Study of a Hybrid Photovoltaic-Wind Smart Microgrid using Data Science Approach

TL;DR

This study analyzes a hybrid photovoltaic-wind microgrid in Peru using data science to identify patterns, seasonality, and correlations, aiming to optimize energy management and improve microgrid performance.

Contribution

It introduces a data-driven analysis of renewable resources and demand patterns in a real-world microgrid, providing insights for better sizing and management strategies.

Findings

Renewable resources and demand show periodicity and seasonality.

Electricity demand increased from 0.7kW to 1.2kW between 2019 and 2021.

Power outages correlate with low resource availability and battery failures.

Abstract

In this paper, a smart microgrid implemented in Paracas, Ica, Peru, composed of 6kWp PV + 6kW Wind and that provides electricity to a rural community of 40 families, was studied using a data science approach. Real data of solar irradiance, wind speed, energy demand, and voltage of the battery bank from 2 periods of operation were studied to find patterns, seasonality, and existing correlations between the analyzed data. Among the main results are the periodicity of renewable resources and demand, the weekly behavior of electricity demand and how it has progressively increased from an average of 0.7kW in 2019 to 1.2kW in 2021, and how power outages are repeated at certain hours in the morning when resources are low or there is a failure in the battery bank. These analyzed data will be used to improve sizing techniques and provide recommendations for energy management to optimize the performance of smart microgrids.