Persistent Homology in Solar Production

Abstract

Persistent homology, an algebraic topology-based mathematical framework, presents an innovative method for capturing and characterizing the inherent topological features present in time series datasets. The research aims to evaluate the efficacy of features derived from persistent homology in enhancing the accuracy and interpretability of classification models. This investigation contributes to the expanding convergence of topology and time series analysis, providing valuable insights into the potential of persistent homology for extracting information from temporal data. The study specifically focuses on the analysis of region-wise solar production data obtained from India for the year 2022. The examination of this data is conducted using R-Software, and the resulting topological properties are represented through persistent diagrams.