Energy Performance Clustering and Data Visualization for Solar-Wind Hybrid Energy Systems

This research article proposes a methodology based on data analytics, to compute well-known energy performance ratios like Capacity Factor (CF), Performance Yields (Yr) and Performance Ratio (PR) used for evaluating solar-wind hybrid energy systems. These terms are developed to study performance in renewable energy systems considering an estimation of energy resources. The methodology implemented is divided twofold, first, we deploy a recognized unsupervised machine learning technique as, k-means data clustering algorithm, considering the following feature space: time, solar radiation, wind speed, and temperature, which are renewable energy potentials available in the campus at Universidad de la Salle in Bogot ' a, Colombia, acquired by a local weather station which is considered as the case study. Second, according to this data-driven model, the performance factors are computed, yielding technological solutions and recommendations considering the data collected, analyzed, and visualized.