Encouraged by the considerable cost reduction, small-scale solar power deployment has become a reality during the last decade. However, grid integration of small-scale photovoltaic (PV) solar systems still remains unresolved. High penetration of Renewable Energy Sources (RESs) results in technical challenges for grid operators. To address this, Virtual Power Plants (VPPs) have been defined and developed to manage distributed energy resources with the aim of facilitating the integration of RESs. This paper introduces a hybrid irradiance forecasting approach aimed at facilitating the integration of PV systems into a VPP, especially when a historical irradiance dataset is exiguous or non-existent. This approach is based on Artificial Neural Networks (ANNs) and a novel similar hour-based selection algorithm, has been tested for a real PV installation, and has been validated also considering irradiance measurements from an aggregation of ground-based meteorological stations, which emulate the nodes of a VPP. Under a reduced historical dataset, the results show that the proposed similar hour-based method produces the best forecasts with regard to those obtained by the ANN-based approach. This is particularly true for one-month and two-month datasets minimizing the mean error by 16.32% and 9.07% respectively. Finally, to demonstrate the potential of the proposed approach, a comparative analysis has been carried out between the hybrid method and the most used benchmarks in the literature, namely, the persistence method and the method based on similar days. It has been demonstrated conclusively that the proposed model yields promising results regardless the length of the historical dataset.

Solar energy penetration has been on the rise worldwide during the past decade, attracting a growing interest in solar power forecasting over short time horizons. The increasing integration of these resources without accurate power forecasts hinders the grid operation and discourages the use of this renewable resource. To overcome this problem, Virtual Power Plants (VPPs) provide a solution to centralize the management of several installations to minimize the forecasting error. This paper introduces a method to efficiently produce intra-day accurate Photovoltaic (PV) power forecasts at different locations, by using free and available information. Prediction intervals, which are based on the Mean Absolute Error (MAE), account for the forecast uncertainty which provides additional information about the VPP node power generation. The performance of the forecasting strategy has been verified against the power generated by a real PV installation, and a set of ground-based meteorological stations in geographical proximity have been used to emulate a VPP. The forecasting approach is based on a Long Short-Term Memory (LSTM) network and shows similar errors to those obtained with other deep learning methods published in the literature, offering a MAE performance of 44.19 W/m² under different lead times and launch times. By applying this technique to 8 VPP nodes, the global error is reduced by 12.37% in terms of the MAE, showing huge potential in this environment.

Bad data as a result of measurement errors in secondary substation (SS) monitoring equipment is difficult to detect and negatively affects power system state estimation performance by both increasing the computational burden and jeopardizing the state estimation accuracy. In this paper a short-term load forecasting (STLF) hybrid strategy based on singular spectrum analysis (SSA) in combination with artificial neural networks (ANN), is presented. This STLF approach is aimed at detecting, identifying and eliminating and/or correcting such bad data before it is provided to the state estimator. This approach is developed to improve the accuracy of the load forecasts and it is tested against real power load data provided by electricity suppliers. Depending on the week considered, mean absolute percentage error (MAPE) values which range from 1.6% to 3.4% are achieved for STLF. Different systematic errors, such as gain and offset error levels and outliers, are successfully detected with a hit rate of 98%, and the corresponding measurements are corrected before they are sent to the control center for state estimation purposes.

This paper proposes a framework, based on the concept of virtual power plant, for promoting the effective participation of remote photovoltaic power generation installations in different energy markets. To this aim, the most significant challenge lies in providing accurate power forecasts for different lead times determined by the markets. To address this challenge, a hybrid forecast strategy based on day-ahead and intra-day prediction models is proposed. For each lead time, a point prediction is provided by choosing, from the outputs of prediction models, the value which most reduces the prediction uncertainty. The framework also requires the cloudiness forecast to improve the accuracy of the prediction, by classifying the days in three categories according to a cloud cover factor, namely, sunny, cloudy and overcast. The predictions are updated every 15 minutes by using real-time information of the day considered. Whenever the algorithm is executed, the generation forecast for the rest of the day is recalculated with a 15-minute resolution. The point prediction is provided with the corresponding confidence interval, which is modelled by a Laplacian distribution function. This interval is of particular importance in the context of energy markets as it allows the risk of penalties for any energy deviation to be modelled. The strategy is evaluated in a VPP working environment demonstrating the potential of the hybrid prediction algorithm.