Generator tripping scheme (GTS) is the most commonly used scheme to prevent power systems from losing safety and stability. Usually, GTS is composed of offline predetermination and real-time scenario ...match. However, it is extremely time-consuming and labor-intensive for manual predetermination for a large-scale modern power system. To improve efficiency of predetermination, this paper proposes a framework of knowledge fusion-based deep reinforcement learning (KF-DRL) for intelligent predetermination of GTS. First, the Markov Decision Process (MDP) for GTS problem is formulated based on transient instability events. Then, linear action space is developed to reduce dimensionality of action space for multiple controllable generators. Especially, KF-DRL leverages domain knowledge about GTS to mask invalid actions during the decision-making process. This can enhance the efficiency and learning process. Moreover, the graph convolutional network (GCN) is introduced to the policy network for enhanced learning ability. Numerical simulation results obtained on New England power system demonstrate superiority of the proposed KF-DRL framework for GTS over the purely data-driven DRL method.
This study proposes a dual-loop supplementary frequency control (SFC) scheme for back-to-back voltage source converter high-voltage DC (BTB-VSC-HVDC) interconnecting asynchronous AC grids, to provide ...frequency support for each other after a large disturbance. The proposed dual-loop SFC consists of the frequency-active power (f–P) loop and the frequency-reactive power (f–Q) loop. The former deployed on the P-loop of VSC-HVDC can provide frequency support for the disturbed grid during the primary frequency regulation and improve the steady-state characteristics of the frequency response, while the latter attached to the Q-loop of VSC-HVDC supports virtual inertia to ameliorate the transient characteristics of the frequency response. Simulation studies are conducted based on the equivalently simplified model of Southwest Power Grid and Hubei Power Grid interconnected by Chongqing-Hubei BTB-VSC-HVDC. Simulation results show that the proposed SFC scheme can effectively improve the transient and steady-state characteristics of the system frequency response under a wide range of operating condition.
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.
This paper presents an adaptive damping controller for the pulse width modulation series compensator (PWMSC) to suppress the low frequency oscillation. The adaptive damping controller is designed by ...using goal representation heuristic dynamic programming (GrHDP), which can facilitate the mapping relationship between input and output signals and improve the control characteristics significantly compared with the conventional heuristic dynamic programming (HDP). Moreover, the adaptive damping controller does not need the system model and has strong learning ability. Simulations results of IEEE 16-machine 68-bus power system show that the proposed GrHDP damping controller can suppress oscillation quickly and has better damping performance than the conventional lead-lag damping controller and HDP based damping controller under different operating conditions and disturbances.
Different communication networks are used in the different application environment of the smart grid. So, how to effectively select the communication networks with the optimal comprehensive ...performance is an important issue for the power management corporations. A novel comprehensive performance evaluation based on Exponential Scale Analytic Hierarchy Process (ESAHP) and Grey Analytic Hierarchy Process (GAHP) for the communication networks selection is proposed in the electricity information acquisition system. The ESAHP is used to calculate the weight of each communication performance index and the economic performance index, and the GAHP is adopted to evaluate the economic cost of the different communication modes. The optimal comprehensive communication model can be selected by comprehensively comparing the communication performance and the economic cost. The test results show that the assessment model can effectively evaluate the comprehensive performance of the different communication networks selection in the electricity information acquisition system of the smart grid.
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.
Display omitted
•A data-driven wide-area damping controller (D-WADC) of BTB-VSC-HVDC is proposed.•The D-WADC makes use of the active and reactive power control of BTB-VSC-HVDC.•The D-WADC can adapt ...to the change of system operating condition.•The D-WADC can adaptively compensate the stochastic time delay.•Both simulation and hardware-in-loop experiment results verify the control effectiveness.
This paper proposes a data-driven adaptive wide-area damping controller (D-WADC) for back-to-back VSC-HVDC to suppress the low frequency oscillation in a large-scale interconnected power system. The proposed D-WADC adopts a dual-loop control structure to make full use of the active and reactive power control of VSC-HVDC to improve the damping of the power system. A data-driven algorithm named the goal representation heuristic dynamic programming is employed to design the proposed D-WADC, which means the design procedure only requires the input and output data rather than the mathematic model of the concerned power system. Thus, the D-WADC can adapt to the change of operating condition through online weight modification. Besides, the adaptive delay compensator (ADC) is added to effectively compensate the stochastic delay involved in the wide-area feedback signal. Case studies are conducted based on the simplified model of a practical power system and the 16-machine system with a back-to-back VSC-HVDC. Both the simulation and hardware-in-loop experiment results verify that the proposed D-WADC can effectively suppress the low-frequency oscillation under a wide range of operating conditions, disturbances, and stochastic communication delays.
Event-based load shedding (ELS) is an important emergency countermeasure against transient voltage instability in power systems. At present, the formulation of ELS measures is usually determined ...offline by the experience of experts, which is inefficient, time-consuming, and labor-intensive. This paper proposes a knowledge-enhanced deep reinforcement learning (DRL) method for intelligent ELS. Firstly, the Markov decision process (MDP) of the knowledge-enhanced DRL model for ELS is established based on transient stability simulation. Different from traditional response-based MDP, this MDP is event-based. Then, compared to conventional exponential decision space, a linear decision space of the DRL agent is established to reduce the decision space and training difficulty. Furthermore, the knowledge of removing repeated and negative actions is fused into DRL to improve training efficiency and decision quality. Finally, the simulation results of the CEPRI 36-bus system show that the proposed method can accurately give effective ELS measures. Compared with the pure data-driven DRL method, the knowledge-enhanced DRL method is more efficient.
Display omitted
•An event-based Markov decision process (MDP) via trial and error sequence is proposed.•A linear decision space for the MDP of event-based load shedding (ELS) is proposed.•A knowledge-enhanced MDP of removing repeated and negative actions is proposed.•The proposed event-based MDP can efficiently obtain effective ELS measures.•The proposed linear decision space can decrease the training difficulty.
The information and communication technology enhances the performance and efficiency of cyber-physical power systems (CPPSs). However, it makes the topology of CPPSs more exposed to malicious cyber ...attacks in the meantime. This article proposes a double deep-Q-network (DDQN)-based resilience assessment method for power systems under sequential attacks. The DDQN agent is devoted to identifying the least sequential attacks to the ultimate collapse of the power system under different operating conditions. A cascading failure simulator considering the characteristics of generators is developed to avoid a relatively optimistic assessment result. In addition, a novel resilience index is proposed to reflect the capability of the power system to deliver power under sequential attacks. Then, an improved prioritized experience replay technique is developed to accelerate the convergence rate of the training process for DDQN agent. Simulation results on the IEEE 39-bus, 118-bus, and 300-bus power systems demonstrate the effectiveness of the proposed DDQN-based resilience assessment method.
To improve the reliability of communication service in smart distribution grid (SDG), an access selection algorithm based on dynamic network status and different service types for heterogeneous ...wireless networks was proposed. The network performance index values were obtained in real time by multimode terminal and the variation trend of index values was analyzed by the growth matrix. The index weights were calculated by entropy-weight and then modified by rough set to get the final weights. Combining the grey relational analysis to sort the candidate networks, and the optimum communication network is selected. Simulation results show that the proposed algorithm can implement dynamically access selection in heterogeneous wireless networks of SDG effectively and reduce the network blocking probability.