Undergraduate Research at Jefferson Lab

Preparation of Hydra for Halls A and C

Student: Austin Kuntz

School: Case Western Reserve University

Mentored By: Alexandre Camsonne and Holly Szumila-Vance

The detector halls at Jefferson Lab produce large amounts of data, and shift workers must regularly check hundreds of plots for any indication of hardware or software issues. Hydra is a software tool developed for Jefferson Lab's detector halls to analyze thousands of data plots efficiently. It utilizes machine learning, being trained on examples of "good" and "bad" plots and showing shift workers ones it deems problematic. This project focused on streamlining the data preparation process and helping produce plots. C++ scripts were created to process large numbers of files and organize plots into relevant directories; they take advantage of the consistency of ROOT files to allow more general use. This significantly reduces the time and effort needed for data preparation and monitoring, allowing shift workers to focus on other tasks. Hydra's ongoing efficiency improvements promise further optimization of data quality monitoring, alleviating the burden on personnel.