ENH: Generic Surfaces and Generic Linear Surfaces

[MateusStano]

Pull request type

• Code changes (bugfix, features)

Checklist

• Tests for the changes have been added (if needed)
• Docs have been reviewed and added / updated
• Lint (black rocketpy/ tests/) has passed locally
• All tests (pytest tests -m slow --runslow) have passed locally
• CHANGELOG.md has been updated (if relevant)

Description

• Added the GenericSurface class that takes in coefficients defined by either a function or a .csv

• Added the LinearGenericSurface class that takes in coefficients derivatives and calculates the total coefficients assuming linearity

• Changed all add_surfaces position arguments to be 3 dimensional

• Improved Flight class structure to be indifferent to the type of AeroSurface

• There is an example simulation in coeff_testing.ipynb of calisto using coefficients that were extracted directly from rocketpy's standard model (Barrowman's). It uses the .csvs that have been committed. These .csvs have a LOT of points, this is due to the shepard interpolation in the Function class, which makes the interpolated values used in the simulation extremely noisy. The files are this big for validation purposes. We should reduce them and create tests with them

Breaking change

• Yes
• No

---

Remaining Tasks

• 1. Add a representation of GenericSurface in the Drawing.
• 2. Create Plots and Prints classes for both GenericSurface and LinearGenericSurface
• 3. Add sideslip and partial angle of attack to flight plots and/or prints
• 4. Create a method in the Rocket class to define the coefficients of the entire rocket. The method should receive a GenericSurface or LinearGenericSurface and use its coefficients for simulation, overwriting any other added aerodynamic surface. The aerodynamic surfaces should still be used for the drawing though. The received generic surface should be positioned in the center of the dry mass of the rocket for the desired behavior
• 5. Currently, we receive cL, cD, cQ, however, user's might prefer to input the coefficients relating to the body axes directly (cX, cY, cZ). This should be an optional parameter in the instantiation of the class.
• 6. Pylint
• 7. Tests
• 8. Doc pages

It would be truly great to get some help on these minor tasks

Future Tasks

1. Create an axisymmetric surface class that needs only cD, cL, either center of pressure or cm, and cl (roll moment coefficient)
2. The Rocket class should have total coefficient attributes: cL, cQ, cD, cm, cn, cl. They should be included in Plots or Prints
3. Add lateral stability analysis. The current center of pressure, static, and stability margins only works for axisymmetric configurations. GenericSurfaces and LinearGenericSurfaces are currently not included in those analysis.

---

Contributors

Special thanks to @kevin-alcaniz and Faraday Rocketry UPV for their contributions!!

[codecov]

Codecov Report

Attention: Patch coverage is 56.57568% with 175 lines in your changes missing coverage. Please review.

Project coverage is 74.80%. Comparing base (dca3a84) to head (e9a8274).
Report is 3 commits behind head on develop.

Files with missing lines	Patch %	Lines
rocketpy/rocket/aero_surface/generic_surface.py	16.48%	76 Missing ⚠️
...etpy/rocket/aero_surface/linear_generic_surface.py	23.91%	35 Missing ⚠️
rocketpy/simulation/flight.py	69.23%	28 Missing ⚠️
rocketpy/prints/aero_surface_prints.py	30.00%	14 Missing ⚠️
rocketpy/environment/fetchers.py	0.00%	6 Missing ⚠️
rocketpy/rocket/rocket.py	86.48%	5 Missing ⚠️
rocketpy/environment/environment_analysis.py	0.00%	2 Missing ⚠️
rocketpy/plots/aero_surface_plots.py	66.66%	2 Missing ⚠️
rocketpy/rocket/aero_surface/nose_cone.py	60.00%	2 Missing ⚠️
rocketpy/sensitivity/sensitivity_model.py	0.00%	2 Missing ⚠️
... and 3 more

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #680      +/-   ##
===========================================
- Coverage    75.40%   74.80%   -0.61%     
===========================================
  Files           96       98       +2     
  Lines        10832    11036     +204     
===========================================
+ Hits          8168     8255      +87     
- Misses        2664     2781     +117     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[Lucas-Prates]

I get the following error from Scipy when trying to load the coefficients using a csv file with insufficient data. Maybe it's worth checking the number of rows on the csv?

[Lucas-Prates]

When I used all arguments as 0 (or null vectors/tuples, depending on the type) on the GenericSurfaces class, the function worked and returned 0 for all its return values. For this function here in LinearGenericSurfaces, I got a division by zero error.

[Lucas-Prates]

Suggested change

	dyn_pressure_area = 0.5 * rho * stream_speed**2 * self.reference_area
	dyn_pressure_area_damping = (
	dyn_pressure_area * self.reference_length / (2 * stream_speed)
	)
	dyn_pressure_area_length = dyn_pressure_area * self.reference_length
	dyn_pressure_area_length_damping = (
	dyn_pressure_area_length * self.reference_length / (2 * stream_speed)
	)
	dyn_pressure_area = 0.5 * rho * stream_speed**2 * self.reference_area
	if stream_speed == 0:
	dyn_pressure_area_length = 0
	dyn_pressure_area_length_damping = 0
	else:
	dyn_pressure_area_damping = (
	dyn_pressure_area * self.reference_length / (2 * stream_speed)
	)
	dyn_pressure_area_length = dyn_pressure_area * self.reference_length
	dyn_pressure_area_length_damping = (
	dyn_pressure_area_length * self.reference_length / (2 * stream_speed)
	)

Possible division by 0. Since the numerator has stream_speed**2 and the denominator stream_speed, the limit is 0 when stream_speed goes to 0, so it could be a good idea to just set these variables to 0 when stream_speed is zero.