• Energy Research

The economic viability of renewable energy is deteriorating due to its curtailment in power systems. Therefore, it is imperative to forecast curtailments for more effective utilization. To alleviate this issue, in this paper, we propose artificial intelligent-based models to accurately predict wind and solar power curtailments (WSPCs), which have not been investigated before. In this regard, a prediction methodology is developed using different types of machine learning (ML) methods and evaluated based on both hold-out (HO) and cross-validation (CV) approaches. The ML methods considered include regression trees (RT), gradient boosting trees (GBT), random forest (RF), feed-forward artificial neural networks (ANN), long short-term memory (LSTM), and support vector machines (SVR). The prediction models are trained based on eight input features, including load demands, the output power of thermal power plants, nuclear units, solar farms, wind turbines, biomass/geothermal units, large hydro units, power imports, and WSPC as two target variables. Based on historical data, i.e., hourly records of California independent system operator (ISO), the predictive models are validated, and the optimal hyperparameters are chosen using Bayesian optimization for each model to attain the best results. Among all the models, the RF model results in the minimum prediction errors and thus the best performance by implementing the proposed CV approach. The obtained results demonstrate the effectiveness of the proposed models in the prediction of WSPCs.

Abstract Hydrogen is considered a potential game changer for world energy systems and a solution to climate change concerns, as it generates zero waste and it is suited for power generation and transportation. Despite its several advantages, there are significant technical challenges in deploying a stable hydrogen economy including improving its process efficiencies, lowering production costs, maintaining cost-effective transmission and distribution, and exploiting inexpensive and sustainable feedstocks. In this context, a detailed study was conducted to analyze the production sources, technologies, storage and transport systems, and global potential exportable feedstocks to produce hydrogen. A comprehensive analysis of current hydrogen production technologies with their energy efficiencies and hydrogen selling prices was reported in this study. Various hydrogen production technologies with their capital investments and CO2 emissions were also presented. Potential feedstocks for hydrogen production were identified and analyzed through a product space model, which characterizes a network of global exportable products based on their similarities and productive knowledge. It was established that the hydrogen production feedstocks and sources currently used are primarily available in six countries: the United States of America, France, Russia, Sweden, the Netherlands, and Spain. Broadly, the results revealed that the United States of America and Russia shared the highest hydrogen feedstock exports, indicating a higher probability of hydrogen production in these countries. Except for Russia, all the studied countries fell in the most desired quadrant, indicating that they can move in all product space directions to exploit unexplored hydrogen feedstocks for better sustainable economic growth.

By growing interest in hydrogen vehicles (HVs), hydrogen fueling stations (HFSs) that convert electric power to hydrogen to supply HVs have emerged as a new asset for power grids. To safely and consistently supply HFSs with power, the use of microgrids (MGs), including various flexible generation units, is considered to be a reliable choice. This paper proposes a competence MG scheduling model. In this model, an optimal coordination of HFSs with demand response (DR), energy storage systems (ESS), and appropriate multi-market mechanisms is addressed. Also, a reformulated version of a risk-constrained stochastic scheduling (RSS) model is used to minimize the MG operation cost. The uncertainties associated with the real-time market price, renewables, electrical loads, and HVs are handled by considering conservativeness parameters. In this model, linearized AC optimal power flow (ACOPF) equations are included in the mixed-integer linear programming (MILP) problem to satisfy the security of MG operation. The proposed model is examined on a 21-bus MG while considering various case studies. The results show that the operation of HFSs provides a profit of 254.5$/day by selling hydrogen for MG operator. In addition, it is found that developing the HFS technology can reduce the total daily operation costs of MG by up to 9.6%. It is also shown that participation of MG in both day-ahead and real-time markets leads to a reduction of 11.7% in the operating costs. Moreover, we show that employing DR programs leads to operation cost reduction and load flattening during high-demand hours. The security constraints keep the voltage between 0.97 p.u. and 1 p.u and the loss of lines within a reasonable range. Finally, a comprehensive comparison between our perceptive RSS with traditional stochastic scheduling and conservative RSS is carried out to show the effectiveness of the proposed method.

Increasing wind and solar generation in power grids leads to more renewable power curtailments in some periods of time due to the fast and unpredictable variations of their outputs. The utilization of these sources for energy storage can unlock huge potential benefits. Therefore, aiming at minimizing the curtailments of renewable power from the viewpoint of an independent system operator (ISO), in this paper, we propose deep learning-driven optimal sizing and operation of alkaline water electrolyzers (AWE) and battery energy storage systems (BESS). For this purpose, a set of actual renewable power curtailment data of California ISO was fully investigated, and deep learning forecast methods were employed to determine the prediction error and its probability distribution function (PDF). Using the fitted PDF, a set of scenarios was generated and reduced to some accurate and probable ones. Consequently, a two-stage scenario-based stochastic model was proposed to determine the optimal planning of this system, and a penalty variable was defined in the second stage to maximize the utilization of curtailed renewable energy sources (RESs). The learning results showed that the prediction errors were minimized using the gated recurrent unit (GRU) method. It was also shown that 97% of curtailments were utilized using AWEs with annual costs of $233.55 million, which had 63.5% fewer costs than using BESSs. Furthermore, using AWEs reduced operational expenses by 89.1% compared with using BESSs, owing to their operational benefits.