How could I connect simulink model to the genetic algorithm toolbox ?

26 views (last 30 days)
Hi, I am currently working on suspension system optimization via Genetic Algorithm. As the suspension model is in Simulink Interface, I couldn't find a way to connect simulink variables with the Population generated by genetic algorithm toolbox. Kindly help me with this.
  2 Comments
wenlong Ma
wenlong Ma on 14 Jun 2018
i have the same Problem with you , do you find solution ? could you please tell me something about it ? or could you send me an mail : mwl.maschine@gmail.com
Andrea Valletta
Andrea Valletta on 16 Jan 2020
Is it possible for you to share the solution? I am currently unsuccesful to run the GA algorithm starting from the matlab script. ga function calls the objective function within which the simulink is started, as shown by Sebastian.
Thanks in advance

Sign in to comment.

Answers (1)

Sebastian Castro
Sebastian Castro on 7 Jun 2017
Edited: Sebastian Castro on 7 Jun 2017
It's all in your cost function. For example:
function cost = costFunction(params)
myValue1 = params(1);
myValue2 = params(2);
% NOTE: This assumes the Simulink model has blocks that
% require variables named 'myValue1' and 'myValue2'
outputs = sim('modelName',<options>);
cost = <some_calculation>(outputs);
end
Hope this helps!
- Sebastian
  2 Comments
Alberto Mora
Alberto Mora on 13 Mar 2020
Edited: Alberto Mora on 13 Mar 2020
I did what you suggested but it seems that the function workspace cannot send the updated design variable to the Simulink workspace.
Therefore I solved the problem, as you can see below.
function cost = costFunction(params)
myValue1 = params(1);
myValue2 = params(2);
hws = get_param('modelName','modelworkspace'); % Handle the model's workspace
hws.DataSource = 'MAT-File';
hws.FileName = 'workspace';
hws.assignin('myValue1', myValue1);hws.assignin('myValue2', myValue2);% Send to Simulink the updated variables
% NOTE: This assumes the Simulink model has blocks that
% require variables named 'myValue1' and 'myValue2'
outputs = sim('modelName',<options>);
cost = <some_calculation>(outputs);
end
It works, but I am not sure that it is the most efficient way to overcome the problem. What do you think about that?
Thanks,
Alberto
Jarin Tasnim
Jarin Tasnim on 7 Aug 2023
@Alberto Mora hi, I am trying to optimize my simulink battery model parameters with genetic algorithm. Can I see your full matlab code from where the cost function is being called? I have written the cost function but I am facing some errors. It would be a great help! Thanks!
Jarin

Sign in to comment.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!