Solar power has rapidly become an increasingly important energy source in many countries over recent years; however, the intermittent nature of photovoltaic (PV) power generation has a significant impact on existing power systems. To reduce this uncertainty and maintain system security, precise solar power forecasting methods are required. This study summarizes and compares various PV power forecasting approaches, including time-series statistical methods, physical methods, ensemble methods, and machine and deep learning methods, the last of which there is a particular focus. In addition, various optimization algorithms for model parameters are summarized, the crucial factors that influence PV power forecasts are investigated, and input selection for PV power generation forecasting models are discussed. Probabilistic forecasting is expected to play a key role in the PV power forecasting required to meet the challenges faced by modern grid systems, and so this study provides a comparative analysis of existing deterministic and probabilistic forecasting models. Additionally, the importance of data processing techniques that enhance forecasting performance are highlighted. In comparison with the extant literature, this paper addresses more of the issues concerning the application of deep and machine learning to PV power forecasting. Based on the survey results, a complete and comprehensive solar power forecasting process must include data processing and feature extraction capabilities, a powerful deep learning structure for training, and a method to evaluate the uncertainty in its predictions.

In recent years, wind energy has become a competitively priced source of energy around the world, which has created increasing challenges for system operators. Accurate wind power generation forecasting plays an important role in power systems to improve the reliable and efficient operation. Therefore, numerous artificial intelligent methods such as machine learning and deep learning have been considered as solutions for accurate wind power forecasts. In addition to deterministic forecasting, the probabilistic forecasting becomes more important, because it indicates the level of uncertainty. In this paper, a hybrid forecasting model considering different Numerical Weather Prediction (NWP) models and the XGBoost training model is proposed for short-term wind power forecasting. The proposed forecasting algorithm includes data preprocessing, in which an autoencoder model is used to reduce the dimension of 20 NWP ensembles. The performance of the proposed method is investigated using historical wind power measurements and NWP results by the Taiwan Central Weather Bureau (CWB); the NWP includes spot wind speeds from WRFD, RWRF, and ensemble wind speeds from WEPS. Based on the forecasting results, the proposed model produces better performance and forecasting accuracy among other forecasting models, which reveals the importance of data preprocessing using autoencoders and the use of deep learning models in deterministic or probabilistic forecasts.

Photovoltaic power has become one of the most popular forms of energy owing to the growing consideration of environmental factors; however, solar power generation has brought many challenges for power system operations. With regard to optimizing safety and reducing the costs of power system operations, an accurate and reliable solar power forecasting model would be a significant step forward. This study proposes a deep learning method to improve the performance of short-term one-hour-ahead solar power forecasting, which includes data preprocessing, feature engineering, kernel principal component analysis, a gated recurrent unit network training mode based on time-of-day classification, and postprocessing with error correction. Both historical solar power, solar irradiance, and numerical weather prediction (NWP) data, such as temperature, irradiance, rainfall, wind speed, air pressure, and humidity, were used as the input dataset in this work. As a case study, the measured power from ten PV sites in Taiwan were collected and predicted with a one-hour resolution. The normalized root mean squared error and normalized mean absolute percent error were chosen to evaluate the performance of the forecasting models. Compared with other benchmark models, including ANN, LSTM, XGBoost, and single GRU, the experimental results showed the proposed model's superior performance. Furthermore, the importance of data preprocessing and postprocessing based on error correction was demonstrated.

Forecasting solar photovoltaic (PV) power generation has become important due to the chaotic and intermittent nature of solar power. With the development of artificial intelligence (AI) techniques, various deep learning algorithms have been shown to be very powerful for renewable power forecasts. This study aims to develop a probabilistic forecasting approach that leverages a hybrid model, named Transformer-Lower Upper Bound Estimation (Transformer-LUBE) with data processing techniques, for one-day-ahead PV power generation forecasts. Especially, this study presents a novel methodology that integrates an effective imputation technique, a modified Transformer architecture, and a combined post-processing technique. During the pre-processing stage, missing data were imputed using XGBoost with the combination of predictive mean matching (PMM) and bootstrapping, then Pearson Correlation Coefficient was utilized to compute the correlation between the input features. In contrast to the original Transformer architecture, the proposed Transformer model incorporates additional normalization and dropout layers between blocks, and also incorporates multi-head attention mechanisms. Finally, a post-processing strategy was implemented based on the utilization of daily numerical fitting curves along with Gated Recurrent Unit (GRU) to correct the bias of the residual errors. The proposed framework was trained with historical measurement data obtained from ten solar farms located in Taiwan, as well as Numerical Weather Predictions (NWPs) data sourced from Taiwan Central Weather Bureau (CWB). The accuracy of the proposed model has been evaluated by comparing other AI models: Artificial Neural Network (ANN), Long Short-Term Memory (LSTM), GRU, and XGBoost. Through a comprehensive experimentation, the proposed framework demonstrates its superiority in accuracy and reliability.

In recent years, solar photovoltaic (PV) generation becomes one of the most relevant energies. However, the intermittent characteristics of solar generation create significant problems to power system operations. To overcome this problem, many solar power forecasting techniques have been developed, and different forecasting horizons require different methodologies. For a short-term prediction, forecasting horizons generally require numerical weather prediction models (NWP) that provide an important estimation of weather variables such as solar irradiance, temperature, wind speed, rainfall, air pressure, etc. This research proposes a machine learning model based on Kernel Principal Component Analysis (PCA)- XGBoost to improve the accuracy of one-hour-ahead solar power forecasts. The model considered the deterministic Weather Research and Forecasting (WRFD) provided by Taiwan Central Weather Bureau (CWB). Furthermore, a XGBoost model was built on an ensemble of decision trees, providing important information and appropriate results in the forecasting process.

Solar power forecasting is essential for optimizing energy management and ensuring stable grid operations. Accurately forecasting solar irradiance is a key factor to improve solar power forecasts because of a strong relationship between solar power generation and solar irradiance. However, the accuracy of solar irradiance forecasting is affected largely by the limits inherent in Numerical Weather Prediction (NWP). Thus, there exists a notable opportunity to improve the forecast beyond NWP by using data processing technologies. Among them, one is based on the classification of weather patterns. This paper aims to propose several PV power forecasting methods based on weather patterns, and to develop appropriate models for each classification. The proposed five clustering methods include the use of K-Means or SOM algorithm, a time-based classification, an amplitude threshold-based classification using PSO and GWO algorithms, and a season-based classification. Moreover, three up-to-date AI models including XGBoost, GRU, and Transformer were then applied to predict one-day-ahead PV power. Through a systematic experimentation and comparative analysis, the developed forecasting method considering weather classifications with Transformer training model achieves the highest forecasting accuracy on both deterministic and probabilistic forecasts. Furthermore, the forecasting results also reveal the potential advantages for different clustering methods. The time-based and season-based classification models can capture specific climate characteristics of different time periods and seasons, respectively.