is the selected solution the optimal one?

1 次查看（过去 30 天）

Housam 2021-3-13

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/771833-is-the-selected-solution-the-optimal-one

评论： Walter Roberson 2021-3-13

linprog always selects the lower limits (zero) for the last two variables and the upper limit for the second variable. However, the values should not be zeros.

question2: how can i visualize the feasible region, plotting the constraints?

objfunc.Constraints.econs1 = var1 + var2 == vector(i);

objfunc.Constraints.cons1 = var1 + var2 <= var3;

objfunc.Constraints.cons2 = var5 >= 3 * var4;

f= [49.1 49.1 98 577 1306];
vector=[950;1200;2000;20000];
for i=1:length(vector)
A= [1 1 -1 0 0
    0 0 0 -3 1]
b= [0 0]
Aeq=[1 1 0 0 0]
beq=vector(i)
lb=[0;0;0;0;0]
ub=[25112; 1255; 25112-0.05*25112; 1255; 3*1255]
options = optimoptions('linprog','Algorithm','dual-simplex','Display','iter');
[x,fval,exitflag,output,lambda] = linprog(f,A,b,Aeq,beq,lb,ub,options)
X{i,1}=x 
FVAL{i,1}=fval + 580 
end

4 个评论
显示 2更早的评论隐藏 2更早的评论

Housam 2021-3-13

Thanks for the reply.

zero is physically not an acceptable solution. then how can i improve my problem formulation to maximize the last two variables instead

Walter Roberson 2021-3-13

在 MATLAB Online 中打开

lb=[0;0;0;0;0]

tells it that 0 is acceptable. If 0 is not acceptable change the appropriate lb entry to hold the smallest acceptable value such as eps(realmin) which technically is not 0. However if the rejection is for physical reasons perhaps you should use the planck distance or 1/avagadro's number or whatever value corresponds to one quanta

请先登录，再进行评论。

请先登录，再回答此问题。