
Re: Monte Carlo simulation with inequality constraints
Posted:
Mar 22, 2012 6:36 AM


Il giorno giovedì 22 marzo 2012 07:24:11 UTC+1, Ray Koopman ha scritto: [..] > > Might it be that whoever wrote the PCC knew what he was doing and > assumed that the users of the program would know all the same tricks?
Yes and no. The PCC was written that way, because from a design point of view, a "tube" (it's not a tube, in reality, but a part of a turbomachine) with inlet smaller than outlet makes no sense. So the PCC prevents the user from assigning design values of D1 and D2 such that D1<D2. Unluckily, just because the design values of D1 is always larger or equal than the design value of D2, this doesn't mean that the manufacturing process, which has a finite process capability, won't produce "tubes" where D1 <D2. As a matter of fact, this sometimes happen when the design value of D1 and D2 are equal.
> It's hard for me to put this intelligibly when I have no idea what > the code is supposed to be doing, but are there any symmetries that > he might have assumed the users would be aware of, so that if you > interchange certain input values then you just have to interchange > or reverse or complement or ... certain values in the output?
Not really, the reason is as I said. I will try to be more specific: it's a computational chain of CAD and fluid dynamics code for a compressor. You define the geometrical details of your design (inputs), such as diameters, blade angles, blade thickness, number of blades, etc. and you generate a CAD part. From the CAD part, some codes extract the input data for the fluid dynamics codes, such as for example blade length, leading edge length normal to the flow. The fluid dynamics codes are then run, and the compressor performance, such as efficiency, head, flow coefficient, etc. are computed. These are the outputs.
I apologize in case this is not enough, but I'm really not allowed to give more details. Of course, I understand if you cannot give me help with these few details. In that case, thanks anyway for your help until now.
deltaquattro

