Leveraging both temporal and spatial correlations to predict wind speed remains one of the most challenging and less studied areas of wind speed prediction. In this paper, the problem of predicting wind speeds for multiple sites is investigated by using the spatio-temporal correlation. We proposed a deep architecture termed predictive spatio-temporal network (PSTN), which is a unified framework integrating a convolutional neural network (CNN) and a long short-term memory (LSTM). Initially, the spatial features are extracted from the spatial wind speed matrices (SWSMs) by the CNN at the bottom of the model. Then, the LSTM captures the temporal dependencies among the spatial features extracted from contiguous time points. Finally, the predicted wind speeds are given by the last state of the top layer of the LSTM, which are generated by using the spatial features and temporal dependencies. Though composed of two kinds of architectures, PSTN is trained with one loss function in an end-to-end (E2E) manner, which can learn temporal and spatial correlations jointly. Experiments for short-term predictions are conducted on real-world data, whose results demonstrate that PSTN outperforms prior methods.

Wind speed prediction with spatio–temporal correlation is among the most challenging tasks in wind speed prediction. In this paper, the problem of predicting wind speed for multiple sites simultaneously is investigated by using spatio–temporal correlation. This paper proposes a model for wind speed prediction with spatio–temporal correlation, i.e., the predictive deep convolutional neural network (PDCNN). The model is a unified framework, integrating convolutional neural networks (CNNs) and a multi-layer perceptron (MLP). Firstly, the spatial features are extracted by CNNs located at the bottom of the model. Then, the temporal dependencies among these extracted spatial features are captured by the MLP. In this way, the spatial and temporal correlations are captured by PDCNN intrinsically. Finally, PDCNN generates the predicted wind speed by using the learnt spatio–temporal correlations. In addition, three error indices are defined to evaluate the prediction accuracy of the model on the wind turbine array. Experiment results on real-world data show that PDCNN can capture the spatio–temporal correlation effectively, and it outperforms the conventional machine learning models, including multi-layer perceptron, support vector regressor, decision tree, etc.

Owing to the recent power outages caused by extreme events, installing battery energy storage and backup generators is important to improve resiliency for a grid‐tied microgrid. In the design stage, the event occurrence time and duration, which are highly uncertain and cannot be effectively predicted, may affect the needed battery and backup generator capacity but are usually assumed to be pre‐determined in utility planning tools. This study investigates the optimal battery and backup generator sizing problem considering the stochastic event occurrence time and duration for the grid‐tied microgrid under islanded operation. The reliability requirement is quantified by the mean value of the critical customer interruption time in each stochastic islanding time window (ITW), whose length is the duration and the centre is the occurrence time. The stochastic ITW constraint is then transformed to a probability‐weighted expression to derive an equivalent Mixed Integer Linear Programming model. Numerical simulations on a realistic grid‐tied PV‐based microgrid demonstrate that the total cost is reduced by 11.5% considering the stochastic ITW, compared with the deterministic ITW under the same reliability requirement.

Leveraging multiple heterogeneous measurements to predict wind power has long been a challenging task in the electrical community. In this paper, a deep architecture incorporated with multitask learning and multimodal learning for wind power prediction, termed predictive stacked autoencoder (PSAE), is presented. PSAE is a unified framework integrating multiple stacked autoencoders (SAEs), one feature fusion layer, and one prediction terminal layer, which expands the architecture from two spatial dimensions, including the depth and width, compared to conventional prediction models. Initially, the SAEs at the bottom of PSAE extracted features from multiple kinds of measurements respectively. Following, the feature fusion layer encodes the high-order features extracted by different SAEs into a unified feature that is more informative and representative for wind power prediction. Finally, the prediction terminal layer functions as a regression machine which generates the predicted targets based on the fusion features. Trained in an end-to-end (E2E) manner, PSAE is capable of learning heterogeneous features jointly and achieving the prediction task of sequence-to-sequence (S2S). Experiments for multi-step short-term predictions are conducted on real-world data, and the results demonstrate the superiority of PSAE to prior methods.

Non-utility owned distributed energy resources (DERs) are mostly untapped currently, but they can provide many grid services such as voltage regulation and service restoration, if properly controlled, and can improve the distribution system's reliability when coordinated with utility-owned assets such as self-healing control and microgrids. This paper integrates transactive energy control into the distribution system reliability evaluation to quantitatively assess the impact of non-utility owned DERs on reliability improvement. A transactive reactive power control strategy is designed to incentivize the DERs to provide reactive power support for improving voltage profiles thus enabling additional customer load restoration during an outage. Also, an operational sequence to coordinate the non-utility owned DERs with the utility owned self-healing control and utility owned microgrids is designed and integrated into the service restoration process with the operational constraints guaranteed by checking the three-phase unbalanced power flow for post-fault network reconfiguration. The reliability indices are then calculated through a Monte Carlo simulation. The transactive reactive power control strategy is tested on a four-feeder distribution system operated by Duke Energy in the U.S. Results demonstrate that the non-utility owned DERs with the transactive control improve the reliability of both the system and critical loads by more than 30%.

Multiple severe forced oscillation events have recently been observed across the North American interconnections and around the world. These forced oscillations have caused power swings, limited power transfer capability, damaged equipment, and persisted indefinitely until the driving source was located and removed. This paper proposes a comprehensive method to mitigate forced oscillations. Once a forced oscillation is detected, a new source location algorithm can be used to locate the source based on the oscillation mode angle without requiring system topology information. If the source cannot be quickly located and removed, a control strategy can be activated to modulate the active power of utility-scale inverter-based battery energy storage systems (BESSs) to reduce the energy of forced oscillations to a safe level and allow sufficient time for locating the exact source. The proposed source location algorithm is validated using measurements collected during the January 11, 2019 forced oscillation event in North America and other actual grid events, while the proposed control strategy is verified using the Eastern Interconnection dynamic model under the replicated January 11, 2019 forced oscillation event. The simulation results demonstrated that the proposed source location algorithm can accurately identify the forced oscillation source, and the proposed control strategy can significantly reduce forced oscillation energy.