ETD

Archivio digitale delle tesi discusse presso l'Università di Pisa

Tesi etd-01232018-121342


Tipo di tesi
Tesi di dottorato di ricerca
Autore
CERRETANI, GIOVANNI
URN
etd-01232018-121342
Titolo
Software of Advanced Virgo Suspension Control System
Settore scientifico disciplinare
FIS/01
Corso di studi
FISICA
Relatori
tutor Passuello, Diego
commissario Prof. Batignani, Giovanni
commissario Prof. Calloni, Enrico
commissario Prof. Sozzi, Marco
commissario Dott. Verkindt, Didier
Parole chiave
  • controlli digitali
  • filtri adattati
  • onde gravitazionali
  • software
  • scada
  • dsp
  • virgo
Data inizio appello
29/01/2018
Consultabilità
Completa
Riassunto
The detection of gravitational waves has been done by very complex detectors based on laser interferometry. Thanks to sophisticated suspension systems, that insulate the optical components of the interferometers from seismic noise, it has been possible to extend the bandwidth of the detectors down to few tens of hertz. This has been one of the most important piece in the discovery puzzle, as the gravitational wave signals detected so far carried most of the information below 100 Hz. From the beginning, the Virgo experiment started to develop the Superattenuators, complex mechanical structures used to improve the sensitivity in this bandwidth. The Suspension Control System has been developed to implement digital control systems on the Superattenuators based on Digital Signal Processors (DSP). The last version of the Superattenuators is the result of the Advanced Virgo project, whose one of the most important upgrade concerned the new control electronics.

In August 2017, during the scientific run O2, the Suspension Control System consisted of 131 DSP-based boards running on the experiment, to control the 10 Virgo Superattenuators. A large number of device requires a supervisory system able to perform automatic tasks, as well as to provide tools that allow users to monitor the state of the system. The first original work presented on the thesis is the new supervisory system, the so called Software Supervisor. It has been designed, developed and deployed on the Advanced Virgo detector. It is a Supervisory Control and Data Acquisition (SCADA) system developed for the Advanced Virgo Suspension Control System, entirely based on TANGO, an open source toolkit for controlling hardware or software widely used in the scientific community. The Software Supervisor actively assisted the installation and the commissioning of the new electronics, and now is integral part of experiment: it is monitoring the Suspension Control System since the installation of new Advanced Virgo electronics; it supported the commissioning, and allowed to have a system ready for the LIGO-Virgo joint observation run in August 2017, when the first binary neutron star merge has been detected, opening a new era of astronomy.

Even if the DSP-based boards have been designed specifically for the digital control of the Superattenuators, they are suitable also to perform on-line data analysis on the Virgo data. A key feature is that they are deeply integrated into the experiment, as they are part of the sensing and control system of the interferometer. This means that they already acquire the Virgo data with almost zero latency. In the third and last part of the thesis we describe the second original work, a feasibility study about how the large computational power of a DSP-based system can be exploited to perform an on-line low-latency search for gravitational waves from compact binary mergers. The lowest latency of the existing searches for signals from compact binary mergers is currently about 16 s, achieved by the GstLAL pipeline. Most of this delay is due to the reconstruction of the gravitational wave strains and to the data transfer from the interferometer sites to the computing farms. We show how a simple matched filter search performed on DSP-based board can dramatically reduce this time to few seconds, or even better in case of a different architecture of the pipeline.
File