Large-scale Power Grid Research Articles

Hollow bimetallic sulfide composite coated separator for hindering the shuttle of polysulfides in Li-S batteries

Due to the increasing demand of new energy vehicles and large-scale power grids for high energy density devices, lithium-sulfur batteries (LSBs) have attracted more and more researchers' attention stemming from their advantages of high theoretical specific capacity, low cost of active substances and environmental friendliness. However, there are still some intractable problems hindering the commercialization of LSBs, among which the most serious problem is “shuttle effect” of polysulfides. For the purpose of solving this problem, this paper prepared hollow structure molybdenum sulfide/cobalt sulfide mixed with amorphous carbon (MoS2/Co9S8/C) (HMCC) nanoparticles using ZIF-67 as the precursor and coated on the side of ordinary PE separator facing the positive electrode by a simple scraper coating process. The combination of MoS2 and Co9S8 bimetallic sulfides can not only adsorb and catalyze the rapid transformation of intermediates, but also effectively conduct lithium ions mainly because of the exposure of many active sites of layered MoS2 in the outermost layer. In addition, a small amount of carbon inside can enhance electrical conductivity of the coating, increasing the utilization of active substances, and the stable hollow structure can act as a "transfer station", greatly inhibiting the large amount of polysulfides to run off. The dual advantages of structure and composition make the modified separator have excellent performance in different current densities. Reversible specific capacities of 930.4 mAh g−1 and 690.7 mAh g−1 are released at 0.5 C and 1 C, and capacity retention of 72 % and 64 % after 500 and 600 cycles, respectively. In addition, due to the rapid conduction of lithium ions, the growth of lithium dendrites is greatly inhibited, which maintains the high safety of the battery. This provides a new research direction for the application of other new bimetallic sulfides in the separator modification of LSBs.

Developing a wind power forecasting system based on deep learning with attention mechanism

Large-scale wind power grid integration is challenging, requiring accurate and stable wind power forecasting to reduce operational risk and improve the scheduling efficiency of the grid. Previous studies have emphasized data preprocessing and parameter optimization. Studies on feature mining for the input layer, particularly the hidden layer, are limited, hindering further forecasting accuracy and stability improvements. To address this gap, a novel wind power forecasting system consisting of feature decomposition, self-attention, forecasting, optimization, and performance evaluation modules, is developed in this study. The feature decomposition module is used to remove superfluous noise components inherent in wind power data. A dual-stage self-attention mechanism is proposed in the self-attention module to determine the importance of driving features in the input and hidden layers, rather than focusing on all f/eatures equally. A deep learning model is used with the optimization module, built into the forecasting module. A performance evaluation module is constructed to provide a synthetical evaluation of the forecasting system. Numerical experiments are conducted to validate the effectiveness of the system and demonstrate its superiority.

Markov Chains for Fault-Tolerance Modeling of Stochastic Networks

Most real-world networks are time-varying, and many are subject to the stochastic functioning of their nodes and edges. Examples can be seen in the human brain undergoing an epileptic seizure, spontaneous infection and recovery in epidemics, and intermittent functioning of devices in the Internet of Things. Moreover, such networks are becoming increasingly large due to rapid technological advances. However, little has been done to study time-varying, large-scale, stochastic networks (SNs) from a reliability engineering perspective. Toward this goal, this article develops a fault-tolerance model for a type of time-varying network in which nodes (and/or edges) stochastically switch between active and inactive states. It considers fault tolerance from a global connectivity point of view, which has applications in many natural and engineered networks. Specifically, this article presents a Markov chain framework that models the dynamic behavior of nodes and allows for the computation of quantitative measures, including availability and time-to-failure metrics. To accommodate large-scale networks and emphasize global connectivity, this framework utilizes percolation theory, which has recently been of interest in the reliability engineering discipline, to characterize network failure. This article makes several contributions: it proposes a Markov chain framework for computing fault-tolerance metrics that is tractable for large-scale networks, it shows the existence of a phase transition in network availability of a time-varying SN, and it accounts for finite-size effects of percolation in the fault-tolerance model. The proposed methodology is applied to Erdös–Rényi random graphs and a real, large-scale power grid. Experimental results provide insights into network design, maintenance, and failure prevention of time-varying SNs. <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">Note to Practitioners</i> —This work develops a fault-tolerance model for time-varying stochastic networks in which nodes (and/or edges) randomly switch between active and inactive states. To address increasingly large-scale networks that are being studied, this article appeals to percolation theory. Fault tolerance is, thus, studied from a global connectivity perspective where the existence of a large connected component containing most of the nodes characterizes the functioning of the network. Specifically, this article presents a continuous-time Markov chain (CTMC) framework that models node dynamics and allows for the computation of fault-tolerance metrics, including network availability and mean time to failure. The proposed framework computes metrics efficiently for large networks and allows for studying their asymptotics. The percolation threshold describing the dissolution of the large connected component is used as the failure criterion in the CTMC. The practitioner should note that several assumptions are made in the proposed CTMC framework: nodes possess identical and time-invariant failure and recovery rates, node failure and recovery times are exponentially distributed, and node dynamics are independent of one another. In addition, fault-tolerance metrics computed for finite networks are estimators of the true metric values. The proposed framework is advantageous for quantifying the fault tolerance of large-scale, time-varying networks where the combinatorial explosion and a changing network topology pose challenges to the use of traditional reliability methods. A case study of a power grid network shows how to apply the proposed methodology to real networks.

