Daily Energy Production Variance Analysis and Statistical Distribution Modeling for Small-Scale Solar Photovoltaic Systems
Arellanos-Tafur E. · Proceedings of the LACCEI international Multi-conference for Engineering, Education and Technology (2026) · DOI: 10.18687/laccei2026.1.1.1720
Resumen
This research presents a detailed statistical analysis of daily energy production variance for the Carhuaquero Photovoltaic Solar Plant, a 565.25 kW small-scale solar installation operated by KONDU SAC in Peru. Using a complete annual dataset spanning from May 17, 2024, to May 16, 2025, this study examines the temporal variability and statistical distribution characteristics of photovoltaic energy generation. The analysis encompasses 365 consecutive daily observations, providing robust insights into production patterns and operational efficiency metrics. Statistical modeling techniques were applied to characterize the probability distributions governing daily energy output, with particular emphasis on variance analysis and distribution fitting methodologies. The results indicate that daily energy production exhibits considerable variability, with values ranging from 0.05 to 6.10 MWh/day and a coefficient of variation of 41.2%. Seasonal patterns are identified through monthly aggregation analysis, revealing distinct performance variations associated with meteorological conditions typical of the Andean region. Distribution modeling shows that energy production follows a right-skewed, leptokurtic distribution, best approximated by a Gamma model, reflecting the alternation between overcast and clear-sky conditions. These findings contribute to an improved understanding of small-scale photovoltaic system performance characteristics and provide valuable inputs for energy forecasting models and grid integration strategies. Furthermore, the results support enhanced operational planning and maintenance scheduling for similar solar installations under comparable geographical and climatic conditions.
Instituciones: Universidad Continental