TAGGED: inverse-design, topology-optimization
-
-
November 25, 2024 at 12:41 pmgangin2Subscriber
Hello,
I am running a topology optimization of a 90 degree bend (see below).
The topology python script I adapted from the example on the website can run the zero-th iteration and it gets to the structure above. However, it cannot run the next iteration.
This is the terminal output:
At iterate   0   f= -1.27792D+01   |proj g|=  2.50000D-01
Making forward solve
Making adjoint solve
Running solves
Processing forward solve
Traceback (most recent call last):
 File "E:\Documents\Research\Lumerical\Inverse\bend\bend_opt2.py", line 74, in
  runSim(initial_cond, eps_min, eps_max, x_pos, y_pos, filter_R*1e-9, working_dir)
 File "E:\Documents\Research\Lumerical\Inverse\bend\bend_opt2.py", line 45, in runSim
  opt.run(working_dir = working_dir)
 File "C:\Program Files\Lumerical\v241\api\python\lumopt\optimization.py", line 582, in run
  self.optimizer.run()
 File "C:\Program Files\Lumerical\v241\api\python\lumopt\optimizers\generic_optimizers.py", line 50, in run
  res = spo.minimize(fun = self.callable_fom,
 File "C:\Program Files\Lumerical\v241\python\Lib\site-packages\scipy\optimize\_minimize.py", line 609, in minimize
  return _minimize_lbfgsb(fun, x0, args, jac, bounds,
 File "C:\Program Files\Lumerical\v241\python\Lib\site-packages\scipy\optimize\lbfgsb.py", line 345, in _minimize_lbfgsb
  f, g = func_and_grad(x)
 File "C:\Program Files\Lumerical\v241\python\Lib\site-packages\scipy\optimize\lbfgsb.py", line 295, in func_and_grad
  f = fun(x, *args)
 File "C:\Program Files\Lumerical\v241\python\Lib\site-packages\scipy\optimize\optimize.py", line 327, in function_wrapper
  return function(*(wrapper_args + args))
 File "C:\Program Files\Lumerical\v241\api\python\lumopt\optimizers\minimizer.py", line 24, in callable_fom_local
  fom = callable_fom(params_over_scaling_factor)
 File "C:\Program Files\Lumerical\v241\api\python\lumopt\optimization.py", line 781, in callable_fom
  fom = self.process_forward_sim(iter)
 File "C:\Program Files\Lumerical\v241\api\python\lumopt\optimization.py", line 714, in process_forward_sim
  Optimization.check_simulation_was_successful(self.sim)
 File "C:\Program Files\Lumerical\v241\api\python\lumopt\optimization.py", line 1027, in check_simulation_was_successful
  raise UserWarning('FDTD simulation did not complete successfully: status {0}'.format(simulation_status))
UserWarning: FDTD simulation did not complete successfully: status 0.0Â
If I check the log file, it states "Error: there was an unknown parallel error. The error code is 0, the process number is 5" for both the forward and adjoint solves.
Â
Once the script fails, I can independently run the simulation files that it has created for the forward and adjoint solves. They will also crash but give more details:
From this error message, since I am doing a narrowband varFDTD simulation I tried following the suggestion and unchecked "can optimize mesh algorithm for extruded structures."Â However doing that does not solve the issue and I'm not sure how I can resolve the next error:
Any help would be appreciated. Thank you!
-
November 28, 2024 at 1:27 amFatema ChowdhuryAnsys Employee
Hello,
According to Optimizable Geometry - Python API – Ansys Optics, one should work with FDTD for all topology optimizations. Therefore, I would first suggest that you kindly try FDTD instead of varFDTD.
For further support, resolving the issue without the access to the file could be difficult. Debugging such issues with a specific file is beyond the scope of ALF. I recommend that you can kinldy create a case in ACSS (https://customer.ansys.com/) to provide you proper support. To know more about ACSS,please use this link(https://optics.ansys.com/hc/en-us/articles/5609426062483-How-to-register-on-Ansys-Customer-Support-Space)
-
- You must be logged in to reply to this topic.
- Errors Running Ring Modulator Example on Cluster
- Difference between answers in version 2024 and 2017 lumerical mode solution
- INTERCONNECT – No results unless rerun simulation until it gives any
- Import material .txt file with script
- Trapezoidal ring
- Help for qINTERCONNECT
- Issues with getting result from interconnent analysis script
- Topology Optimization Error
- Edge Coupler EME Example Issue
- The two modes overlap the integral
-
1191
-
508
-
488
-
225
-
209
© 2024 Copyright ANSYS, Inc. All rights reserved.