Predicting the motion of a high-Q pendulum subject to seismic perturbations using machine learning

TL;DR

This paper demonstrates that machine learning can effectively predict and reduce seismic-induced motion of high-Q pendulums in gravitational-wave detectors, enhancing their sensitivity by significantly lowering displacement noise.

Contribution

The study introduces a machine learning approach to predict high-Q pendulum motion driven by seismic activity, achieving substantial noise reduction at resonance frequencies.

Findings

40dB reduction in displacement spectral density at 1.4Hz

6dB reduction at 11Hz

Machine learning enables significant seismic noise mitigation in gravitational-wave detectors

Abstract

The seismically excited motion of high-Q pendula in gravitational-wave observatories sets a sensitivity limit to sub-audio gravitational-wave frequencies. Here, we report on the use of machine learning to predict the motion of a high-Q pendulum with a resonance frequency of 1.4Hz that is driven by natural seismic activity. We achieve a reduction of the displacement power spectral density of 40dB at the resonant frequency 1.4Hz and 6dB at 11Hz. Our result suggests that machine learning is able to significantly reduce seismically induced test mass motion in gravitational-wave detectors in combination with corrective feed-forward techniques.