Fidelity Zero Aerodynamic analysis

[Divyam Verma]

Hi,

Upon running my mission, I am getting 'nan' for my total aircraft drag and total aircraft lift. Why is this so? The module is computing the individual drag components (eg.: parasitic drag), but I am getting this error:

Segment did not converge. Segment Tag: cruise1

Error Message:

The iteration is not making good progress, as measured by the 

  improvement from the last ten iterations.

surrogate

Much help appreciated.

Kind regards,

Divyam

[ebot...@gmail.com]

A NaN like that usually means a dimension is set to 0. Usually places to look for would be reference areas, lengths, or chords.

-Emilio

[Divyam Verma]

Thanks Emilio.

[Divyam Verma]

Hi Emilio,

Sorry to bother you again. I am running a Fidelity_Zero() aerodynamic module. In running the mission analysis, the mission iteration sometimes gives me this below error:

"Propeller BEMT did not converge to a solution (Stall)"

Could you please guide why and which Python file is this error arising form? I have noticed that these errors are sensitive to the inputs we give. Hence, do I need to play with the inputs to avoid getting these errors?

Thanks a ton,

Divyam

On Wednesday, October 13, 2021 at 12:48:36 PM UTC+5:30 ebot...@gmail.com wrote:

[Divyam Verma]

Hi again,

Could you also please tell me where the drag_breakdown.induced.inviscid drag component is computed? 

Is it computed from the Vortex_Lattice.py file (I am using Fidelity_Zero as the aerodynamic module) ? Because sometimes I am getting NaN values for it. 

I have checked the reference areas, lengths and chords values. They are finite but on occasions I am still getting a NaN value.

Thanks,

Divyam

On Wednesday, October 13, 2021 at 12:48:36 PM UTC+5:30 ebot...@gmail.com wrote:

[ebot...@gmail.com]

Hi,

The location of both of those files and the behavior depends on the version of SUAVE. I'll show you the 2.4 relevant info.

The propeller shows this warning when the blades have stalled. It may or may not be an issue. If you do not have a mission convergence error, it's fine, ignore it. What happened was the mission solver tried a condition that made the blades stall, but then found the right solution. If the mission fails and you get that warning (and/or it prints that warning dozens of times), then your propeller is not well designed for your vehicle and mission. So yes, in that case adjust your mission, your vehicle, and the initials guess on the mission.

The inviscid drag comes from the VLM. If that's NaN it's highly likely that lift is also NaN. In a mission it is computed via a surrogate. It is possible the surrogate points are no good for your problem (different AoA/Mach's). There's an option to disable the surrogate. Missions will take a long time to solve in this case. But it will rule that out, and then you can adjust the surrogate. When you setup your analyses, set this option: aerodynamics.settings.use_surrogate = False

The other thing to help you pinpoint would be to check at the wing level, to know which wings are getting NaN's. The induced.inviscid drag data structure should split it out.

-Emilio

[Divyam Verma]

Hi Emilio,

First of all, thanks a ton. You have been really helpful!

Funny you should talk about the convergence mission error. I've got rid of the NaN's but I am getting this Error message:

Segment did not converge. Segment Tag: cruise1

Error Message:

The iteration is not making good progress, as measured by the 

  improvement from the last ten iterations.

It is  computing the lift and drag well but am still getting this error. Even this error (The propeller shows this warning when the blades have stalled) has disappeared.

Why is this happening do you think?

-Divyam

[ebot...@gmail.com]

Hi,

Try changing your cruise segment. This means the airplane doesn't want to fly. Try a very short cruise segment and vary the speeds. Then play around with initial guesses on unknowns. If you have a converged climb segment before that can help you find what's reasonable.

-Emilio