Photovoltaic (PV) power forecasting urges in economic and secure operations of power systems. To avoid an inaccurate individual forecasting model, we propose an approach for a one-day to three-day ahead PV power hourly forecasting based on the stacking ensemble model with a recurrent neural network (RNN) as a meta-learner. The proposed approach is built by using real weather data and forecasted weather data in the training and testing stages, respectively. To accommodate uncertain weather, a daily clustering method based on statistical features, e.g., daily average, maximum, and standard deviation of PV power is applied in the data sets. Historical PV power output and weather data are used to train and test the model. The single learner considered in this research are artificial neural network, deep neural network, support vector regressions, long short-term memory, and convolutional neural network. Then, RNN is used to combine the forecasting results of each single learner. It is also important to observe the best combination of the single learners in this paper. Furthermore, to compare the performance of the proposed method, a random forest ensemble instead of RNN is used as a benchmark for comparison. Mean relative error (MRE) and mean absolute error (MAE) are used as criteria to validate the accuracy of different forecasting models. The MRE of the proposed RNN ensemble learner model is 4.29%, which has significant improvements by about 7–40%, 7–30%, and 8% compared to the single models, the combinations of fewer single learners, and the benchmark method, respectively. The results show that the proposed method is promising for use in real PV power forecasting systems.

One of the most critical aspects of integrating renewable energy sources into the smart grid is photovoltaic (PV) power generation forecasting. This ensemble forecasting technique combines several forecasting models to increase the forecasting accuracy of the individual models. This study proposes a regression-based ensemble method for day-ahead PV power forecasting. The general framework consists of three steps: model training, creating the optimal set of weights, and testing the model. In step 1, a Random forest (RF) with different parameters is used for a single forecasting method. Five RF models (RF1, RF2, RF3, RF4, and RF5) and a support vector machine (SVM) for classification are established. The hyperparameters for the regression-based method involve learners (linear regression (LR) or support vector regression (SVR)), regularization (least absolute shrinkage and selection operator (LASSO) or Ridge), and a penalty coefficient for regularization (λ). Bayesian optimization is performed to find the optimal value of these three hyperparameters based on the minimum function. The optimal set of weights is obtained in step 2 and each set of weights contains five weight coefficients and a bias. In the final step, the weather forecasting data for the target day is used as input for the five RF models and the average daily weather forecasting data is also used as input for the SVM classification model. The SVM output selects the weather conditions, and the corresponding set of weight coefficients from step 2 is combined with the output from each RF model to obtain the final forecasting results. The stacking recurrent neural network (RNN) is used as a benchmark ensemble method for comparison. Historical PV power data for a PV site in Zhangbin Industrial Area, Taiwan, with a 2000 kWp capacity is used to test the methodology. The results for the single best RF model, the stacking RNN, and the proposed method are compared in terms of the mean relative error (MRE), the mean absolute error (MAE), and the coefficient of determination (R2) to verify the proposed method. The results for the MRE show that the proposed method outperforms the best RF method by 20% and the benchmark method by 2%.

As meteorological conditions can be unique in different countries and may have influence on electricity demand, providing demand model to analyze characteristic of demand is useful as obtained information can be used to manage related power systems better. This paper proposes regression based demand model to identify typical characteristic of demand in Indonesia more detail. Three different demand areas (Racing Area, Poltek-Area, and Paropo Area) in Makassar, Indonesia including their total demand (Total-Area) are analyzed by creating demand model. The demands are correlated with meteorological parameters (temperature functions, relative humidity, and wind speed) and holidays. Individual characteristics are firstly observed to obtain main drivers and their typical effect on each demand area. Furthermore, general characteristics are analyzed to find common characteristic of demand such as what variables influence electricity demand generally. Several options for model are calculated and assessed by statistical tests to get best model. Results indicate more information concerning characteristic of demands can be revealed by models which are well validated. Each demand area has individual characteristic as demand drivers and their effect are relatively different between areas. Other results concerning general characteristic confirm temperature functions, relative humidiy, and holidays are important driver for demand. The variables are quite good to explain electricity demand generally as adjusted coefficient of determination of model (R 2 , ) is 76.42%.An electric power system is expected can effectively service load demand in all time during its operation. To achieve such expected condition, knowledge about characteristics of connected demand in the system is an important thing, as it can be used by power utilities to manage their power systems better. By performing electricity demand analysis particularly characteristic analysis, typical information such as key drivers for demand and how far their effect under a certain condition are possibly known. One of the method that can be used for the task above is regression approach. However, providing a good demand model as analysis tool is not an easy task as load 978·1·4799-6432-1114/$31.00 mOl41EEE 383