Search

Transient stability (TS) and short-term voltage stability (STVS) assessment are of fundamental importance for the operation security of power systems. Both phenomena can be mutually influenced in weak power systems due to the proliferation of power electronic interface devices and the phase-out of conventional heavy machines (e.g., thermal power plants). There is little research on the assessment of both types of stability together, despite the fact that they develop over the same short-term period, and that they can have a major influence on the overall transient performance driven by large electrical disturbances (e.g., short circuits). This work addresses this open research challenge by proposing a methodology for the joint assessment of TS and STVS. The methodology aims at estimating the resulting short-term stability state (STSS) in stable, or unstable conditions, following critical events, such as the synchronism loss of synchronous generators (SG) or the stalling of induction motors (IM). The estimations capture the mechanisms responsible for the degradations of TS and STVS, respectively. The paper overviews the off-line design of the data-driven STSS classification methodology, which supports the design and training of a hybrid deep neural network RCNN (recurrent convolutional neural network). The RCNN can automatically capture spatial and temporal features from the power system through a time series of selected physical variables, which results in a high estimation degree for STSS in real-time applications. The methodology is tested on the New England 39-bus system, where the results demonstrate the superiority of the proposed methodology over other traditional and deep learning-based methodologies. For reference purposes, the numerical tests also illustrate the classification performance in special situations, when the training is performed by exclusively using measurements from generation and motor load buses, which constitute locations where the investigated stability can be observed. [ABSTRACT FROM AUTHOR]

Short-term load forecasting (STLF) has a significant role in reliable operation and efficient scheduling of power systems. However, it is still a major challenge to accurately predict power load due to social and natural factors, such as temperature, humidity, holidays and weekends, etc. Therefore, it is very important for the efficient feature selection and extraction of input data to improve the accuracy of STLF. In this paper, a novel hybrid model based on empirical mode decomposition (EMD), a one-dimensional convolutional neural network (1D-CNN), a temporal convolutional network (TCN), a self-attention mechanism (SAM), and a long short-term memory network (LSTM) is proposed to fully decompose the input data and mine the in-depth features to improve the accuracy of load forecasting. Firstly, the original load sequence was decomposed into a number of sub-series by the EMD, and the Pearson correlation coefficient method (PCC) was applied for analyzing the correlation between the sub-series with the original load data. Secondly, to achieve the relationships between load series and external factors during an hour scale and the correlations among these data points, a strategy based on the 1D-CNN and TCN is proposed to comprehensively refine the feature extraction. The SAM was introduced to further enhance the key feature information. Finally, the feature matrix was fed into the long short-term memory (LSTM) for STLF. According to experimental results employing the North American New England Control Area (ISO-NE-CA) dataset, the proposed model is more accurate than 1D-CNN, LSTM, TCN, 1D-CNN–LSTM, and TCN–LSTM models. The proposed model outperforms the 1D-CNN, LSTM, TCN, 1D-CNN–LSTM, and TCN–LSTM by 21.88%, 51.62%, 36.44%, 42.75%, 16.67% and 40.48%, respectively, in terms of the mean absolute percentage error. [ABSTRACT FROM AUTHOR]

This paper proposes a data-driven cooperative method for load frequency control (LFC) of the multi-area power system based on multi-agent deep reinforcement learning (MA-DRL) in continuous action domain. The proposed method can nonlinearly and adaptively derive the optimal coordinated control strategies for multiple LFC controllers through centralized learning and decentralized implementation. The centralized learning is achieved by MA-DRL based on a global action-value function to quantify overall LFC performance of the power system. To solve the MA-DRL problem, multi-agent deep deterministic policy gradient (DDPG) is derived to adjust control agents’ parameters considering the nonlinear generator behaviors. For implementation, each individual controller only needs local information in its control area to deliver optimal control signals. Numerical simulations on a three-area power system and the fully-modeled New-England 39-bus system demonstrate that the proposed method can effectively minimize control errors against stochastic frequency variations caused by load and renewable power fluctuations. [ABSTRACT FROM AUTHOR]

