Date of Award
Spring 2025
Project Type
Departmental Honors Paper/Project
Department
Department of Physics & Astronomy
First Advisor
Todd C. Hillwig
Abstract
The Advanced Laser Interferometer Gravitational Wave Observatory (aLIGO) made its first detection of gravitational waves in 2015. Since then, the rate of event detection has only increased, with a detection being made every 2-3 days during the current observing run, O4. This rapid influx of data has the potential to create bottle-necks in data analysis efforts, and can delay the scientific progress which require those efforts. Traditional gravitational wave data analysis techniques, such as matched filtering, require extremely large template banks of synthetic gravitational waveforms and can often fail to provide meaningful limits on system parameters. Not only is this process computationally intensive, but it also requires a preprocessing of the data. Given the amount of effort this analysis takes, the results leave much to be desired. With recent advances in machine learning, there have been hopes that many of the bottle-necks currently afflicting "big data" may be effectively resolved. This has proven to hold true in many areas of science, even gravitational physics. Machine learning neural networks have already demonstrated the ability to flag whether or not a signal is buried within noise, denoise time-series data and extract the signal, and make accurate parameter estimates. However, the majority of these neural networks still rely on data preprocessing or transformation prior to analysis, raising the question of whether a neural network could instead directly intake raw, noisy, unprocessed time-series signals from a laser interferometer and accurately estimate key system parameters. This thesis covers the development, performance, and analysis of a neural network that, given a time-series of raw, noisy, unprocessed signal from a laser interferometer, can accurately predict the chirp mass of the binary black hole (BBH) system that produced the signal. This advancement has the capability of significantly increasing the computational efficiency of gravitational wave data analysis and yielding more accurate parameter estimates than current techniques.
Recommended Citation
Scheel, Lane, "Parameter Estimation of Binary Black Hole Coalescence Using LSTM Neural Networks" (2025). Undergraduate Honors Papers. 10.
https://scholar.valpo.edu/undergrad_capstones/10
