Search

Time series forecasting of meteorological variables, such as the hourly air temperature, has multiple benefits for industry, agriculture, and the environment. Due to the high accuracy required for the associated short-term predictions, traditional methods cannot satisfy the requirements and generally ignore spatial dependencies. This paper proposes a deep Graph Convolutional Long Short Term Memory Neural Network (GCN-LSTM) technique to tackle the time series prediction problem in air temperature forecasting. In the proposed methodology, temporal and spatial-based imputation approaches have been employed to recover the weather variables missing values. The proposed approach is validated using real, open weather data from 37 meteorological stations in Spain. Performed analysis indicates that GCN-LSTM showed superior performance when compared with various state-of-the-art Deep Learning based models found in the literature, resulting in a more robust and computationally efficient model for forecasting air temperature in many meteorological stations simultaneously. [ABSTRACT FROM AUTHOR]

Wind power is one of the most important renewable energy sources worldwide. However, integrating these technologies in electricity markets faces challenges due to the intermittent wind speed, resulting in potential imbalances within power grids and subsequent inefficiencies. Traditional approaches for modeling the uncertainty simulate fixed wind power values in operational and planning models, which work well for the short term but are inadequate for medium-term decision-making due to lower forecasting accuracy and the inability to differentiate the imbalance risk faced in each simulated value. This paper proposes a novel methodology for incorporating wind energy imbalance risk into medium-term generation planning models. For this, we first simulate realistic wind power scenarios and perform a fitting distribution process to characterize the hourly wind power behavior. Then, we model the imbalance risk using quadratic-uncertainty cost functions for each simulated value. Subsequently, we conduct a temporal aggregation to reduce the computational burden, preserving the conceptual framework of the imbalance risk modeling. Finally, we propose a medium-term generation planning model that includes the imbalance risk model and the temporal aggregation strategy. Our approach is evaluated in a real-size case study formed by Spain, Portugal, and France's electricity markets, demonstrating accurate wind power imbalance risk modeling and lowered planning costs. • A new methodology to include wind energy imbalance risk in generation planning models. • Introducing a fitting process for modeling the distribution asymmetries of wind power. • Proposing a temporal aggregation method for efficient imbalance risk modeling. • The methodology is validated in Spanish, Portuguese, and French electricity markets. [ABSTRACT FROM AUTHOR]

Wind power has been increasing its participation in electricity markets in many countries around the world. Due to its economical and environmental benefits, wind power generation is one of the most powerful technologies to deal with global warming and climate change. However, as wind power grows, uncertainty in power supply increases due to wind intermittence. In this context, accurate wind power scenarios are needed to guide decision-making in power systems. In this paper, a novel methodology to generate realistic wind power scenarios for the long term is proposed. Unlike most of the literature that tackles this problem, this paper is focused on the generation of realistic wind power production scenarios in the long term. Moreover, spatial-temporal dependencies in multi-area markets have been considered. The results show that capturing the dependencies at the monthly level could improve the quality of scenarios at different time scales. In addition, an evaluation at different time scales is needed to select the best approach in terms of the distribution functions of the generated scenarios. To evaluate the proposed methodology, several tests have been made using real data of wind power generation for Spain, Portugal and France. [ABSTRACT FROM AUTHOR]