Estimation of Solar Panel Output based on Weather Parameters using Machine Learning Algorithms

다국어 초록 (Multilingual Abstract)

Solar energy is one of the most extensively used renewable energy sources. However, it is highly variable and needs accurate estimation for its wide range of integration into the electricity grid. Solar voltage and current are estimated in areas where only sunlight is considered as a primary solar parameter, and information about their weather conditions are unknown. Weather plays a vital role in the prediction of solar panel output. In this paper, we propose solar panel output prediction considering the solar panel and weather parameters using machine learning algorithms. We estimate the solar panel voltage and current consider the weather parameters such as temperature, humidity, rain rate, wind speed, and wind direction. For estimating the output voltage and current, Linear Regression (LR) and Artificial Neural Network (ANN) are applied on weather and solar data. The datasets are extracted from Bancroft close 49KW substation, which is placed in the UK, for three months. The performance of the given model is evaluated using two matrices Root Mean Square Error (RMSE) and Absolute Error (AE). The Neural Network shows better accuracy compared to the linear regression.

목차 (Table of Contents)

• ABSTRACT
• 1-INTRODUCTION
• 2-THE PROPOSED ARCHITECTURE FOR SOLAR PANEL OUTPUT PREDICATION
• 3-RESULTS AND DISCUSSION
• CONCULSION
• References