Hello Friends,
I am working on fractals and I am using GA for optimization of my patch antenna. What I am thinking is to first perform a number of iterations on the antenna for a particular parameter, say slot width or fractal length. Then I find the impact of that parameter on the resonant frequency or S11 or bandwidth of the patch and deduce an equation or relationship using Curve Fitting toolbox in MATLAB. Then, using that equation apply GA on the same and obtain an optimized value.
I am not aware whether I am right or not.
Kindly guide me if I am going in wrong direction.
Thanks in advance.
Hi, Kumar.
Today see me two times, I'm happy about that.
More than one study the influence of parameters on the performance of an antenna. but using curve fitting is good ideas. All the above is traditional work. But using it in GA is a very good thing, and I hope good results for you.
And I think I will hear happy news from you.
With my best regards.
Hi Kumar
Your alternative seems ok, other solution (I used in the past) was to use the genetic alg. toolbox in matlab to control the EM Field solver (used CST and the VBA automation code). So basically matlab used the info contained in the GA code to build the antenna (a yagi), automatically building the geometry inside CST with Visual Basic (not hard to do). Then Matlab starts CST, waits till its done and collects the results, by reading the S parameter files. Then using the info in the S parameter files one's able to evaluate a cost function and start it over again.
There is also a DLL in CST MWS that helps out working with automation from outside.
good luck
Thanks Hayder and Marcelo for sharing your views. Do any of you have any solution in HFSS as I am not working on CST? Any information regarding GA in HFSS is highly appreciated.
Thanks once again.
- Badges
- Science topic