Answers are grounded in this paper and cite their sources.
The regionalized energy practices profit greatly from the predictive analysis. The estimation of influential independent variables of solar energy production and its consumption is needed to predict for scheduling the energy harvesting. It helps to diagnose energy efficiency and manage electrical and thermal grids. In the current research work, a total of four machine learning (ML) approaches KNN, MLP, M5, and SMV Reg are applied to predict solar panel efficiency. The input data i.e. Time (mins), Temp (oC), Wind speed (km/h), Humidity (%), and Air pressure (mbar) are filtered to exclude the outliers and extreme values using various algorithms. The most sensitive input parameter with respect to the output voltage has been found (Time > Temp > Humidity > Air pressure) by performing data ranking. Wind speed has a minimal effect on the final output value. Among the ML models, the k Nearst Neighbours Algorithm (KNN) produces the best correlation coefficient value (0.9772) with relatively smaller time consumption.
No takes yet. Share an insight, caveat, or question.
Hossain et al. (2024) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: