[WillemFloris]

Hello guys,

I run a simulation whereby the drag has to be optimized. Which is done for the first +- 60 designs ( 'NIT' = 8 ). However after this point the objective function (drag) increases again. I checked the residuals of the direct- and adjoint calculations, both were nicely converged.
At this point, I don't know what is happening inside 'the black box' and any help is welcome. I would like to know why the objective function is increasing after a couple of 'NIT'. Also, some tips to avoid this behavior are welcome .

Kind regards,

Floris

(config file: turb_SA_RAE2822.txt )

[WillemFloris]

Hello all,

This time I ran another case, which contains the same behavior as the test case before. The objective of this case is to minimize the drag, just for testing purposes. During the line search process the optimizer does not choose the design with the lowest drag, but it chooses the design with the highest efficiency. Can anybody tell me, why the optimizer is choosing for highest efficiency instead of choosing the design with lowest drag?
Any help is appreciated

Cheers!

Floris

config file: turb_SA_RAE2822.txt
history_project.dat:history_project.txt



log.txt:

New Project: ./
Sequential Least SQuares Programming (SLSQP) parameters:
Number of design variables: 22 ( 22 )
Objective function scaling factor: [0.1]
Maximum number of iterations: 100
Requested accuracy: 1e-11
Initial guess for the independent variable(s): [0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]
Lower and upper bound for each independent variable: [(-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01), (-0.01, 0.01)]

NIT FC OBJFUN GNORM
1 1 1.412388E-03 1.995497E-01
2 12 1.426525E-03 2.079778E-01
3 23 1.426297E-03 2.077896E-01
4 34 1.425957E-03 2.075662E-01
5 40 1.406145E-03 2.047457E-01
6 51 1.405817E-03 2.046707E-01
7 62 1.405490E-03 2.045980E-01
8 73 1.405164E-03 2.045267E-01
9 84 1.404844E-03 2.044569E-01
11 95 1.404627E-03 2.042543E-01