Authors: Liang Zhao (U. of Michigan), Susan T. Lepri (U. of Michigan), and Henry Han (Baylor University, TX)

In-situ observations of solar wind plasma exhibit statistical differences according to their coronal origins. These in-situ conditions are a direct result of various processes such as ionization and acceleration that occur in the inner corona. Machine learning (ML) and Artificial Intelligence (AI) methods have been successful in characterizing  solar wind in-situ observations using unsupervised deep clustering and dimensionality reduction techniques, but it remains unclear as to how solar wind data embedding and downstream clustering could be improved while providing better interpretability in ML/AI process. In this study, we explore the impact of distance metrics on solar wind in-situ data (ACE and Solar Orbiter) clustering. We evaluate the metric performance by applying it to dimension-reduction-stacking and deep clustering techniques and comparing it with state-of-the-art methods using solar wind in-situ measurements. Our work demonstrates the potential for customized distance metrics to improve the interpretability and performance of deep clustering approaches applied in solar wind in-situ observations.