3 Replies Latest reply on Dec 5, 2016 10:34 AM by Alex Clarke

    Can I batch run multiple parametric studies?

    Brent Milbaugh



      I really enjoy the functionality that FlowSim provides with parametric studies, especially goal optimization. So far, I have been running only one at a time and waiting for the current one to finish before manually closing it then opening another study.  This method is sometimes inefficient because if one finishes overnight, hours are wasted before I get back to the office to start another one.


      I would like to know if there is a way to essentially batch run multiple goal optimizations so that when one finishes, another is automatically started.


      My alternative is creating tons and tons of configurations and batch running those. However, this severely clutters the configuration and project windows with studies I don't need at the end since I am only interested in the "goal" configurations.


      If it is not possible to run back to back goal optimizations, I would also greatly appreciate any tips on how to handle a large amount of configurations and trying to identify the "goal" configuration in the set of extraneous ones.


      I can provide more specific information on my project goals if needed.


      Thank you in advance.

        • Re: Can I batch run multiple parametric studies?
          Alex Clarke



          I've put this into the "New Ideas" section, as I have similar problems. Can I ask that you vote for this enhancement too?


          SOLIDWORKS World 2017 Idea Submission: My Favorites


          What is the nature of the project you are working on? Are you varying boundary conditions, trying to maximise flow rate, trying to reach a given pressure? I think the limitation to one goal for goal optimisation makes sense - Imagine asking Solidworks to aim for minimum flow velocity at the same time as minimum pipe diameter - Those goals are clearly at cross purposes - How does the program know which one to aim for?


          Let me know a little more about your project, it would be interesting to see what you are trying to achieve.



            • Re: Can I batch run multiple parametric studies?
              Brent Milbaugh



              Thanks for your reply.


              For my project, I have a nozzle with a pressure inlet and outlet and the mass flow rate is controlled by a stopper rod.  My goal optimization parameter is the mass flow rate and the design parameter is the stopper rod lift.  I have multiple stopper rod designs and for each I am trying to find the lift required to attain a desired flow rate (specified by the customer).  It would be nice to have those cases start up automatically.  In post-processing, I am interested in the velocity and pressure in the throttle region as well as the MFR vs. Lift curve for control purposes.  I agree that multiple goal optimization parameters would be difficult to implement although I've never had the need for such.  I'd be glad to answer more questions if you have any.


              I read over your linked post and I think it's safe to say your design process is a little more involved than mine but we would both still benefit from batch running parametric studies.  Unfortunately, I am getting denied from the Top Ten list so I can't vote right now.  I see you had the same problem. Were you able to get that fixed?




                • Re: Can I batch run multiple parametric studies?
                  Alex Clarke

                  Hi Brent,


                  I haven't registered my copy of Solidworks because of a mix up with company names (Turns out, our parent company own all the licences!). Getting it sorted at the moment by our reseller.


                  I see what you're trying to do now, you've got multiple designs and want to optimise and calibrate each one. Makes sense! Solidworks seems to classify a parametric study as a sub-study and not allow selection of it in the batch run menu.


                  One way you could approach it is to open another session of Solidworks on the same PC and do a parametric study on that at the same time. This would make the simulation run at half the speed, but if you're getting dead time overnight anyway it might not be such an issue. You'll want to save your work first, as I can't guarantee that this will work on every computer.


                  As far as identifying the "Goal" configuration, I tend to use Compare Results. Tools>Flow Simulation>Results>Compare.


                  Let me know how you get on.