Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page
Ask Your Question

Revision history [back]

More complex constraints in genOp

Dear everyone,

I am using genOpt to perform an optimization of a shading device. This device has been parametrized, and 5 parameters have been found.

Even though all these parameters are independent of each other (they are a base), one of them is constrained to be half of another. Otherwise, the geometry does not make sense.

Accordingly, what I need to achieve is something like:

Parameter{ 
    Name    = spacing;
    Min     = 0.01;
    Ini     =  0.25;
    Max     =  2 * radius;
    Step    =  0.1;
}

Parameter{
    Name    =  radius;
    Min     =  -3;
    Ini     =  0;
    Max     =  3;
    Step    =  0.5;
}

I have not been able to get a clear understanding of constraints by reading the genOpt manual, so I have not been able to implement this.

I think I can create some auxiliary variables for solving this, but the results will be somehow complex and "unnatural", since this new variables will not really mean much in the visual part of the geometry... thus, I want to avoid this.

Thanks in advance!

More complex constraints in genOp

Dear everyone,

I am using genOpt to perform an optimization of a shading device. This device has been parametrized, and 5 parameters have been found.

Even though all these parameters are independent of each other (they are a base), one of them is constrained to be half of another. Otherwise, the geometry does not make sense.

Accordingly, what I need to achieve is something like:

Parameter{ 
    Name    = spacing;
    Min     = 0.01;
    Ini     =  0.25;
    Max     =  2 * radius;
    Step    =  0.1;
}

Parameter{
    Name    =  radius;
    Min     =  -3;
    Ini     =  0;
    Max     =  3;
    Step    =  0.5;
}

I have not been able to get a clear understanding of constraints by reading the genOpt manual, so I have not been able to implement this.

I think I can create some auxiliary variables for solving this, but the results will be somehow complex and "unnatural", since this new variables will not really mean much in the visual part of the geometry... thus, I want to avoid this.

Thanks in advance!