This paper addresses the problem of frequency stability prediction (FSP) following active power disturbances in power systems by proposing a vision transformer (ViT) method that predicts frequency stability in real time. The core idea of the FSP approach employing the ViT is to use the time-series data of power system operations as ViT inputs to perform FSP accurately and quickly so that operators can decide frequency control actions, minimizing the losses caused by incidents. Additionally, due to the high-dimensional and redundant input data of the power system and the O(N2) computational complexity of the transformer, feature selection based on copula entropy (CE) is used to construct image-like data with fixed dimensions from power system operation data and remove redundant information. Moreover, no previous FSP study has taken safety margins into consideration, which may threaten the secure operation of power systems. Therefore, a frequency security index (FSI) is used to form the sample labels, which are categorized as "insecurity", "relative security", and "absolute security". Finally, various case studies are carried out on a modified New England 39-bus system and a modified ACTIVSg500 system for projected 0% to 40% nonsynchronous system penetration levels. The simulation results demonstrate that the proposed method achieves state-of-the-art (SOTA) performance on normal, noisy, and incomplete datasets in comparison with eight machine-learning methods. [ABSTRACT FROM AUTHOR]

Electric Vehicles (EVs) and their charging infrastructure have witnessed rapid adoption into the smart grid owing to their ability to reduce the transportation sector's emissions. However, this hasty deployment has left the grid prone to attacks initiated through the charging ecosystem. This paper develops a family of robust dynamic load-altering attacks that manipulate the EV load to destabilize the power grid. The attack strategy is formulated based on feedback control theory and Linear Matrix Inequalities (LMIs) and incorporates mathematical modeling of the power grid uncertainties. The effectiveness of our proposed robust EV attack strategy is demonstrated through extensive simulations of attack scenarios against the New England (NE) 39-bus grid. The results demonstrate that the proposed attacks were able to destabilize the grid frequency inducing a deviation of 1.5 Hz and forcing the generators to trip after compromising as little as 6.6% of the total grid load. We also present a two-tier Deep Learning (DL) attack detection mechanism. We optimize this mechanism to achieve optimal results against dynamic and switching attacks. The proposed mechanism was evaluated against several other models and proved to be superior with an accuracy of 99.9%. • We propose a robust EV Dynamic Attack Design that takes advantage of compromised EVs to cause power grid instability. • The controller design accounts for several uncertainties that can impact the attack success. • We perform extensive time domain simulations to demonstrate the devastating impact of such attacks even in the presence of uncertainties. • The proposed scheme demonstrates its ability to cause generator tripping even with just 6.7% of the total grid load. • We then propose a two-tier detection mechanism that achieved a detection accuracy above 99.9%. [ABSTRACT FROM AUTHOR]

• The vulnerable branches which affect the transient voltage security are identified. • Image-form samples combine grid structure and power flow information visually. • Knowledge visualization makes CNN-based transient voltage stability assessment model more understandable and reliable. • A comprehensive vulnerability indicator integrates the knowledge in CNN, grid topology and branch state. • The proposed method is demonstrated on a real-life provincial grid with renewable energy integration. Because of the increasing proportion of grid-connected converters, the transient voltage problem has increased the risk of the secure and stable operation of power systems. As the extension of transient voltage stability assessment (TVSA), transient voltage vulnerability assessment is the key technique to detect the crucial elements which can cause the insecurity risk of power system transient voltage. Although deep learning-based TVSA models have demonstrated the high accuracy, it is more important to display the knowledge learnt by the models and utilize it to recognize the weakness of transient voltage. This paper presents a transient voltage vulnerability assessment method by knowledge visualization of convolutional neural network (CNN). First, the image-form sample construction method is proposed to characterize the power system operating condition, which prepares for training CNN-based TVSA model. Then, the knowledge in TVSA model is visualized based on Gradient-weighted Class Activation Mapping (Grad-CAM) and Activation Maximization (AM) algorithms, which is analyzed from the perspective of power system stability. Finally, a comprehensive vulnerability indicator is proposed for identify the weakness of transient voltage, which combines the knowledge inside the TVSA model and the indicators with clear physical meaning. Simulation results of the New England 39-bus test system and the real Shandong provincial power grid in China demonstrate that the visualization makes the model more reliable and the proposed method can accurately identify the vulnerable branches. [ABSTRACT FROM AUTHOR]

