Search Results (4)

Photovoltaic panels have become a promising solution for generating renewable energy and reducing our reliance on fossil fuels by capturing solar energy and converting it into electricity. The effectiveness of this conversion depends on several factors, such as the quality of the solar panels and the amount of solar radiation received in a specific region. This makes accurate solar irradiance forecasting essential for planning and managing efficient solar power systems. This study examines the application of machine learning (ML) models for accurately predicting global horizontal irradiance (GHI) using a three-year dataset from six distinct photovoltaic stations: NELHA, ULL, HSU, RaZON+, UNLV, and NWTC. The primary aim is to identify optimal shared features for GHI prediction across multiple sites using a 30 min time shift based on autocorrelation analysis. Key features identified for accurate GHI prediction include direct normal irradiance (DNI), diffuse horizontal irradiance (DHI), and solar panel temperatures. The predictions were performed using tree-based algorithms and ensemble learners, achieving $R^2$ values exceeding 95% at most stations, with NWTC reaching 99%. Gradient Boosting Regression (GBR) performed best at NELHA, NWTC, and RaZON, while Multi-Layer Perceptron (MLP) excelled at ULL and UNLV. CatBoost was optimal for HSU. The impact of time-shifting values on performance was also examined, revealing that larger shifts led to performance deterioration, though MLP performed well under these conditions. The study further proposes a stacking ensemble approach to enhance model generalizability, integrating the strengths of various models for more robust GHI prediction. Full article

The intermittent nature of renewable energy sources (RES) restricts their widespread applications and reliability. Nevertheless, with advancements in the field of artificial intelligence, we can predict the variations in parameters such as wind speed and solar irradiance for the short, medium and long terms. As such, this research attempts to develop a machine learning (ML)-based framework for predicting solar irradiance at Muscat, Oman. The developed framework offers a methodological way to choose an appropriate machine learning model for long-term solar irradiance forecasting using Python’s built-in libraries. The five different methods, named linear regression (LR), seasonal autoregressive integrated moving average with exogenous variables (SARIMAX), support vector regression (SVR), Prophet, k-nearest neighbors (k-NN), and long short-term memory (LSTM) network are tested for a fair comparative analysis based on some of the most widely used performance evaluation metrics, such as the mean square error (MSE), mean absolute error (MAE), and coefficient of determination (R²) score. The dataset utilized for training and testing in this research work includes 24 years of data samples (from 2000 to 2023) for solar irradiance, wind speed, humidity, and ambient temperature. Before splitting the data into training and testing, it was pre-processed to impute the missing data entries. Afterward, data scaling was conducted to standardize the data to a common scale, which ensures uniformity across the dataset. The pre-processed dataset was then split into two parts, i.e., training (from 2000 to 2019) and testing (from 2020 to 2023). The outcomes of this study revealed that the SARIMAX model, with an MSE of 0.0746, MAE of 0.2096, and an R² score of 0.9197, performs better than other competitive models under identical datasets, training/testing ratios, and selected features. Full article

The growing interest in solar energy stems from its potential to reduce greenhouse gas emissions. Global horizontal irradiance (GHI) is a crucial determinant of the productivity of solar photovoltaic (PV) systems. Consequently, accurate GHI forecasting is essential for efficient planning, integration, and optimization of solar PV energy systems. This study evaluates the performance of six machine learning (ML) regression models—artificial neural network (ANN), decision tree (DT), elastic net (EN), linear regression (LR), Random Forest (RF), and support vector regression (SVR)—in predicting GHI for a site in northern Saudi Arabia known for its high solar energy potential. Using historical data from the NASA POWER database, covering the period from 1984 to 2022, we employed advanced feature selection techniques to enhance the predictive models. The models were evaluated based on metrics such as R-squared (R²), Mean Squared Error (MSE), Root Mean Squared Error (RMSE), Mean Absolute Percentage Error (MAPE), and Mean Absolute Error (MAE). The DT model demonstrated the highest performance, achieving an R² of 1.0, MSE of 0.0, RMSE of 0.0, MAPE of 0.0%, and MAE of 0.0. Conversely, the EN model showed the lowest performance with an R² of 0.8396, MSE of 0.4389, RMSE of 0.6549, MAPE of 9.66%, and MAE of 0.5534. While forward, backward, and exhaustive search feature selection methods generally yielded limited performance improvements for most models, the SVR model experienced significant enhancement. These findings offer valuable insights for selecting optimal forecasting strategies for solar energy projects, contributing to the advancement of renewable energy integration and supporting the global transition towards sustainable energy solutions. Full article

Solar irradiation (Rs) is the electromagnetic radiation energy emitted by the Sun. It plays a crucial role in sustaining life on Earth by providing light, heat, and energy. Furthermore, it serves as a key driver of Earth’s climate and weather systems, influencing the distribution of heat across the planet, shaping global air and ocean currents, and determining weather patterns. Variations in Rs levels have significant implications for climate change and long-term climate trends. Moreover, Rs represents an abundant and renewable energy resource, offering a clean and sustainable alternative to fossil fuels. By harnessing solar energy, we can actively reduce greenhouse gas emissions. However, the utilization of Rs comes with its own challenges that must be addressed. One problem is its variability, which makes it difficult to predict and plan for consistent solar energy generation. Its intermittent nature also poses difficulties in meeting continuous energy demand unless appropriate energy storage or backup systems are in place. Integrating large-scale solar energy systems into existing power grids can present technical challenges. Rs levels are influenced by various factors; understanding these factors is crucial for various applications, such as renewable energy planning, climate modeling, and environmental studies. Overcoming the associated challenges requires advancements in technology and innovative solutions. Measuring and harnessing Rs for various applications can be achieved using various devices; however, the expense and scarcity of measuring equipment pose challenges in accurately assessing and monitoring Rs levels. In order to address this, alternative methods have been developed with which to estimate Rs, including artificial intelligence and machine learning (ML) models, like neural networks, kernel algorithms, tree-based models, and ensemble methods. To demonstrate the impact of feature selection methods on Rs predictions, we propose a Multivariate Time Series (MVTS) model using Recursive Feature Elimination (RFE) with a decision tree (DT), Pearson correlation (Pr), logistic regression (LR), Gradient Boosting Models (GBM), and a random forest (RF). Our article introduces a novel framework that integrates various models and incorporates overlooked factors. This framework offers a more comprehensive understanding of Recursive Feature Elimination and its integrations with different models in multivariate solar radiation forecasting. Our research delves into unexplored aspects and challenges existing theories related to solar radiation forecasting. Our results show reliable predictions based on essential criteria. The feature ranking may vary depending on the model used, with the RF Regressor algorithm selecting features such as maximum temperature, minimum temperature, precipitation, wind speed, and relative humidity for specific months. The DT algorithm may yield a slightly different set of selected features. Despite the variations, all of the models exhibit impressive performance, with the LR model demonstrating outstanding performance with low RMSE (0.003) and the highest R2 score (0.002). The other models also show promising results, with RMSE scores ranging from 0.006 to 0.007 and a consistent R2 score of 0.999. Full article