Blind False Data Injection Attacks Against State Estimation Based on Matrix Reconstruction

Cyber, especially false data injection attacks (FDIAs), are gradually becoming a kind of severe threat to modern power grids. This paper proposes a blind FDIA approach against the state estimation of power grids based on matrix reconstruction and subspace estimation. As a blind FDIA method, it requires no information about the system parameters and topology, but only needs a limited period of measurement data. In particular, this paper investigates the interference of random measurement noise for subspace based blind FDIA methods, and designs a scheme to alleviate the influence of this random noise in the process of FDIA. Compared with previous blind FDIA methods, the proposed method overcomes the shortcoming that measurement noise is difficult to address when data is not abundant. Thus the presented approach performs high successful rate of FDIA, and it is able to operate when measurement data is very limited. Besides, it performs great robustness to large-scale power grids and high-level measurement noise. Numerous cases demonstrate the effectiveness and advantages of the proposed method.

Wind Speed Prediction via Collaborative Filtering on Virtual Edge Expanding Graphs

Accurate and stable wind speed prediction is crucial for the safe operation of large-scale wind power grid connections. Existing methods are typically limited to a certain fixed area when learning the information of the wind speed sequence, which cannot make full use of the spatiotemporal correlation of the wind speed sequence. To address this problem, in this paper we propose a new wind speed prediction method based on collaborative filtering against a virtual edge expansion graph structure in which virtual edges enrich the semantics that the graph can express. It is an effective extension of the dataset, connecting wind turbines of different wind farms through virtual edges to ensure that the spatial correlation of wind speed sequences can be effectively learned and utilized. The new collaborative filtering on the graph is reflected in the processing of the wind speed sequence. The wind speed is preprocessed from the perspective of pattern mining to effectively integrate various information, and the k-d tree is used to match the wind speed sequence to achieve the purpose of collaborative filtering. Finally, a model with long short-term memory (LSTM) as the main body is constructed for wind speed prediction. By taking the wind speed of the actual wind farm as the research object, we compare the new approach with four typical wind speed prediction methods. The mean square error is reduced by 16.40%, 11.78%, 9.57%, and 18.36%, respectively, which demonstrates the superiority of the proposed new method.

Parameter identification of small distributed hydro generators during excitation system start‐up

Concerning the penetration of renewable energy resources into the power grid, investigation of the power system dynamics to meet the power demand and maintain stability has gathered momentum recently. On the other hand, evaluating the generator parameters in the distributed generation (DG) units as a subsystem of a large-scale power grid is crucial for control and stability analysis purposes. To this aim, a simple and efficacious algorithm to identify the operational parameters of synchronous generators (SGs) in small hydro DGs is proposed here. The required data for the SG identification using the recommended approach is gathered during excitation system start-up while the DG is disconnected from the grid. Using the proposed test procedure, the adverse impact of magnetic saturation on estimation results can be avoided. Furthermore, a straightforward parameter estimation (PE) procedure is established here to obtain the intended parameters from the recorded data. Finally, the proposed method was applied to a 10 MW hydro DG, and the results are discussed.