To solve the problem of manpower and time consumption caused by power flow state adjustment in a large-scale power grid, a power system operation state adjustment method considering the static stability constraint based on parallel deep reinforcement learning is proposed. By introducing the process of adjusting the power flow state that satisfies static stability, the Markov decision-making process of adjusting power flow is constructed. Then, based on the positioning of the adjustment target, the selection of actionable devices and the calculation of the amount of action, a power flow state adjustment strategy is developed. The adjustment process is accelerated through sensitivity, transfer ratio and load margin. Then, a parallel deep reinforcement learning model is established, and it maps actions to power flow adjustment to form a pair of generator actions and realises parallel adjustment of multi-sectional objectives. In addition, the reinforcement learning strategy and the deep learning network are improved to promote learning efficiency. Finally, the New England 39-bus standard system and actual power grid are used to verify the effectiveness of the method. [ABSTRACT FROM AUTHOR]

• A hierarchical framework to accurately detect, classify, and locate the FDEs. • All FDEs (Generator Trip, Line Outage, Load Disconnection, and fault) are evaluated. • A limited number of PMUs are applied to collect few samples of the ROCOF time series. • High-precision RNN and LSTM models are employed online to classify and locate FDEs. Frequency Disturbance Events (FDEs) occur due to various events such as Generator Trip (GT), Line Outage (LO), and Load Disconnection (LD), which affect the stability of power systems. Depending on disturbance severity, it has small to severe effects on an integrated system's performance. Accurate and fast detection and classification of the event and its location are crucial in monitoring resource adequacy, preventing economic losses and blackouts. This paper, focusing on an online cutting-edge hierarchical methodology, first detects the event, then identifies its classification. Eventually, the exact location of FDE is found based on Phasor Measurement Units (PMUs) data. Compared to other counterparts, which focused only on classifying or locating events separately, in this paper, both aims are achieved using the suggested novel hierarchical framework. This research employs Deep Learning (DL) advances to develop a Recurrent Neural Network (RNN) model and a Long Short-Term Memory (LSTM) model to distinguish and locate FDEs with significant accuracy. In the research, only a few time-series of Rate-Of-Change-Of-Frequency (ROCOF) received from a limited number of PMUs is used as the DL algorithm's inputs. This hierarchical methodology is tested in New England 39-bus, IEEE 14 bus systems, and modified IEEE 118-bus system. The evaluation results demonstrate the potential application of the proposed models to detect and classify FDEs compared to conventional algorithms and the frequency-based DL model. The proposed models have achieved significant classification accuracy. Simultaneously, the models locate the events with competitive performance in terms of precision, recall, accuracy, and F1-score. © 2017 Elsevier Inc. All rights reserved. [ABSTRACT FROM AUTHOR]

Short-Term Load Forecasting is critical for reliable power system operation, and the search for enhanced methodologies has been a constant field of investigation, particularly in an increasingly competitive environment where the market operator and its participants need to better inform their decisions. Hence, it is important to continue advancing in terms of forecasting accuracy and consistency. This paper presents a new deep learning-based ensemble methodology for 24 h ahead load forecasting, where an automatic framework is proposed to select the best Box-Jenkins models (ARIMA Forecasters), from a wide-range of combinations. The method is distinct in its parameters but more importantly in considering different batches of historical (training) data, thus benefiting from prediction models focused on recent and longer load trends. Afterwards, these accurate predictions, mainly the linear components of the load time-series, are fed to the ensemble Deep Forward Neural Network. This flexible type of network architecture not only functions as a combiner but also receives additional historical and auxiliary data to further its generalization capabilities. Numerical testing using New England market data validated the proposed ensemble approach with diverse base forecasters, achieving promising results in comparison with other state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Increased integration of renewable energy sources brings new challenges to the secure and stable power system operation. Operational challenges emanating from the reduced system inertia, in particular, will have important repercussions on the power system transient stability assessment (TSA). At the same time, a rise of the "big data" in the power system, from the development of wide area monitoring systems, introduces new paradigms for dealing with these challenges. Transient stability concerns are drawing attention of various stakeholders as they can be the leading causes of major outages. The aim of this paper is to address the power system TSA problem from the perspective of data mining and machine learning (ML). A novel 3.8 GB open dataset of time-domain phasor measurements signals is built from dynamic simulations of the IEEE New England 39-bus test case power system. A data processing pipeline is developed for features engineering and statistical post-processing. A complete ML model is proposed for the TSA analysis, built from a denoising stacked autoencoder and a voting ensemble classifier. Ensemble consist of pooling predictions from a support vector machine and a random forest. Results from the classifier application on the test case power system are reported and discussed. The ML application to the TSA problem is promising, since it is able to ingest huge amounts of data while retaining the ability to generalize and support real-time decisions. [ABSTRACT FROM AUTHOR]

