Implementing Nonlinear Control in a Classical Experiment to Reduce Measurement Noise

Document #:

LIGO-T2300288-v1

Document type:

T - Technical notes

Abstract:

This is a SURF project in the summer of 2023. We aim to implement a nonlinear controller based on Reinforcement Learning to achieve better performance than the linear controllers. The task is to control the temperature of a test mass over long time spans (>10 days).

Files in Document:

• SURF Final Report (LIGO_SURF_Final_Report-2.pdf, 6.6 MB)

Other Files:

• SURF First Interim Report (interim1.pdf, 1.0 MB)
• SURF Project Proposal (Proposal.pdf, 80.4 kB)

Topics:

• Machine Learning
• State Control and Monitoring

Authors:

• Andrei Diaconu
• Rana Adhikari

Keywords:

SURF23 nonlinear controls

Referenced by:

• LIGO-T2300416: Final reports from LIGO SURF program, Summer 2023