ENH: variable gravity

[FranzYuri]

Pull request type

Please check the type of change your PR introduces:

• Code base additions (bugfix, features)
• Code maintenance (refactoring, formatting, renaming, tests)
• ReadMe, Docs and GitHub maintenance
• Other (please describe):

Pull request checklist

Please check if your PR fulfills the following requirements, depending on the type of PR:

• ReadMe, Docs and GitHub maintenance:

  • Spelling has been verified
  • Code docs are working correctly

• Code base maintenance (refactoring, formatting, renaming):

  • Docs have been reviewed and added / updated if needed
  • Lint (black rocketpy) has passed locally and any fixes were made
  • All tests (pytest --runslow) have passed locally

• Code base additions (for bug fixes / features):

  • Tests for the changes have been added
  • Docs have been reviewed and added / updated if needed
  • Lint (black rocketpy) has passed locally and any fixes were made
  • All tests (pytest --runslow) have passed locally

What is the current behavior?

currently gravity is a constant equal to 9,80665 m/s²

What is the new behavior?

Now, env.g is a callable that returns the gravity given the height.
On the other hand, env.standard_gravity will store the constant value of 9,80665 m/s².

Does this introduce a breaking change?

• Yes
• No

Other information

Enter text here...

[MateusStano]

I don't really understand what is going on here. Does the gravity only vary if the user passes gravity = "variable"? Shouldn't it be calculated just by getting the elevation? The gravity argument should be an optional argument that sets it to a constant value, just in case anyone wants to try to simulate another planet or something.

Am I getting something wrong here?

[phmbressan]

@MateusStano

I believe this PR may be reaching its final steps. Just a few things that I wanted to point out before finishing reviews and merging:

Flight class PotentialEnergy changes:

• The method was wrong since inputting self.z (a Function) to the environment gravity Function yielded an error at interpolation.
• Trying to multiply Function by possible arrays (such as self.env.g(self.z.source[,:1])) does not help.
• Temporary fix was to perform the multiplication "manually".

Environment class gravity parameter:

• I personally don't think that I took the best decision here, i.e., the default None leads to variable gravity, however, since the parameter is a Function input, the options were limited. I am open to better suggestions.

Gravity Function being discrete:

• The Environment discretizes the gravity function height domain for two main reasons: 1) It is likely faster in the case of Somigliana's formula (but there might be a small performance hit compared a constant valued case); 2) Using lambda sources raises an error at the plotting class when the .source[] attributes are used, since it turns many Functions of the Flight class also lambda based (being a performance hit also).

Nota bene: the fin flutter tests still do not pass for me locally.

[MateusStano]

Nota bene: the fin flutter tests still do not pass for me locally.

Hey, seems that in this case the test actually did its job. The change in the gravity model actually changed the simulation by just enough to make the test break. It seems flutter analysis is very very sensitive

I think I fixed it now, please just try to run it locally to make sure

[giovaniceotto]

@MateusStano, I see you changed the values for the fin flutter test. Shouldn't we keep a test with the old values, but using constant gravity, just to be safe?

To be clear, I mean we should keep both tests: the old one and the new one.

[giovaniceotto]

I am afraid this formula, $U_g (z) = mg(z)z$, is no longer valid for variable gravity. Instead, we should use $U_g (R) = \frac{GMm}{R}$ where $R$ is the rocket's center of mass position vector magnitude relative to Earth's center.

Reference: https://en.wikipedia.org/wiki/Gravitational_energy

[MateusStano]

So I implemented this in 90b2bdb with the following code:

        potentialEnergy = (
            GM
            * totalMass
            * (1 / (self.z + self.env.earthRadius) - 1 / self.env.earthRadius)
        )

So what we are calculating here is the potential energy in relation to the sea level. All good now right?

[MateusStano]

This is the result for Calisto: