Solve function supports only four decimal points

5 次查看（过去 30 天）
V. Freyr Gudmundsson
评论： V. Freyr Gudmundsson 2022-5-31
Is there a way to solve a system of equations like this with double precision without rounding? The theta vector here has precision double. But unless I limit the number of decimal points to 4 or fewer, I dont get an answer.
ts = 0.002;
syms a b;
eqn1 = 2+a*ts == 1.97720865074869;
eqn2 = -1-a*ts == -0.977208098949782;
eqn3 = ts^2*b == 0.00103667507747696;
%Note: solve function seems to accept only 4 decimal places.
%The round command is to accomodate this. Has nothing to do with the math.
sol = solve([eqn1, eqn2, eqn3],[a,b]); %solve for a and b, ts is known
aSol=double(sol.a)
bSol=double(sol.b)
If I just use double or increase the number of decimals to 5 I get : 0×1 empty double column vector.

请先登录，再回答此问题。

回答（2 个）

Star Strider
Star Strider 2022-5-31
I cannot run that because the ‘theta’ vector is missing.
However, the Symbolic Math Toolbox calculates in extended precision. If the results are limited to 4 decimal places, change the format setting to display more of them. Use the vpa function to change what the Symbolic Math Toolbox displays.
.
  1 个评论
V. Freyr Gudmundsson
V. Freyr Gudmundsson 2022-5-31
I failed at submitting a sample that would run. I have fixed that in the original post.
You will now see the problem when you run the code. It returns a struct with two 0x1 sym fields.
However, if you reduce the number of decimals down to four, it will come up with the right answer in two 1x1 sym fields.

请先登录，再进行评论。

Walter Roberson
Walter Roberson 2022-5-31
Add the first two equations. The a*ts cancel out, leaving the equation 1 = theta1 + theta2. If that equation is not satisfied then there cannot be any solution to the system.

类别

在 Help Center 和 File Exchange 中查找有关 Conversion Between Symbolic and Numeric 的更多信息

标签

产品


版本

R2022a

Community Treasure Hunt

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

Start Hunting!

Translated by