Robust Transmission Expansion Planning of Ultrahigh-Voltage AC–DC Hybrid Grids

With the ever-increasing installation of large-capacity ultrahigh-voltage direct current (UHVdc) links, the “strong dc link and weak ac network” issue has become prominent for many regional grids in China. The transmission expansion planning (TEP) problem of such grids has been facing a significant challenge to cope with the coordination between the UHVdc links and the ultrahigh/high voltage alternating current (UHVac/HVac) network. To handle this issue, this article proposes a multistage robust TEP (RTEP) model of ultrahigh-voltage ac–dc hybrid grids. This model explicitly takes into account multiple types of uncertainties, i.e., credible contingencies occurring in the ac network, large power imbalances due to the UHVdc bipole blocking, as well as the long-/short-term uncertainty. Wasserstein metric is used to reformulate the uncertainty set and the worst-case constraints. A column-and-constraint generation based solution approach is developed in order to ensure tractability of the RTEP model. The effectiveness of the proposed RTEP is verified by numerical testing on a practical large-scale power grid.

Data‐model driven rescheduling considering both rotor angle stability and transient voltage stability constraints

Facing the problem of workforce and time costs caused by the rescheduling of large-scale power grids with renewable energy penetration, and the lack of parallel control research on rotor angle instability and transient voltage instability, a rescheduling method based on trajectory sensitivity and deep reinforcement learning (DRL) is proposed to meet the requirements of rotor angle stability and transient voltage stability at the same time. By introducing the process of rescheduling to meet the transient stability constraint, the Markov decision-making process of rescheduling is first constructed. Then, a simple dominant instability type identification method is proposed according to the occurrence time, location, voltage pattern, and position of the oscillation centre. Moreover, a rescheduling strategy is formulated based on the dominant instability mode identification, trajectory sensitivity calculation, actionable device selection, action amount computation, and action of the device. Next, according to the improved distributed distributive deep deterministic policy gradients (D4PG), a DRL model is established to map the action to actionable generator pairs and capacitors/reactors, so as to realize parallel rescheduling of rotor angle instability and transient voltage instability. Finally, the improved New England 39-bus system and an actual power grid are used to verify the effectiveness and advantages of the method.

A dynamic and integrated approach of safety investment decision-making for power grid enterprises

Recently, the consequences caused by frequent large-scale power grid accidents are more significant with arousing great attention of power grid enterprises. Scientific safety investments are essential to improve the overall safety level of the power grid enterprises and promote safe production. This study proposed a dynamic and integrated approach for safety investment decision-making of power grid enterprises based on the entropy weight method (EWM) and the system dynamics (SD) theory. By applying the accident loss assessment and the safety performance evaluation, the SD-based model for the safety investment decision-making for power grid enterprises was developed. The results show that risk assessment is the critical factor for power grid safety, followed by safety training, organizational investment, and technological investment. Moreover, the proposed safety investment decision-making model can be continuously updated and optimized with future emerging data to support the dynamic prediction of power grid safety investment.

Measurement method of inertia constant of power system based on large-scale wind power grid connection

After the large-scale wind power airport is connected to the power grid, the wind farm cannot actively provide inertia support to the system, resulting in a significant decline in the inertia of the power system and weakening the frequency regulation ability of the power system. In order to explore the measurement and evaluation of inertia constant after large-scale wind farm is connected to the power grid. This paper first introduces the composition of inertia of power system at the present stage, then expounds the response process of inertia to disturbance after large disturbance of power system, and then deduces the power system model with large-scale wind power access through the analysis of traditional inertia measurement. Finally, the simulation model is used to compare the traditional measurement estimation with this paper. The maximum error is 9.18%, the minimum error can reach 2.4%, and the minimum error of the traditional estimation method can also reach 18%. Compared with traditional methods, it can be concluded that the error between the theoretical value of inertial time constant and the estimation of inertial time constant of the method described in this paper is significantly reduced.

PowerNet: Multi-Agent Deep Reinforcement Learning for Scalable Powergrid Control

This paper develops an efficient multi-agent deep reinforcement learning algorithm for cooperative controls in powergrids. Specifically, we consider the decentralized inverter-based secondary voltage control problem in distributed generators (DGs), which is first formulated as a cooperative multi-agent reinforcement learning (MARL) problem. We then propose a novel on-policy MARL algorithm, PowerNet, in which each agent (DG) learns a control policy based on (sub-)global reward but local states and encoded communication messages from its neighbors. Motivated by the fact that a local control from one agent has limited impact on agents distant from it, we exploit a novel spatial discount factor to reduce the effect from remote agents, to expedite the training process and improve scalability. Furthermore, a differentiable, learning-based communication protocol is employed to foster the collaborations among neighboring agents. In addition, to mitigate the effects of system uncertainty and random noise introduced during on-policy learning, we utilize an action smoothing factor to stabilize the policy execution. To facilitate training and evaluation, we develop PGSim, an efficient, high-fidelity powergrid simulation platform. Experimental results in two microgrid setups show that the developed PowerNet outperforms the conventional model-based control method, as well as several state-of-the-art MARL algorithms. The decentralized learning scheme and high sample efficiency also make it viable to large-scale power grids.