This paper proposes an effective deep learning framework for Short-Term Load Forecasting (STLF) of multivariate time series. The proposed model consists of a hybrid Convolutional neural network-Bidirectional Long Short-Term Memory (CBiLSTM) based on the Evolution Strategy (ES) method and the Savitzky–Golay (SG) filter (SG-CBiLSTM). The adopted methodology incorporates the virtue of different prepossessing blocks to enhance the performance of the CBiLSTM model. In particular, a data-augmentation strategy is employed to synthetically improve the feature representation of the CBiLSTM model. The augmented data is forwarded to the Partial Least Square (PLS) method to select the most informative features above the predefined threshold. Next, the SG algorithm is computed for smoothing the load to enhance the learning capabilities of the underlying system. The structure of the SG-CBiLSTM for the ISO New England dataset is optimized using the ES technique. Finally, the CBiLSTM model generates output forecasts. The proposed approach demonstrates a remarkable improvement in the performance of the original CBiLSTM model. Furthermore, the experimental results strongly confirm the high effectiveness of the proposed SG-CBiLSTM model compared to the state-of-the-art techniques. [ABSTRACT FROM AUTHOR]

As the transitions of the power industry to decarburization and distributed energy systems, the future uncertainty information of electric load is becoming essential in power systems planning and operation. However, a great number of studies focus on point forecasting, which only provides the expected value at each time step and it cannot provide uncertainty information. This paper proposed a novel probabilistic load forecasting model by combining quantile regression (QR) with a hybrid model to improve smart grid reliability. In addition, to further improve accuracy and solve the problem that the optimal model is not unique, we propose a new combined probabilistic forecasting model (CPFM). The CPFM employs the traditional statistical models and QR-machine learning models as alternative models; several alternative models with the best performance are combined through the improved multi-objective optimizer to obtain the final forecasting results. The ISO New England data is modeled as a case study to verify the effectiveness of the proposed CPFM. The comparative study includes 13 models, and the results show that the proposed CPFM has better performance in reliability, resolution, and sharpness. • Developed a deep QR hybrid model that can perform accurate probabilistic forecast. • A combined probabilistic forecasting model was proposed. • The problem that the optimal model is not unique in interval forecast is solved. • Improved the rat swarm algorithm with three strategies. [ABSTRACT FROM AUTHOR]

This paper uses deep (machine) learning techniques to develop and test how motor behaviors, derived from location and movement sensor tracking data, may be associated with falls, delirium, and urinary tract infections (UTIs) in long-term care (LTC) residents. Longitudinal observational study. A total of 23 LTC residents (81,323 observations) with cognitive impairment or dementia in 2 northeast Department of Veterans Affairs LTC facilities. More than 18 months of continuous (24/7) monitoring of motor behavior and activity levels used objective radiofrequency identification sensor data to track and record movement data. Occurrence of acute events was recorded each week. Unsupervised deep learning models were used to classify motor behaviors into 5 clusters; supervised decision tree algorithms used these clusters to predict acute health events (falls, delirium, and UTIs) the week before the week of the event. Motor behaviors were classified into 5 categories (Silhouette score = 0.67), and these were significantly different from each other. Motor behavior classifications were sensitive and specific to falls, delirium, and UTI predictions 1 week before the week of the event (sensitivity range = 0.88–0.91; specificity range = 0.71–0.88). Intraindividual changes in motor behaviors predict some of the most common and detrimental acute events in LTC populations. Study findings suggest real-time locating system sensor data and machine learning techniques may be used in clinical applications to effectively prevent falls and lead to the earlier recognition of risk for delirium and UTIs in this vulnerable population. [ABSTRACT FROM AUTHOR]

