Particle swarm optimization of the sensitivity of a cryogenic gravitational wave detector

TL;DR

This paper demonstrates that particle swarm optimization can effectively enhance the sensitivity and sky localization of cryogenic gravitational wave detectors like KAGRA by optimizing multiple parameters.

Contribution

It introduces the application of particle swarm optimization to tune parameters of cryogenic gravitational wave detectors, improving their performance.

Findings

Binary neutron star inspiral range improved by 10%

Sky localization of GW170817-like binaries improved by a factor of 1.6

Optimization of seven parameters yields significant sensitivity gains

Abstract

Cryogenic cooling of the test masses of interferometric gravitational wave detectors is a promising way to reduce thermal noise. However, cryogenic cooling limits the incident power to the test masses, which limits the freedom of shaping the quantum noise. Cryogenic cooling also requires short and thick suspension fibers to extract heat, which could result in the worsening of thermal noise. Therefore, careful tuning of multiple parameters is necessary in designing the sensitivity of cryogenic gravitational wave detectors. Here, we propose the use of particle swarm optimization to optimize the parameters of these detectors. We apply it for designing the sensitivity of the KAGRA detector, and show that binary neutron star inspiral range can be improved by 10%, just by retuning seven parameters of existing components. We also show that the sky localization of GW170817-like binaries can be further improved by a factor of 1.6 averaged across the sky. Our results show that particle swarm optimization is useful for designing future gravitational wave detectors with higher dimensionality in the parameter space.