Furkan Tercanlı 0 Report post Posted January 31 Hello, I have result pool and I want to create a response surface optimization using the results. The results have been come from the sampling method with running CFD. When I read the article which is titled "How To Create a Response Surface in 4 Easy Steps" on your website, it tells that response surface method can be used to optimize the design without CFD run. I guess the article is belong to old version of CAESES. I cannot use the method without CFD. It uses the result pool and creates new designs and after that, it runs the design iteration on CFD. How can I use the response surface optimization without CFD run? Best Regards, Furkan Share this post Link to post Share on other sites
Mr. Heinrich von Zadow 9 Report post Posted January 31 Hi Furkan, If you set up a "Response Surface Optimization" and choose to "use result pool", a surroate will be created (if your pool is large enough, otherwise samples will be added first). An optimization on that surrogate is performed automatically in the background (a Genetic Algorithm is used here) and the optimal candidate (or multiple if you choose so for a multi-objective problem) is returned. The design you see created in CAESES is therefore either an additional sample or already the potential optimum. Check the dakota.out file in your design directory to see the details. Also, the finaldata file will show you the predicted performance of the optimum... If you keep this Algorithm running, the actual (CFD) result(s) of the optimum-candidate(s) will be added to the pool for the next iteration and the process repeats. Cheers, Heinrich 1 Share this post Link to post Share on other sites
Mr. Sebastian Uharek 1 Report post Posted January 31 Hi, you can select the option "Response Surface Optimization in the Optimization workspace. Afterwards, you can check the option "Use Result Pool" It will ask you to select the pools which you want to utilize for the surrogate based optimization after starting the design engine. Please note that if you do not have sufficient samples in your result pool, it will do additional samples before starting the optimization. It will then perform the optimization on the surrogate in the background without calling CFD. After that, a new design will show up in CAESES, which is then evaluated using the CFD to check if it matches the prediction from the surrogate. 1 Share this post Link to post Share on other sites
Furkan Tercanlı 0 Report post Posted February 1 Thank you for your explanatory answers. I have another problem. Initially, the model added new samples until it had enough in the pool. Then, the iterations began. However, the resulting design from the RSM convergence violates my constraints. How can I prevent the model from creating optimal designs that violate constraints? Best Regards, Furkan Share this post Link to post Share on other sites
Mr. Heinrich von Zadow 9 Report post Posted February 1 Hi Furkan, if you have set the constraints in the Design Engine, the algorithm predicts an optimum within the feasible domain. However, predicting constraints (which is technically the same as predicting objectives) comees with uncertainty and hence, sometimes none of the predicted optimal designs is actually feasible. What might help is, to set a narrower margin in the Design Engine and try again. Cheers, Heinrich Share this post Link to post Share on other sites
