-0.0006211*(t2+4.77)) + 0.05262*4.4253);

total = total + double(int(4.8825*(exp(-0.0954*x)+12.845/4.8825*exp(-0.0006211*x)),x,t2,t2+4.6059));

end

??? Error using ==> solve>getEqns at 178

' 1.590127e+000 ' is not a valid expression or equation.

Error in ==> solve at 67

[eqns,vars] = getEqns(varargin{:});

[Sean de Wolski]

just rewrite it in it's non-exponential form:

1.590127

[Bart]

As I need to eventually run this algorithm 159 times, I was hoping if there is a script for this that i could use, as I don't exactly know where the 1.590127 comes from

[Sean de Wolski]

use sprintf to generate the string if necessary.

[Bart]

how could I incorporate sprintf in the above for-loop? I think the error comes forth from the calculation of 't' in the 'solve()' function. But I have no idea how I could implement sprintf in that case.

[Walter Roberson]

The real problem is that you are re-using variable names. You have t=0 and then in t2 you want to solve for t -- but t is 0. This leaves you trying to solve a constant rather than an expression.

Your expression for p calculates a particular expression when t=0, and then in t2 you are trying to solve the same expression minus p to get t out. If t was symbolic there, of course the solution is going to be t=0