Display omitted
•CNN is designed for both transient stability and instability mode prediction.•Stochastic gradient descent with warm restarts training algorithm is employed.•Case studies are ...conducted on both benchmark and practical larger power system.•Superior accuracy and robustness to input signal noise and loss.
Online transient stability assessment (TSA) is vital for power system control as it provides the basis for operators to decide emergency control actions. But none of previous TSA research has taken into consideration the difference between two instability modes (aperiodic instability and oscillatory instability), which may threaten secure operation of power system. To address this problem, a TSA and instability mode prediction method based on convolutional neural network is proposed. The method takes the bus voltage phasor sampled by phasor measurement units (PMUs) during a short observation window after disturbance as input, and outputs the prediction result promptly: stable, aperiodic unstable or oscillatory unstable. The end-to-end model automatically extracts needed features from the raw measurement data, thus freeing itself from reliance on expertise. At the offline training stage, stochastic gradient descent with warm restart (SGDR) optimization algorithm is employed so that the model tends to converge to 'flat' and 'wide' minima with better generalization ability. Case studies conducted on New England 39-bus system and Western Electricity Coordinating Council (WECC) 179-bus system demonstrate superior accuracy, adaptability and scalability of the proposed method compared with conventional machine learning methods. Furthermore, the proposed model is empirically proven to be robust to PMU noise and loss.
Although the harmonics interaction plays a vital role in causing commutation failure (CF) in an ultra HVDC (UHVDC) system with hierarchical connection, it has not been sufficiently considered in ...previous work. In this paper, the electric interaction among the inverter valves of the UHVDC system with hierarchical connection during receiving system fault transient process is demonstrated. Moreover, the mechanism of CFs in UHVDC system with hierarchical connection is also investigated via the analysis of harmonics interaction. After the AC fault occurs, the appearance processes of the local commutation failure (LCF) and concurrent commutation failure (CCF) are divided into several stages and interpreted by schematic diagrams. And the characteristics of each stage are also concluded. In addition, a coordinated CF prevention scheme based on harmonics and DC current detection is proposed to mitigate the CFs in UHVDC system with hierarchical connection. Finally, the effectiveness of the proposed scheme is verified via a <inline-formula><tex-math notation="LaTeX">\pm 1100</tex-math></inline-formula> kV UHVDC transmission system under various short circuit ratio values and AC-side coupling strengths.
This paper proposes an adaptive dual droop control (ADDC) scheme, it can provide the disturbed onshore system with fast frequency support from both other undisturbed onshore systems, and voltage ...source converter-based multi-terminal direct current (VSC-MTDC) integrated offshore wind farms (OWFs). With conventional droop control at onshore converters, the DC voltage and power flow will change once frequency events occur, it will lead to frequency variations in other undisturbed systems. The proposed ADDC scheme firstly detects the disturbed and undisturbed systems, and then makes the undisturbed onshore system provide more frequency support power, while ensuring safe operation by settling the support power limitation and regulating the droop coefficients. Moreover, the offshore stations will estimate the onshore DC voltage as the control signal for fast frequency support. After that, the OWFs will recover their rotor speed with an asymptotic control scheme to reduce the second frequency drop. Case studies are carried out on 3-terminal and 5-terminal test systems, and the Opal-RT real-time simulation platform, respectively. Different control schemes are compared, and the parameter uncertainty and noise disturbance are considered to illustrate the performance and effectiveness of the proposed ADDC scheme.
Simulation analysis is critical for identifying possible hazards and ensuring secure operation of power systems. In practice, large-disturbance rotor angle stability and voltage stability are two ...frequently intertwined stability problems. Accurately identifying the dominant instability mode (DIM) between them is important for directing power system emergency control action formulation. However, DIM identification has always relied on human expertise. This article proposes an intelligent DIM identification framework that can discriminate among stable status, rotor angle instability, and voltage instability based on active deep learning (ADL). To reduce human expert efforts required to label the DIM dataset when building DL models, a two-stage batch-mode integrated ADL query strategy (preselection and clustering) is designed for the framework. It samples only the most helpful samples to label in each iteration and considers both information contents and diversity in them to improve query efficiency, significantly reducing the required number of labeled samples. Case studies conducted on a benchmark power system (China Electric Power Research Institute (CEPRI) 36-bus system) and a practical large-area power system (Northeast China Power System) reveal that the proposed approach outperforms conventional methods in terms of accuracy, label efficiency, scalability, and adaptability to operational variability.
Deep learning (DL) is a useful tool for power system stability assessment (PSSA) and dominant instability mode (DIM) identification. However, when faced with operational variability, the performance ...of DL models degrades. This paper proposes a bidirectional active transfer learning (Bi-ATL) framework for more adaptive PSSA and DIM identification, where the DL model is easier to adapt to unlearned operating conditions with fewer newly labeled instances. At the instance level, forward active learning and backward active learning are integrated to progressively build a mixed instance set by actively including the most label-worthy instances of new operating conditions and actively eliminating the most useless original operating condition instances. Then at the model parameter level, the mixed instance set is utilized to fine-tune the original DL model to new operating conditions. The Bi-ATL framework synthesizes three-way information of the instances and model of the original operating condition, and a few labeled instances of new operating conditions for more efficient adaptation. Intensive case studies conducted on a benchmark power system (CEPRI 36-bus system) and a real-world large-scale power system (Northeast China Power System-2131 bus) validate the efficacy and efficiency of the Bi-ATL framework as well as the role of the three-way information.
