Machine Learning Based Energy Prediction of Hybrid Solar-Wind Energy Harvesters Using MPPT

The ever-growing demand for reliable and sustainable energy solutions has shifted the trend from standalone power systems towards higher efficient Hybrid Solar-Wind Power Systems. This paper presents a comparative analysis of a machine learning based energy prediction for a hybrid solar-wind energy harvester using MPPT. The solar-wind hybrid systems interfaced with MPPT can simultaneously capture both solar and wind power, and machine learning algorithms are involved for optimized multi-mode operation across different environmental conditions. The models used including Support Vector Machine (SVM), Random Forest Regression (RF), K-Nearest Neighbors (KNN), Linear Regression (LR), Decision Tree (DT), and Linear Discriminant Analysis (LDA) to analyze the performance of hybrid energy harvester. This type of energy combination is beneficial for coastal areas and optimal solution for sustainable energy production from renewable sources.