• Model aggregated synthetic inertia and PFR support from WPPs and PVPs. • Represent aggregated synthetic inertia and PFR in stochastic SCUC framework. • Estimate the PFR requirement based on system inertia condition. • Impact of frequency security parameter variation on system performance. • Uncertainty characterization of RE power via Bayesian statistics using MCMC model. Increasing grid integration of uncertain Renewable Energy (RE) sources reduces system's inherent Frequency Response (FR) and restricts System Operators'(SOs) ability to maintain frequency stability. This condition further aggravates as existing operational practices do not consider FR from RE based generation. Technical feasibility of RE generation to facilitate synthetic inertia and Primary Frequency Response (PFR) is well established. However, FR support from RE plants must be integrated in scheduling decisions to ensure inertial and PFR adequacy in operational time frame. Further, unlike conventional units, RE plants' output is volatile due to inherent variability and uncertainty. Hence, the uncertainty of RE generation needs to be considered in scheduling horizon to accurately assess system-wide FR requirement and RE Plants' potential in its' contribution. In this context, the paper proposes a novel day-ahead stochastic security-constrained unit commitment framework to incorporate synthetic inertia and PFR support from wind and PV plants, and ensure frequency response adequacy. RE generation is forecasted using long-short-term memory, a deep learning approach. A machine learning-based Bayesian inference statistics is used to characterize RE power uncertainty. Simulations are carried out on New England test system for various RE integration levels. Numerical results demonstrate the effectiveness of frequency support from RE plants in terms of improvement in frequency security parameters. Operating cost and RE curtailment are reduced 11% and 68%, respectively with PFR support from RE plants. [ABSTRACT FROM AUTHOR]

• The hybrid information-driven DSA model is fast, lightweight and fault-independent. • One-hot encoding raises the learning efficiency of the DSA model for fault info. • Improved ADASYN alleviates the problem of data imbalance for data-driven DSA. Online dynamic security assessment is one of the important applications of online situation awareness for power systems, providing essential information for secure operation and preventive control. Different from the existing methods which are fault-dependent or requiring for transient-state data, this paper proposes a novel deep learning model for online power system multi-fault dynamic security assessment driven by hybrid information of anticipated faults and static operating points. The proposed model is universal and suitable for different fault situations. In addition, this model avoids the acquisition of transient information, thus reducing the time-consuming process of online evaluation. For the discrete nominal features in the hybrid information, one-hot encoding is introduced to preprocess the discrete nominal features. Both the performance of the dynamic security assessment model and the principal component analysis are promoted. Furthermore, an improved adaptive synthetic sampling algorithm is proposed and applied to alleviate the problem of power system data imbalance. Finally, the feasibility and effectiveness of the proposed method is verified by testing on the anticipated faults in the New England 39-bus system and the IEEE 54-machine 118-bus system. [ABSTRACT FROM AUTHOR]

Short-term load forecasting (STLF) is very important for an efficient operation of the power system because the exact and stable load forecasting brings good results to the power system. This manuscript presents the application of two new models in STLF i.e. Cross multi-models and second decision mechanism and Residential load forecasting in smart grid using deep neural network models. In the cross multi-model and second decision mechanism method, the horizontal and longitudinal load characteristics are useful for the construction of the model with the calculation of the total load. The dataset for this model is considered from Maine in New England, Singapore, and New South Wales of Australia. While, In the residential load forecasting method, the Spatio-temporal correlation technique is used for the construction of the iterative ResBlock and deep neural network which helps to give the characteristics of residential load with the use of a publicly available Redd dataset. The performances of the proposed models are calculated by the Root Mean Square Error, Mean Absolute Error, and Mean Absolute Percentage Error. From the simulation results, it is concluded that the performance of cross multi-model and second decision mechanism is good as compare to the residential load forecasting. [ABSTRACT FROM AUTHOR]