Digital simulation is very important for the safe and economic operation of power systems. Power system dominant instability mode (DIM) identification is an intractable problem in large-scale ...transient simulation analysis, because transient instability and short-term voltage instability are often intertwined. Deep learning is a promising way to achieve accurate DIM identification. By encoding time series as images, this paper proposes a robust and transferable DIM identification framework combining curve filtering rules and a dual-channel VGGNet (DCVGG) image recognition network. The curve filtering rules can effectively reduce the redundancy of massive time-domain features, the model parameters, and training time. A skillfully designed method is proposed to encode time series as images and improve the feature extraction capability of convolutional neural networks. The proposed DCVGG network is capable of processing rotor angle and voltage images simultaneously and has the potential to become a pre-trained model, making transfer learning between different power systems possible. Case studies conducted on 8-machine 36-bus system and Northeast China Power Grid demonstrate the proposed framework has better performance, scalability, robustness, and transferable ability than contrastive machine learning methods.
With the expansion of power system simulation scale, intelligent stability analysis based on simulation data becomes more and more important. However, the changeable operating conditions and fault ...conditions will seriously affect the performance of the intelligent analysis model. This paper proposes an advanced unsupervised transfer learning (UTL) method combining dynamic graph attention network (DGAT) and data augmentation to identify the instability mode towards multiple known and unknown scenarios. For the pre-training stage, the DGAT model with data augmentation is used for instability mode identification based on the basic dataset. For the transfer learning stage, it starts when faced with new operating conditions or fault conditions different from the basic dataset. The UTL scheme takes the distribution differences among different categories of source domain and target domain as the loss to fine-tune the pre-trained model. Moreover, considering the difficulty of sample-labeling, the UTL scheme can make the pre-trained model adapt to them without additional labeled samples. Case studies are conducted on 8-machine 36-bus system and Northeast China Power Grid. The results verify the superiority of the DGAT model with data augmentation and confirm the UTL scheme can help the model adapt to multiple unknown scenarios.
Digital simulation is significant for the operating mode and control decision-making of power systems. In the process of simulation data analysis, stability analysis is an essential part. One of the ...most challenging tasks is to distinguish between transient rotor angle instability and short-term voltage instability. This paper proposes a graph attention networks (GATs)-based method to overcome this ticklish problem via integrating power grid topology information into the neural networks. Compared with the conventional graph convolutional networks (GCNs), the attention mechanism is introduced into the GATs to learn the weights among different neighbor vertices in the graph. Due to the difficulty of distinguishing between the rotor angle instability and voltage instability in some samples, a label-smoothing method is adopted to alleviate the influence caused by label inaccuracy. Case studies are conducted on an 8-machine 36-bus system and Northeast China Power System. Simulation results show that the proposed method has better performance than conventional GCNs and other machine learning methods.
Display omitted
•A deep learning-based large-scale power grid simulation analysis model is provided.•A graph attention network (GAT) integrating power grid topology is proposed.•The proposed label-smoothing loss function can tolerate label inaccuracy.•The proposed GAT model can efficiently identify the dominant instability mode.
Deep learning (DL) is effective in identifying the dominant instability mode (DIM) of power systems. However, regular supervised learning for training DL models requires a large number of training ...samples with accurate labels provided by power system experts, which is prohibitively costly in practice. To address this issue, this paper proposes a weakly supervised learning framework to train DL models for DIM identification based on cheap but potentially inaccurate (noisy) labels from non-expert engineers. The framework comprises two stages to mitigate the detrimental effects of label noise. At Stage I, an auxiliary model is proposed to intelligently detect detrimental noisy samples while preserving truly-labeled informative hard samples based on the entire training loss dynamics of base DL models. Then Stage II incorporates virtual adversarial training to utilize all samples, including noisy ones with labels removed, to train a smooth DIM identification model in a semi-supervised learning way. It can help further mitigate the effects of undetected label noise. Case studies are conducted on CEPRI 36-bus system and Northeast China Power System (2131 buses). The results verify that the proposed framework can tolerate high-intensity feature-(in)dependent label noise and build reliable DL models for DIM identification with significantly less reliance on experts.
•Robust deep learning for power system dominant instability mode identification.•Two-stage weakly supervised learning to mitigate label noise from non-experts.•Noisy samples are detected while hard ones are preserved.•Virtual adversarial training is proposed to use noisy samples as unlabeled ones.•Intelligent dominant instability mode identification becomes more affordable.
Display omitted
With the increasing penetration of renewable energy, power system instability factors are rising. Transient voltage instability is a common power system problem that can cause blackouts and severe ...economic losses. An important measure to ensure transient voltage stability during emergencies is event-based load shedding (ELS). However, formulating ELS measures by experts gradually becomes inadaptable and time-consuming currently. With the increased complexity and uncertainty of modern new power systems, there is an urgent need for more intelligent and rapid ELS. This paper proposes a knowledge-enhanced parallel branching dueling Q-network (BDQ) framework for intelligent and rapid ELS against transient voltage instability. Firstly, an event-based Markov decision process (MDP) that differs from the conventional response-based MDP is established, which can effectively guide the training process. Secondly, to condense the huge conventional exponential decision space, a multi-branch BDQ structure is designed, which has higher training effectiveness and decision capability compared to branchless agents. Then, the domain proprietary knowledge that low-voltage buses are prioritized for ELS is incorporated into the BDQ agent. In comparison with a purely data-driven BDQ approach, incorporating knowledge can significantly enhance both training effectiveness and decision capability. Next, to further improve the applicability in large-scale real power systems, the parallel BDQ is proposed. Finally, the advantages of the proposed approach are demonstrated in the China Electric Power Research Institute 36-bus system and the Western Electricity Coordinating Council 179-bus system.
Display omitted
•Knowledge-enhanced parallel BDQ framework is proposed for transient voltage instability.•Multi-branch BDQ structure is designed to handle the huge exponential decision space.•Knowledge of prioritizing ELS on low-voltage buses can enhance decision capability.•Parallel BDQ training is proposed to improve application in actual large-scale power system.