Update test-mass noise model

Closes #136 (closed)

• Add option testmass_shape (similar to laser_shape) to add low-frequency relaxation to test-mass noise model on Instrument. Default remains at 'original' , one can enable this extra low-frequency component with 'lowfreq-relax' .
• Add missing high-frequency component to test-mass noise (both original and lowfreq-relax models).

changed milestone to %v1.6

added feature label

assigned to @j2b.bayle

added 1 commit

• e03e458a - Add option to add relaxation to TM noise

Compare with previous version

changed the description

requested review from @ohartwig

Hi JB!

As quickly discussed on slack, I would suggest to use this opportunity to update the TM noise to also increase the high frequency part of the noise shape. It's usually assumed to have this shape (though this is a requirement/allocation, but what we use right now is based on the same formula where we just ignored the high-frequency part):

changed title from Resolve "Allow for a more pessimistic model of test-mass noise" to Update test-mass noise model

changed the description

added 1 commit

• e0b3af95 - Add high-frequency part of test-mass noise

Compare with previous version

You're giving me more work here @ohartwig .

That should all be implemented now. I've tested it by generating a noise time series from lisainstrument.noises.testmass() with both noise shapes, at 16 Hz and then 0.01 Hz to look at the high and low frequency parts. I've superimposed the model, and everything seems in order.

I played to change the various parameters and it seems to work as well.

I've also taken a look at the report, and that also seems to work.

Would you mind taking a second look and merging this @ohartwig (a bit of work for you too)!

added 5 commits

• e0b3af95...4775f4c0 - 3 commits from branch master
• 889722c0 - Add option to add relaxation to TM noise
• 9d5c5f82 - Add high-frequency part of test-mass noise

Compare with previous version

added 1 commit

• 914a43fd - Optimize relaxed model for TM noise

Compare with previous version

As discussed on Slack, I've factorized the terms for the relaxed model, so that we only compute 6 independent components.

I've tested that it still matches our models (both models), which seems to be the case.

What do you think now @ohartwig ?

added 1 commit

• cdf6f797 - Fix levels in factorization

Compare with previous version

added 6 commits

• cdf6f797...78137385 - 2 commits from branch master
• dd84759a - Add option to add relaxation to TM noise
• 04802e1b - Add high-frequency part of test-mass noise
• 38ce74b2 - Optimize relaxed model for TM noise
• 35b324b5 - Fix levels in factorization

Compare with previous version

removed review request for @ohartwig

resolved all threads

enabled an automatic merge when the pipeline for 35b324b5 succeeds

merged

mentioned in commit 03831647