Discover MakerZone

MATLAB and Simulink resources for Arduino, LEGO, and Raspberry Pi

Learn more

Discover what MATLAB® can do for your career.

Opportunities for recent engineering grads.

Apply Today

Could not find a feasible initial point after change the lower and upper boundaries.

Asked by Sook Teng on 28 Dec 2012

Hi all,

I have a problem on running the multiobjective genetic algorithm. I would like to minimise three objective functions, if I set the lb and ub to [0,0],[1,1], there will be no problem.

However, I am only interested in optimising the variables range within lb = [0.55,0.7]; ub = [0.9,0.9]. After I set it, I got messages as follow, Could not find a feasible initial point. The number of points on the Pareto front was: 0

As I though if I set options=gaoptimset(options,'PopInitRange',[0.55,0.7;0.9,0.9]) I will able to get the initial point and pareto front within that area. May I know why this happened?

Following is the code that I written for the multiobjective GA.

A = [-0.162 1;-0.31 1]; b = [0.75;0.31];
Aeq = []; beq = [];
lb = [0.55,0.7];
ub = [0.9,0.9];
options = gaoptimset(options,'TolFun',1e-4,'StallGenLimit',200);
options = gaoptimset(options,'ParetoFraction',0.5);
FitnessFunction = @objectives;
numberOfVariables = 2;
[x,fval,exitflag,output,population,score] = ...

Could anyone kindly to answer my questions? Thank you.

Regards, vunteng

1 Comment

Matt J on 29 Dec 2012

And the fitness function is definitely defined and finite-valued everywhere in the constrained region? Nothing that would produce NaNs or Infs?

Sook Teng

0 Answers

Contact us