Simplify ssto #534
Simplify ssto #534
Conversation
| elif cb == 'moon': | ||
| rho_ref = 0.0 | ||
| h_scale = 1.0 | ||
| else: |
There was a problem hiding this comment.
Don't need the else branch because self.options already does value-checking to assure that "central_body" is only in ['earth', 'moon']
There was a problem hiding this comment.
Honestly we're not even running a test against "moon" any more (and haven't been for some time). There's no need to add this complication here, just remove cb as an option and replace with rho_ref and h_scale options which default to the Earth values.
There was a problem hiding this comment.
The test_parameters failure is just due to the targets of the states and parameters changing since there's no longer a nested EOM. Since the variables have the same names as the inputs of the ODE, targets should just be omitted. State rates will need to be changed as well. Units are the same as those defined in the ODE and can also be removed.
| elif cb == 'moon': | ||
| rho_ref = 0.0 | ||
| h_scale = 1.0 | ||
| else: |
There was a problem hiding this comment.
Honestly we're not even running a test against "moon" any more (and haven't been for some time). There's no need to add this complication here, just remove cb as an option and replace with rho_ref and h_scale options which default to the Earth values.
| F_T = inputs['thrust'] | ||
| Isp = inputs['Isp'] | ||
|
|
||
| g = _g[self.options['central_body']] |
There was a problem hiding this comment.
also include g as an option, since we're only testing against Earth.
| self.declare_partials(of='mdot', wrt='Isp', rows=ar, cols=ar) | ||
|
|
||
| # # Complex-step derivatives | ||
| # self.declare_partials(of='*', wrt='*', method='cs') |
There was a problem hiding this comment.
With this PR (OpenMDAO/OpenMDAO#1862) these lines should work now - can remove the other declare_partials calls, and removal of compute_partials
| @@ -45,18 +43,18 @@ def test_doc_ssto_earth(self): | |||
| phase.set_time_options(initial_bounds=(0, 0), duration_bounds=(10, 500)) | |||
|
|
|||
| phase.add_state('x', fix_initial=True, ref=1.0E5, defect_ref=1.0, | |||
| rate_source='eom.xdot', units='m') | |||
| rate_source='xdot', units='m') | |||
There was a problem hiding this comment.
targets and units here are no longer necessary, since the targets have the same name as the states/controls/parameters.
…it with options for rho_ref and h_scale. Removed targets and units where not longer necessary
updated some docs for kaushiks SSTO case
Summary
Modified SSTO problem to be single component rather than a group. Added code to declare derivatives to use complex step
Related Issues
Status
Backwards incompatibilities
None
New Dependencies
None