WORK IN PROGRESS: module for measuring and analyzing linear optics and equilibrium parameters from turn-by-turn data acquisitions.
The measurement script has been tested online and works well. Only linear optics analysis is available for now (beta beating and phase advance errors), as probed by either sinusoidal fitting, PCA or ICA. Equilibrium parameters fitting (energy spread and emittances) is not yet implemented.
TODOs
[ ] Phase unwrapping:
numpy.unwrap() does not work properly for large phase advances (~2.26 rad). This is observed mostly in acquisitions of
vertical motion
[ ] Uncertainties and error bars:
expressions for the uncertainties and error bars for the beta-functions and phase-advances are known, but have not yet
been implemented.
[ ] Linear optics analysis sensitivity & robustness
the estimates for beta-functions and phase advance have not yet been scrutinized against models simulations (where
ground-truth is know) to access their performance (robustness to noise & artifacts, precision and accuracy).
From the early experience in handling data from the recent acquisitions, the fitting approach (harmonic_analysis) seems
particularly sensitive to artifacts, and often has poorer performance at fitting beta functions, overestimating beta-beating.
PCA & ICA perform better. PCA, however, seems particularly sensitive to contaminating source signals with variances
close to the betatron signals. In this scenario, ICA is better in separating the signals and accurately accessing the optics.
Additionally, with the application of these analysis as a characterization tool to evaluate changes and corrections in the
machine, it is important to study their sensitivity: how well can they access optics perturbations?
[ ] Equilibrium Parameters
analysis of the data at larger timescales (hundreds & thousands of turns) to access energy-spread and emmitance
through the effects of decoherence modulation has not been implemented yet.
[ ] Coupling
analysis of TbT data can also reveal information about betatron coupling. This has not yet been studied & implemented yet.
WORK IN PROGRESS: module for measuring and analyzing linear optics and equilibrium parameters from turn-by-turn data acquisitions.
The measurement script has been tested online and works well. Only linear optics analysis is available for now (beta beating and phase advance errors), as probed by either sinusoidal fitting, PCA or ICA. Equilibrium parameters fitting (energy spread and emittances) is not yet implemented.
TODOs
[ ] Phase unwrapping:
numpy.unwrap()does not work properly for large phase advances (~2.26 rad). This is observed mostly in acquisitions of vertical motion[ ] Uncertainties and error bars: expressions for the uncertainties and error bars for the beta-functions and phase-advances are known, but have not yet been implemented.
[ ] Linear optics analysis sensitivity & robustness the estimates for beta-functions and phase advance have not yet been scrutinized against models simulations (where ground-truth is know) to access their performance (robustness to noise & artifacts, precision and accuracy). From the early experience in handling data from the recent acquisitions, the fitting approach (
harmonic_analysis) seems particularly sensitive to artifacts, and often has poorer performance at fitting beta functions, overestimating beta-beating. PCA & ICA perform better. PCA, however, seems particularly sensitive to contaminating source signals with variances close to the betatron signals. In this scenario, ICA is better in separating the signals and accurately accessing the optics. Additionally, with the application of these analysis as a characterization tool to evaluate changes and corrections in the machine, it is important to study their sensitivity: how well can they access optics perturbations?[ ] Equilibrium Parameters analysis of the data at larger timescales (hundreds & thousands of turns) to access energy-spread and emmitance through the effects of decoherence modulation has not been implemented yet.
[ ] Coupling analysis of TbT data can also reveal information about betatron coupling. This has not yet been studied & implemented yet.