help for subs and sym commands ???

Question

Hatem Çoban 2017-2-16

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/325466-help-for-subs-and-sym-commands

评论： Hatem Çoban 2017-2-17

if true
  clc;clear;close;
format long
syms x
f=0.0000095862*((x+1.4051)^0.44267)*((6.4051-x)^5.5867);
g=diff(f,x);
g2=diff(f,2);
h=solve(g==0)
h2=solve(g2==0)
A=subs(f,x,h)
end

A must be double? Why it is sym in workplace ? Besides I have some warnings; Warning: Cannot solve symbolically. Returning a numeric approximation instead.

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

John D'Errico 2017-2-16

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/325466-help-for-subs-and-sym-commands#answer_255127

编辑：John D'Errico 2017-2-16

在 MATLAB Online 中打开

Why MUST A be a double? Did you not create it as symbolic? Even though the result is a "number", it can still be stored as a symbolic form, so many digits. It will be, since how does MATLAB know that at the end, you want it as a double? How does MATLAB know that you will not now do something else with A that requires symbolic computation?

For example, you compute both h and h2 along the way. Both of them are only precursors to the final value of A, yet both of them were in symbolic form too. Would you have wanted MATLAB to arbitrarily decide that h and h2 are numbers,so it should convert them in to doubles,losing all of those extra digits along the way? While it might not matter here, it surely will matter some other time.

However, nothing stops you from now converting A INTO a double, as long as it is purely numeric. Try these things: