Makine öğrenme yaklaşımlarına dayalı güneş enerjisi gücü tahmin sistemi

Solar energy is a crucial and sustainable source for electricity generation. The optimization of solar energy systems is essential to maximize their efficiency and utilization. This thesis explores the application of machine learning algorithms, specifically Support Vector Machine (SVM) and Artificial Neural Network (ANN), for the purpose of solar energy optimization. The research focuses on predicting various meteorological parameters that significantly impact solar energy generation, including air temperature, surface humidity, radiance intensity, ozone levels, total precipitable water vapor, and wind speed. The performance of the SVM and ANN models is evaluated using the Root Mean Squared Error (RMSE) metric, which quantifies the average deviation between the predicted and actual values. The results demonstrate the effectiveness of the SVM and ANN models in accurately predicting meteorological parameters for solar energy optimization. The SVM model achieves an overall RMSE ranging from 2.14 to 16.49, while the ANN model exhibits superior performance with RMSE ranging from 1.80 to 14.23. These metrics signify the model's ability to capture variations in meteorological parameters and provide reliable predictions. Moreover, the thesis delves into the data preprocessing techniques utilized, such as feature scaling using StandardScaler, and the hyperparameter tuning process, which enhances the models' performance. Visualization techniques, including line plots and scatter plots, are employed to facilitate the interpretation and analysis of the results. The findings of this study offer valuable insights into the optimization of solar energy systems using machine learning techniques. Accurate prediction of meteorological parameters enables the efficient harnessing of solar energy, leading to increased sustainability and improved energy vii generation. This research contributes to ongoing endeavors in the field of renewable energy and paves the way for further exploration and advancements in solar energy optimization.