[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Orekit Users] Recommendation for lifetime analysis
Hi Bogdan,
Welcome to Orekit!
On Mon, 2018-03-26 at 02:20 +0200, bogdan.udrea@vissidus.com wrote:
Dear Colleagues,
First of all I would like to express my gratitude to the crew at CS and their
collaborators for making Orekit available to the community. I became aware of
Orekit in 2013 at the AAS/AIAA Space Flight Mechanics Meeting in Hawaii and
since then I have been lurking in the shadows waiting for the right time to
start working with it. I recently downloaded Orekit 9.1 and I became
productive in less than a day.
I am interested in performing a trade study for the lifetime of a LEO mission
using Orekit propagators. Preliminary design, using tables from Wertz' New
SMAD, shows a lifetime of a few months to a year. The stop condition is a
periapsis altitude of 100km.
I am writing to ask if anyone has a recommendation for a propagator for such a
study. On one hand it seems that DSST would be the fastest but I haven't used
it before. On the other hand, I am tempted by numerical propagators, for their
field propagation, because later on I would like to refine the study and
propagate uncertainties resulting from orbit injection.
I think it depends on what you're trying to achieve and which errors you expect to be most significant. The DSST tends to be good for long term studies and the NumericalPropagator for short high-fidelity analyses, but it really comes down to how you configure
each propagator. Each propagator will have modeling error (in the force model), truncation error (in the integrator), and rounding error (due to finite precision). Assessing modeling error comes from your understanding of astrodynamics and evaluating different
force models. The relation that truncation error tends towards zero as the step size tends towards zero can be used to reduce truncation error to an acceptable level. Rounding error is probably the hardest to evaluate because it increases with a decreasing
step size. Perhaps using a FieldPropagator with Dfp would allow you to separate out the rounding error from the other errors.
Does that help answer your question?
Best Regards,
Evan
Thank you and best regards,
Bogdan