Authenticating source information of distribution synchrophasors at intra-state locations for cyber-physical resilient power networks

As power systems are gradually evolving into more efficient and intelligent cyber-physical energy systems with the large-scale penetration of renewable energies and information and communication technologies, they become increasingly reliant upon more accurate monitoring and fast control. Phasor Measurement Units (PMUs) collect high-precision Distribution Synchrophasors (DS) data regarding the system dynamics and provide real-time situational awareness for better monitoring and control of large-scale power grids. The accuracy and generalizability of the PMUs heavily rely upon the data quality of DS measurements, which is very susceptible to newly emerging “Source ID Mix” data spoofing attacks. Such attacks could maliciously alter a large portion of supposedly protected data, which may not be easily detected by existing operational practices, thereby jeopardizing most DS-based applications and even causing catastrophic power interruptions. This paper proposes a novel data-driven source authentication method to automatically identify the source information of DS collected from multiple intra-state locations and thus enhance the reliability and cybersecurity of power systems. The proposed method integrates Mathematical Morphological Decomposition (MMD) and Multi-Weighted Deep Stacking Forest (MLW-DSF) for providing accurate source authentication with a low computational cost which requires neither system’s models nor parameters. The effectiveness of the proposed method is corroborated through seven “state-of-the-art” data-driven models by using real-life DS measurements of power networks in Queensland state.© 2017 Elsevier Inc. All rights reserved.

On the impact of load profile data on the optimization results of off-grid energy systems

Access to electricity via large scale power grids is seen as one of the solutions for a fully renewable power system. However, it remains a huge technical, economical, and geopolitical challenge. In the meantime, millions of people across the world have none or limited access to electricity and quite often to rely on autonomous solutions such as diesel generators. With the decreasing cost of renewable energy generation technologies in recent years, one could observe a simultaneous increase in studies dedicated to optimal sizing of renewable off-grid systems. Many of these studies rely on the usage of typical daily load profiles to model the electricity demand, sometimes enhanced with seasonal or random components. Such approaches tend to overlook the existing potential case-specific correlation between availability of renewable energy and energy demand and in particular the natural variability of the load in terms of its extreme values or ramp rates. The objective of this study is to investigate the impact of different types of load input data (for instance real load, monthly adjusted typical load, and typical daily load) on the cost of energy provided by off-grid PV-battery systems supplying various loads with different reliability levels. For this purpose, we determine the optimal capacity of PV-battery systems based on commonly used energy management strategies and optimization algorithms. The analysis of the obtained results indicates that, on average, using daily load profiles tends to underestimate the cost by 1.2% points (pp) for a system with 100% reliability and by over 5 pp for a system characterized by 95% reliability. Using monthly adjusted typical daily load profiles leads to slight differences compared to the results obtained by using real load as input. Although the obtained average values indicate a tendency of underestimating the energy cost, some outliers have been also observed reaching values of up to 15% of overestimating the cost of energy.

Extreme Solar Flare as a Catastrophic Risk

Space weather, or the disturbances of the plasma environment driven by the magnetic activities in the Sun in geospace, has become a potential source of disaster for modern society, which is increasingly dependent on its space infrastructure and large-scale power grids. Recently, independent pieces of evidence have been found that support the possibility of extremely intense space weather driven by a “superflare,” a solar phenomenon that modern society has never experienced. This paper reviews state-of-art studies of superflares and their potential impacts.

Remote error estimation of smart meter based on clustering and adaptive gradient descent method

At present, the main way for electric power companies to check the accuracy of electric meters is that professionals regularly bring standard electric meters to the site for verification. With the widespread application of smart meters and the development of data processing technology, remote error estimation based on the operating data of smart meters becomes possible. In this paper, an error estimation method of smart meter based on clustering and adaptive gradient descent method is proposed. Firstly, the fuzzy c-means clustering method is used to preprocess the data to classify the operating conditions of each measurement, and then the adaptive gradient descent method is used to establish the error estimation model. The simulation results show that this method has high error estimation accuracy. This method has a small amount of calculation and high reliability and is suitable for large-scale power grids.