• Propose a deep-learning intelligent system incorporating data augmentation for STVS assessment. • Build a BiGRU-attention-based assessment model to extract temporal dependencies. • Effectiveness and superiority of the approach is verified by the New England 39-bus system. • Statistical tests have been performed to examine the performance of the proposal. Facing the difficulty of expensive and trivial data collection and annotation, how to make a deep learning-based short-term voltage stability assessment (STVSA) model work well on a small training dataset is a challenging and urgent problem. Although a big enough dataset can be directly generated by contingency simulation, this data generation process is usually cumbersome and inefficient; while data augmentation provides a low-cost and efficient way to artificially inflate the representative and diversified training datasets with label preserving transformations. In this respect, this paper proposes a novel deep-learning intelligent system incorporating data augmentation for STVSA of power systems. First, due to the unavailability of reliable quantitative criteria to judge the stability status for a specific power system, semi-supervised cluster learning is leveraged to obtain labeled samples in an original small dataset. Second, to make deep learning applicable to the small dataset, conditional least squares generative adversarial networks (LSGAN)-based data augmentation is introduced to expand the original dataset via artificially creating additional valid samples. Third, to extract temporal dependencies from the post-disturbance dynamic trajectories of a system, a bi-directional gated recurrent unit with attention mechanism based assessment model is established, which bi-directionally learns the significant time dependencies and automatically allocates attention weights. The test results demonstrate the presented approach manages to achieve better accuracy and a faster response time with original small datasets. Besides classification accuracy, this work employs statistical measures to comprehensively examine the performance of the proposal. [ABSTRACT FROM AUTHOR]

This paper presents a multi-agent Double Deep Q Network (DDQN) based on deep reinforcement learning for solving the transmission network expansion planning (TNEP) of a high-penetration renewable energy source (RES) system considering uncertainty. First, a K-means algorithm that enhances the extraction quality of variable wind and load power uncertain characteristics is proposed. Its clustering objective function considers the cumulation and change rate of operation data. Then, based on the typical scenarios, we build a bi-level TNEP model that includes comprehensive cost, electrical betweenness, wind curtailment and load shedding to evaluate the stability and economy of the network. Finally, we propose a multi-agent DDQN that predicts the construction value of each line through interaction with the TNEP model, and then optimizes the line construction sequence. This training mechanism is more traceable and interpretable than the heuristic-based methods. Simultaneously, the experience reuse characteristic of multi-agent DDQN can be implemented in multi-scenario TNEP tasks without repeated training. Simulation results obtained in the modified IEEE 24-bus system and New England 39-bus system verify the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

Electricity load forecasting plays a vital role in power system planning and operations. Probabilistic forecasting is expected to become a popular load prediction form due to providing more uncertainty information for the decision of smart grid. As one of the promising forecasting methods, the convolutional neural network has an outstanding advantage in feature extraction. However, there is a critical problem that needs to be solved when using a convolutional neural network for probabilistic load forecasting. The classical parametric and nonparametric techniques for generating probability distribution suffer from the predetermined load probability distribution types or the nondifferentiable training function, which might affect the prediction accuracy of the convolutional neural network. In this paper, a load range discretization method is proposed to generate load probability distributions. The method constructs discrete load probability distributions by segmenting the load range. Then, the optimal estimation is employed to optimize the load probability distributions for training samples. As a result, the samples can be utilized to train the convolutional neural network, so that the network can forecast load probability distributions directly. There is no probability distribution assumption and nondifferentiable training function in the proposed method. Based on the data of independent system operators in New England, the superiority of the proposed method is verified by comparing with 7 well-established benchmarks. The proposed method acquires more reliable and sharper load probability distributions, which can be beneficial to various decision-making activities in power systems. Image 1 • A novel probabilistic load forecasting model is proposed based on an improved convolutional neural network (CNN). • A universal load range discretization (LRD) method is proposed to construct the probabilistic training samples for CNN. • The optimal estimation is employed to search for the optimal load probability distributions for training samples. • Seven typical methods are adopted for comparing and analyzing the forecasting results. [ABSTRACT FROM AUTHOR]