Synergistic use of intrusive and non-intrusive model order reduction techniques for dynamical power grids

This manuscript combines the recently developed nonlinear moment-matching (NLMM) technique with dynamic mode decomposition (DMD) to obtain a simulation-free reduction framework for power systems. Unlike the conventional model reduction methods for power systems, where the external area is linearized, we consider the nonlinear effective network (EN) and the synchronous motor (SM) power grid models. First, the reduced system is constructed from the solution of the underlying approximate Sylvester equation by exciting the system with user-defined inputs from a representative exogenous system. Then, a non-intrusive reduction is performed using DMD to approximate the nonlinear function via Koopman modes in an equation-free manner. The advantage is that a “simulation-free” nonlinear model order reduction framework is obtained to approximate the response of the large-scale power grid models. Finally, we substantiate our observations using numerical simulations of reduced EN and SM models of the IEEE 118 and IEEE 300 bus systems for realistic fault scenarios. Results show that the overall CPU times of the reduced-order models are lowered to half as compared to the original models while maintaining the fidelity. The results are also compared with POD-DEIM for reference.

A safety investment optimization model for power grid enterprises based on System Dynamics and Bayesian network theory

In recent years, frequent large-scale power grid accidents have caused serious economic losses and bad social impact, which has drawn great attention from power grid enterprises. As one of the key elements of production, safety investment plays an important role in improving the safety level and reducing accident loss. In this paper, System dynamics (SD) and Bayesian network (BN) are integrated to develop a novel safety investment optimization model for power grid enterprises, which takes into account the impact of safety investment factors on accidents and the interactions between them. Based on sensitivity analysis, critical safety investment factors are determined to form the subsystem of the SD model. Subsequently, the optimal safety investment strategy is determined by a three-step simulation. The simulation results show that there are barrel effects and a diminishing marginal utility in safety investment. The proposed safety investment optimization model is practical to provide technical supports and guidance for determining an effective safety investment strategy in power grid enterprises.

Potential-Based Large-Signal Stability Analysis in DC Power Grids With Multiple Constant Power Loads

The increasing adoption of power electronic devices may lead to large disturbance in DC power grids. Traditionally, the Brayton-Moser mixed potential theory is utilized to address large-signal stability (LSS) analysis. However, this theory proposes sufficient criteria only for large-signal global stability, which provides general but fuzzy information about the stability of DC microgrids. To solve this issue, we propose for the first time a comprehensive hyperlocal large-signal stability analysis (HL-LSS) that is adaptable to large-scale complex power grids. The novel proposed analysis studies the complex situation in nonlinear DC grids with multiple equilibrium points, including stability analysis of each equilibrium point and convergence analysis of state trajectories. The region of attraction (ROA) of microgrids is also estimated.

Overview of High-voltage Large-capacity DC Transformer

DC transformer is the core equipment to realize the convergence and transmission of new energy such as solar energy, wind energy, etc. It also plays a key role in the construction of large-scale DC power grid in the future. Therefore, DC transformer has a broad application prospects in the future energy Internet era. This paper briefly summarizes the current research on DC transformer at home and abroad, and also summarizes the current research on DC transformer in the future. On the basis of the basic principle of decomposing DC transformers, the characteristics and applications of common DC transformers are classified and the problems to be solved are summarized.

RMIT University’s practical space weather prediction laboratory

Space weather is a key component in the daily operation of many technological systems and applications, including large-scale power grids, high-frequency radio systems, and satellite systems. As the international space sector continues to boom, accessible space weather products, tools and education are increasingly important to ensure that space actors (both old and new) are equipped with the knowledge of how space weather influences their activities and applications. At RMIT University, the initiative was taken to develop a Space Weather Prediction Laboratory exercise for students as part of its new offering of a Bachelor’s Degree in Space Science in 2020. This new Space Weather Prediction Lab exercise is offered as part of an undergraduate course on “Space Exploration”, which has a diverse student in-take, including students with no background in physics; a key detail in the design of the Lab. The aims of the Space Weather Prediction Lab were to: (1) provide a short and intense introduction to the near-Earth space environment and its impact on various human technologies; (2) give students “hands-on” training in data analysis, interpretation and communication; and (3) create an immersive space science experience for students that encourages learning, scientific transparency and teamwork. The format of the lab that was developed can be easily scaled in difficulty to suit the students’ technical level, either by including more/less space weather datasets in the analysis or by analyzing more/less complicated space weather events. The details of the Space Weather Prediction Lab developed and taught at RMIT in 2020, in both face-to-face and online formats, are presented.